T
Experimental Social Psychology
VOLUME 31
This Page Intentionally Left Blank
ADVANCES IN
Experimental Social Ps...
38 downloads
1987 Views
28MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
T
Experimental Social Psychology
VOLUME 31
This Page Intentionally Left Blank
ADVANCES IN
Experimental Social Psychology
DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF WATERLOO WATERLOO. ONTARIO. CANADA
VOLUME 31
ACADEMIC PRESS London Boston New York Sydney Tokyo Toronto San Diego
This book is printed on acid-free paper.
@
Copyright 0 1999 by ACADEMIC PRESS All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients~This consent is given on the condifion, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923). for copying beyond that permitted by Sections 107 or 108 of the US.Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-1999 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0065-260 1/99 $30.00
Academic Press a division of Harcourt Brace & Company
525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.apnet.com
Academic Press 24-28 Oval Road, London NWl 7DX, UK http://www.hbuk.co.uWap/ International Standard Book Number: 0-12-01 523 1-2 PRINTED IN THE UNITED STATES OF AMERICA 99 0 0 0 1 02 03 0 4 BB 9 8 7 6
5
4
3
2
1
CONTENTS
Contributors
........................,...............................
ix
Affect and Information Processing Robert S. Wyer, Jr., Gerald L. Clore, and Linda M. Isbell I. 11. 111. IV. V. VI. VII. VIII. IX. X.
Preliminary Considerations . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Affect and Information Processing: A Conceptual Integration . . . . . . . . . . Determinants of Affect . , . . . . . . . . . , . . , . . . . . , . , . . . . , . , . . . . , . . . . . . . . . Cognitive Consequences of Affect: Memory and Encoding Processes . . . . The Use of Affect as Information in Making Judgments (Path 10) . . . . . . Effects of Affect on Motivation and Ability (Path 9) . . . . . . . . . . . . . . . . . . The Influence of Affect on Response Evaluation (Path 11): A Performance-Feedback Model of Affect as Information . . . . . . . . . . . . . . . . .............. Multiple Influences of Affect on Judgment . . . Task Analyses of the Impact of Affect on Infor Concluding Remarks . . . . . . . . . . . . . . . . , . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3 8 9 14 29
38 41 54
58 67 68
Linguistic Intergroup Bias: Stereotype Perpetuation through Language Anne Maass I. 11.
111. IV. V. V1. VII.
The Linguistic Intergroup Bias Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............. Research Paradigm . . . . . . . . . . . . . . . . . . . . . . . Generalization across oup Settings, Populations. and Linguistic Groups . . ........................................ Generalization across rernent . . . . . . . . . .... External Validity: Linguistic Intergroup Bias an .......... Underlying Mechanisms: In-Group Protection or .... Differential Expectancy? . . . , , . . . . . . . . . . . . . . . . . . . . . . . . The Consequences of Abstract versus Concrete Language Use: Feeding into a Self-Perpetuating Cycle . ....................... V
79 82 84 85
87 91 102
vi
CONTENTS
VIII. IX.
Open Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
105 116 I18
Relationships from the Past in the Present: SignificantOther Representations and Transference in Interpersonal Life Serena Chen and Susan M. Andersen 1. 11. 111.
IV. V. VI. VII. VIII. IX.
X. XI.
The Transference Phenomenon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Social-Cognitive Model of Transference . . . . . . . . The Basic Transference Paradigm . . . . . . . . . . . . . . . . . . Motivation, Expectancies, and Affect in Transference . . . . . . . . . . . . . . . . . Transient Mood States and Interpersonal Roles in Transference . . . . . . . . The Self in the Transference Context . . . . . ....................... Basic Social-Cognitive Principles Underlying Transference . . . Applicability-Based Triggering of Transference . . . . . . . . . . . . . A Theory-Based Approach to Significant-Other Representations and Transference ................................................. Finding Common Ground: Transference and Close Relationships . . . . . . . Closing Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References ..... ....................................
123
139 146 149
165 170 179
The Puzzle of Continuing Group Inequality: Piecing Together Psychological, Social, and Cultural Forces in Social Dominance Theory Felicia Pratto 1. 11. Ill.
Individual Prejudice ............................................... The Integration: Research on Social Dominance Theory . . . . . . . . . . . . . . . The Completed Puzzle: How Group Dominance Assembles Itself . . . . . . . References .......................................................
192 205 244 254
Attitude Representation Theory Charles G . Lord and Mark R. Lepper 1. Introduction ...................................................... 11. Attitude Representation Theory ....................................
265 268
111.
IV. V. VI.
CONTENTS
vii
Research Relevant to the Representation Postulate . . . . . . . . . . . . . . . . . . . Research Relevant to the Matching Postulate ........................ Issues Related to the Attitude Representation Theory Model . . . . . . . . . . Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
274 295 318 329 330
Discontinuity Theory: Cognitive and Social Searches for Rationality and Normality-May Lead to Madness Philip G. Zimbardo 1. 11. 111.
IV. V. VI.
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discontinuity Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case Studies Illustrating Aspects of the Theory . . . Experimentally Creating Discontinuities, Search Fra Their Consequences . . . . . . . . . . ......................... Raising, Reconciling Some C ues ..................... Conclusions and Reflections ................................. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Index .............................................................. Contents of Other Volumes .........................................
345 347 389 406 458 467 473
487 501
This Page Intentionally Left Blank
CONTRIBUTORS
Numbers in parentheses indicate the pages on which the authors’ contributions begin.
SUSAN M. ANDERSEN (123), Department of Psychology, New York University, New York, New York 10003 SERENA CHEN(123), Department of Psychology, University of Michigan, Ann Arbor, Michigan 48109 GERALD L. CLORE(l),Department of Psychology, University of Illinois, Champaign, Illinois 61820 LINDAM. ISBELL(l), Department of Psychology, University of Illinois, Champaign, Illinois 61820 MARKR. LEPPER(265), Department of Psychology, Stanford University, Stanford, California 94305 CHARLES G. LORD(265), Department of Psychology, Texas Christian University, Fort Worth, Texas 76129 ANNEMAASS(79), Department of Psychology, University of Padua, 35131 Padua, Italy
FELICIA PRATTO(191), Department of Psychology, University of Connecticut, Storrs, Connecticut 06269 ROBERT S. WYER,JR. (l),Department of Psychology, University of Illinois, Champaign, Illinois 61820 PHILIPG. ZIMBARDO (245), Department of Psychology, Stanford University, Stanford, CA 94305
ix
This Page Intentionally Left Blank
AFFECT AND INFORMATION PROCESSING
Robert S. Wyer, Jr. Gerald L. Clore Linda M. Isbell
The most active area of research and theory of social information processing to emerge in the past two decades (for reviews, see Fiske & Taylor, 1991; Higgins & Sorrentino, 1990; Sorrentino & Higgins, 1986; Wyer & Srull, 1984, 1994) concerns the cognitive determinants and consequences of affect and emotion. A seemingly unlimited number of issues have been investigated. Some research addresses the dimensionality of emotional reactions to stimuli (C. Smith & Ellsworth, 1985) and the factors that give rise to these reactions (Abelson, 1983; Ortony, Clore, & Collins, 1988; Roseman, 1984, 1991). Other work focuses on the influence of affect on the encoding and organization of new information in memory (Bower, 1981, 1991), its impact on the retrieval of previously acquired concepts and knowledge (Blaney, 1986), the use of affect as a basis for judgments and decisions (Clore, 1992; Schwarz & Clore, 1983, 1988), and its influence on the cognitive strategies that individuals adopt when they perform cognitive tasks (Isen, 1987; Mackie & Worth, 1989; Schwarz & Bless, 1991). This research has shown how affective reactions can affect creativity (Isen, Daubman, & Nowicki, 1987; Isen, Johnson, Mertz, & Robinson, 19851, the impact of persuasive messages (Bless, Bohner, Schwarz, & Strack, 1990; Mackie & Worth, 1989), impression formation (Isbell, Clore, & Wyer, 19981, stereotyping (Bodenhausen, 1993; Bodenhausen, Sheppard & Kramer, 1994), self-evaluations (Levine, Wyer, & Schwarz, 1994; Schwarz & Clore, 1983), and political judgment (Abelson, Kinder, Peters, & Fiske, 1982; Isbell & Wyer, 1999; Ottati & Isbell, 1996; Ottati & Wyer, 1992). ADVANCES IN EXPERIMENTAL SOClAL PSYCHOLOGY VOL 31
1
Copyright D 1999 hy Academic Press All rights of reproduction in any form reserved 00h5-2MlliY9 $30 (x)
2
R. S. WYER, Jr., ET AL.
In this chapter, we develop a conceptualization that incorporates the implications of these diverse phenomena and the cognitive processes that underlie them. Our formulation specifies the possible determinants and consequences of the affect that individuals experience in both laboratory and daily life situations. In doing so, it takes into account the phenomena implied by many existing theories of affect and cognition. In many cases, however, it suggests interpretations of these phenomena that differ from those that have typically been proposed. One basic assumption of our conceptualization distinguishes it from other formulations of affect and cognition (e.g., Bower, 1981; Forgas, 1992, 1995; Isen, 1987; Wyer & Srull, 1989). Specifically, although affective reactions can be responses to previously acquired concepts and knowledge that are activated in memory, and although one can have concepts about their own and others’ reactions, affect per se is not itself part of the cognitive system. This assumption places restrictions on the ways that affect can influence information processing. As we will argue, many phenomena in which affect has been hypothesized to play a role may be due instead to the semantic concepts that accompany different affective and emotional states. In fact, we propose that affective reactions per se enter into information processing in two primary ways: 1. Positive and negative affect can be learned responses to external or internal stimuli. Once acquired, these responses may become preconditions for cognitive operations of the sort that compose procedural knowledge (J. Anderson, 1983; E. Smith, 1990) and govern behavior in specific types of situations to which they are relevant. Once these procedures are activated, they can influence responses to new information without conscious awareness. 2. The affective reactions that one experiences at a given moment can be used as information about one’s attitude toward either oneself, other persons, situations with which one is confronted, the outcomes of behavior in these situations, and the appropriateness of certain strategies for attaining specified processing objectives. Consequently, affect can influence both judgments of these entities and behavioral decisions that concern them. Moreover, it can affect the type and amount of goal-directed activity one performs. These conclusions are embodied in six postulates that concern the impact of affect on goal-directed information processing and the conditions in which this influence occurs. These postulates, along with empirical evidence bearing on their validity, are presented in the context of issues to which they are relevant.
AFFECT AND INFORMATION PROCESSING
3
I. Preliminary Considerations
A. AFFECT VERSUS COGNITION
1. Definition Affect refers to the positively or negatively valenced subjective reactions that a person experiences at a given point in time. These reactions are experienced as either pleasant or unpleasant feelings. Affective reactions can be elicited either by perceptions of one’s immediate stimulus environment or by thoughts about past or future events. Affect can also be elicited by psychedelic drugs, painful or pleasurable physical stimulation, or proprioceptive cues (e.g., facial expressions; see Strack, Martin, & Stepper, 1988; Zajonc, Murray, & Inglehart, 1989), but it is generally an internal response to a set of newly formed or previously acquired cognitions. Although our conceptualization of affect seems relatively noncontroversial, it should be distinguished from the implicit or explicit definitions employed in other formulations. For example, some theories (e.g.. Fiske & Pavelchak, 1986) equate affective reactions to a stimulus with evaluations of the stimulus. In contrast, others (e.g., Zajonc, 1980) assume that these reactions are totally separate from the cognitive system that assigns “meaning” to these stimuli. We also do not view affective reactions per se as part of the cognitive system. We nevertheless assume that these reactions are frequently elicited by cognitions, and that they provide the basis for assigning evaluative meaning in ways we elaborate presently. Affective reactions to a stimulus configuration can often reflect learned responses to a subset of the features that compose this configuration (Clore & Byrne, 1974; Lott & Lott, 1985; Staats & Staats, 1958). For example, if a man has a positive affective reaction to a woman he meets for the first time, he may later experience a similar reaction upon seeing the woman across the room or hearing someone mention her name. Thoughts about the woman or about aspects of the original interaction might elicit a similar reaction. The intensity of these latter reactions depends in part on the proportion of original stimulus features that are represented in the thoughts that occur. The features of an affect-eliciting stimulus event do not need to be clearly articulated. That is, a person may respond configurally (or schematically) to a stimulus situation without identifying its individual components (Clore & Ortony, in press). This appraisal often occurs spontaneously, without conscious awareness of the particular situational features that elicit it. An
4
R. S. WYER, Jr., ET AL.
articulation of these specific features might occur only subsequently, if there is a particular reason to identify them. 2. Affect and Declarative Knowledge Affect can be a response to either a previously formed mental representation of a person or event that is retrieved from memory, or a new representation that is constructed of one’s immediate environment. Moreover, it can occur in the course of imagining hypothetical situations. (This is presumably accomplished by combining affect-eliciting features of existing concepts and knowledge to form a new representation.) It is important to note, however, that although affective reactions can be responses to concepts and schemata that compose declarative knowledge, and can also be the referents of concepts and knowledge, they are not themselves part of the cognitive system. This distinction is central to the conceptualization we propose. That is, people obviously form concepts about affective reactions, and they may use these concepts to interpret their responses to new experiences. Moreover, they may spontaneous apply these concepts (like other types of concepts) to configurations of stimulus features that exemplify them. Thus, one may spontaneously apply the concept “happy” to a configuration of subjective reactions in much the same way one applies “chair” to a physical stimulus with features that exemplify this concept. Thus, a person who subjectively experiences a certain configuration of affective reactions may label these reactions as “angry,” “happy,” or “feeling great,” based on a previously formed semantic concept that refers to them. Moreover, once applied, this concept can become part of a mental representation of the experience in which this affect occurred. However, the reactions themselves are no more part of the cognitive system than are the referents of other concepts and knowledge structures (e.g., the persons, objects, or events that one encounters in daily life). The only difference is that the referents of the latter concepts are typically external to the perceiver, whereas the referents of affect-related cognitions are internal. Other similarities in the use of affect-related concepts and other concepts are worth noting. First, both types of concepts can be applied to a referent without consciously assigning the referent a verbal label. Thus, one might spontaneously recognize a stimulus object as a chair without consciously labelling it as such. Similarly, one might activate and use a concept of one’s affective state without consciously labeling the state as “happy” or “sad.” Second, several different sets of features can exemplify the same concept, and exposure to these features might activate it. Affect-related concepts, for example, might be activated by external stimulus features, such as
AFFECT AND INFORMATION PROCESSING
5
another person’s facial expression, behavior, or tone of voice, as well as by internal reactions. Thus, these concepts can potentially be applied to other persons as well as to oneself. Our assumption that affective reactions are not themselves features of the mental representations that compose declarative knowledge distinguishes our theory from other conceptualizations of the role of affect in information processing (e.g., Bower, 1981; Forgas, 1992, 1995; Isen, Shal Ker, Clark, & Karp, 1978; Wyer & Srull, 1989). These theories assume that affect functions as a concept in memory that, like other concepts, can influence the interpretation of new information and can cue the retrieval of previously acquired knowledge. If, on the other hand, affect is not itself a feature of the concept and knowledge representations that are stored in memory, there is no reason to suppose it will have these effects. We discuss these alternative possibilities in more detail later in this chapter. 3. Affect and Procedural Knowledge
Although affective reactions are not represented in declarative knowledge, they could potentially play a role in procedural knowledge. J. Anderson (1983: see also E. Smith, 1984) postulates the existence of cognitive “productions.” These productions can be conceptualized metaphorically as “IF[X], THEN[Y]” rules, where [Y] is a sequence of cognitive or motor acts and [XI is a configuration of internally generated stimuli (thoughts, perceptions, or bodily sensations) that serve as a precondition to this sequence. The features of such a precondition can include both affective reactions and cognitions about the situation in which the reactions occur. However, these reactions need not be interpreted in terms of concepts drawn from declarative knowledge; persons might not even be consciously focusing on them. To this extent, the affective reactions that people experience in a situation could influence their responses to the information they receive independently of the interpretation, if any, that they place on these reactions. These possibilities are also considered in the conceptualization we propose.
B. RELATED CONSTRUCTS 1. Moods and Emotions Many types of affective reactions can be experienced, each of which is elicited by a different configuration of cognitions. These cognitions could concern either specific aspects of a real or imagined situation or a more
6
R. S . WYER, Jr., ET AL.
global impression of the situation as a whole (Lazarus, 1982, 1991). The cognitions could also pertain to past or future experiences. Once affective reactions are elicited, they may persist for a period of time after the cognitions that gave rise to them are no longer activated. Affective states that occur in response to cognitions about actions, events, or objects, these affective reactions are termed emotions (Clore, 1994). The particular emotion that one experiences depends on not only the stimulus event itself but also the interpretation that is placed on it (Ortony et al., 1988; see also Abelson, 1983; Roseman, 1991). This means that the same event can often elicit different emotions, depending on the perspective from which it is viewed and the aspects of the event to which one attends. For example, suppose a boy’s alcoholic uncle has pawned the child’s favorite toy to buy a bottle of scotch. The boy is likely to experience sadness if he focuses on the consequences of the event for himself, but might experience anger if he considers the uncle’s behavior. On the other hand, the uncle might experience happiness if he thinks about the event’s consequences for himself but feel guilty if he thinks about his behavior and its consequences for the boy. One implication of this conceptualization is that an individual’semotional reactions to a situation can change abruptly, depending on the cognitions about it that happen to come to mind. Once the cognitions associated with an emotion-eliciting experience are no longer activated, however, the reactions that were stimulated by them take time to dissipate. These reactions, which are experienced without concurrent activation of concepts about their source, are referred to as moods (for an elaboration of this distinction, see Clore, 1994; Clore, Schwartz, & Conway, 1994). This implies that every emotion that is elicited by a cognitive appraisal may have a corresponding mood, or residual affective state, that is experienced for a period of time after a representation of the appraised event is no longer activated. Although persons can theoretically experience only one emotion at a time, their moods could in principle reflect the residual affect of several different emotions. The theoretical and empirical implications of this possibility have rarely if ever been explored. For one thing, the affective reactions associated with different cognitive appraisals of a situation can potentially differ in terms of both the time required to reach their maximum intensity and their rate of decay once the cognitions that elicited them are no longer activated (Frijda, Ortony, Sonnemans, & Clore, 1992). An understanding of the different rates of onset and decay that characterize different types of affect could provide insight into both the immediate and long-range effects of emotional experiences and the impact of new events that distract one from thinking about them. Thus, suppose a man whose wife has left him experiences both anger and sadness in close temporal contiguity. The
AFFECT AND INFORMATION PROCESSING
7
relative magnitudes of these feelings once the precipitating event is no longer thought about could depend on the relative rates of decay of the two types of reactions as well as the recency of the thoughts that elicited them. Similar considerations underlie the conditions in which positive feelings induced by a happy experience will override the negative feelings that result from an earlier, unhappy event.
2. Attitudes and Evaluations One’s affective reactions to an object can often provide a basis for evaluating the object. These evaluations are often assumed to express attitudes (Eagly & Chaiken, 1993; Fazio, 1986).’ We do not distinguish between attitudes and evaluations in the theory we propose. However, we d o distinguish between the criteria that are used as a basis for these judgments, and also between the conditions in which these criteria are applied (for a similar distinction, see Niedenthal & Setterlund, 1994; Zanna & Rempel, 1988). Specifically, affect-based evaluations are based principally if not exclusively on affective reactions. The most obvious evaluations of this type pertain either directly or indirectly to one’s feelings about oneself or others. For example, they include one’s feelings about one’s general life situation (e.g., subjective well-being; see Schwarz & Clore, 1983; Seidlitz, Wyer, & Diener, 1997), about oneself (e.g., judgments of self-esteem) or about other persons (e.g., judgments of love, anger, or amusement). In contrast, descriptively based evaluations are typically based on declarative knowledge about the person or object being judged. For example, judgments of positively or negatively valenced personality traits (honesty, intelligence, etc.) might be based on specific behaviors that exemplify them. Or, the consequences of an action (having an abortion, going to the dentist, etc.) might be evaluated on the basis of beliefs that they would facilitate or interfere with one’s social or personal objectives. The descriptive and affective bases for an evaluation may have different implications. Going to the dentist, for example, might elicit negative affective reactions but be necessary in order to attain a desired goal (not losing one’s teeth). By the same token, eating a hot fudge sundae might elicit positive affect but interfere with one’s goal of losing weight. When both descriptive and affective criteria are potentially relevant to an evaluation, their relative impact might depend in part on which criterion comes to mind most quickly and easily (Levine et al., 1994).
‘
This definition of course differs from the tripartite conception of attitudes as having cognitive and behavioral components as well as an affective one (Rosenberg & Hovland, 1960). However, the latter conception can lead to conceptual confusion for reasons noted by Wyer and Carlston (1979; see also Petty, Priester, & Wegener, 1994). We prefer to restrict our definition to the affective dimension. thereby relegating the relations among attitudes, beliefs, and overt behavior to empirical or theoretical inquiry.
8
R. S. WYER, Jr., ET AL.
11. Affect and Information Processing: A Conceptual Integration
The role of affect in goal-directed information processing must be understood in the context of a more general conceptualization of the cognitive activities that mediate goal-related judgments and decisions. A number of recent processing models rely on current connectionist and paralleldistributed-processing assumptions (e.g., Kashima & Kerekes, 1994; E. Smith, 1996;Vallacher & Nowak, 1994).For our present purposes, however, a sequential-stage processing model is particularly useful. The stages of processing we postulate to occur in response to stimulus input information are similar to those assumed by Wyer and Srull(l986; 1989;see also Wyer & Radvansky, in press), specifically: 1. When people encounter new stimulus information, they retrieve semantic and episodic knowledge from memory that they can use to interpret it and construe its implications. The likelihood of using any particular subset of knowledge is a joint function of its applicability and its accessibility in memory. Thus, when more than one set of previously formed cognitions is equally applicable, the set that comes to mind most quickly and easily is most likely to be applied.’ 2. Once people have interpreted new information in terms of previously acquired concepts and knowledge, they operate upon it in ways they believe will attain the specific objectives to which the information is relevant. These objectives can be either internally generated or externally imposed. The specific sequence of cognitive operations that is performed, which may often constitute a cognitive procedure as conceptualized earlier ( J. Anderson, 1983; E. Smith, 1984,1990,1994), depends on the goal being pursued as well as the type of information being processed (for analyses of the role of goals in information processing, see E. Smith, 1994; Srull & Wyer, 1986; Wyer & Srull, 1989).3 3. People generate a subjective output, or “result,” from the procedure they have employed and evaluate its utility for attaining the goal to which The accessibility of these concepts, in turn, can be influenced by (a) the goal that recipients happen to be pursuing, (b) transitory situational factors that have led these cognitions to be activated and applied (Bargh, 1984,1994,1997; Higgins, Bargh, & Lombardi, 1985; Higgins & King, 1981; Wyer & Srull, 1989), and (c) the frequency with which the concepts have been used in the past (Higgins et al., 1985; Wyer & Srull, 1989). The frequency of using concepts and knowledgt can, in turn, influence their “chronic” accessibility in memory (Bargh, Bond, Lornbardi, & Tota, 1986; Bargh, Lombardi, & Higgins, 1988 Klinger, 1975). The procedures can sometimes be activated and applied without awareness of the goal to which they are relevant (Bargh, 1997). For our present purposes, however, we assume that individuals are typically conscious of the goals toward which their behavior is directed, although many of the cognitive steps involved in attaining these goals might be performed automatically (Wyer & Srull, 1989).
AFFECT AND INFORMATION PROCESSING
9
it is relevant. If they decide that this output is sufficient for attaining their goal, they make an overt response (e.g.. a judgment or behavior) and terminate goal-directing processing. When multiple responses are required for goal attainment, they may use the initial output they have generated to evaluate whether the strategy they have used to produce this output is appropriate and, therefore, to decide if they should continue exploring this strategy or adopt a different one. The processes described in step 3 compose a feedback loop whereby the outputs of processing are continually being evaluated and processing continues until a result that fulfills one’s processing objective has been obtained. (For other theories that postulate such a mechanism, see Carver & Scheier, 1981, 1990a,b, Miller, Galanter & Pribram, 1960: Simon, 1967.) The criterion for evaluating this result is not always clear. A man who is preparing for an examination, for example, may wish to continue studying until he feels that he “knows enough.” However, whether this goal has been attained cannot easily be determined objectively. Rather, it may be based on the student’s subjective “feelings” about his knowledge of the material. Several specific theories of affect and cognition can be viewed as hypotheses about the role of affect to one or more of the stages of information processing described above. Figure 1 conveys these stages of processing and the sequence in which they typically occur. The numbered paths indicate the cognitive determinants of affect as well as its consequences for goal-directed processing. Thus they indicate the various ways in which both situational and individual difference variables could potentially influence the outcomes of goal-directed processing through the mediating influence of these variables on the affect that persons experience. Each numbered path is the subject of one or more theories of affect and cognition. On the other hand, not all of these paths are equally likely to exist. In the theory we propose, only those paths denoted by solid lines are postulated. This becomes clear in the discussion to follow.
111. Determinants of Affect
Affective reactions can theoretically be elicited by stimulus events either directly, with a minimum of cognitive mediation (Paths 1 and 3 ) , or indirectly as a result of the interpretation that is placed on these events (Path 6). Affect can also be elicited spontaneously by thoughts about previously encountered people and events, or by imagined future events whose
I....y Proprioceptive
Fig. 1. Conceptual path diagram showing the cognitive determinants of affect and the cognitive consequences of affect for processes involved in goaldirected cognitive activity.
AFFECT AND INFORMATION PROCESSING
11
features have become associated with affective reactions through prior learning (Path 5). Proprioceptive feedback can provide yet another source of affect (Path 2). Although some of these effects are self-evident, others require discussion.
A. NONCONSCIOUSLY MEDIATED AFFECT ELICITATION (PATHS 1-3) The possibility that internally generated proprioceptive stimulation elicits affect without any conscious cognitive mediation (Path 2) is a basic tenet of the facial-feedback hypothesis (Laird, 1974; for evidence supporting its validity, see Strack et al., 1988; Zajonc et al., 1989). Zajonc (1980; Murphy & Zajonc, 1993; Zajonc & Markus, 1984) postulated that externalstimulus events can also elicit affect without any cognitive mediation (Path 3). Early support for this hypothesis was based on evidence that liking for novel stimuli increases with the frequency of exposure to these stimuli independently of the ability to recognize the stimuli as ever having been seen before (Kunst-Wilson & Zajonc, 1980; Moreland & Zajonc, 1979; Murphy & Zajonc, 1993). However, refined analysis of the data from these studies raise questions about the validity of the conclusions that were drawn (Birnbaum, 1981; Birnbaum & Mellers, 1979a, b; Wyer & Srull, 1989). In a quite different paradigm (Bargh, Chaiken, Raymond, & Hymes, 1996), participants on a series of trials were first exposed to a positively or negatively valenced word below recognition threshold, and then were asked to pronounce a similarly or differently valenced target word that followed it. Participants responded more quickly to a target word when it was evaluatively congruent with the word that preceded it than when it was incongruent. This difference occurred although participants had no insight into the fact that the valence of the words was of any concern. Bargh et al. (1996) interpreted these data as evidence that the initial word, although not consciously recognized, spontaneously elicited an evaluative reaction. This reaction either facilitated or interfered with the identification of the word that followed it, depending on whether the evaluative response elicited by the second word was similar or dissimilar. Although these results could be interpreted as evidence for Paths 1 and 3, they do not necessarily indicate that affective and cognitive systems are independent. For one thing, cognitive responses to stimuli can also occur outside awareness (Bargh, 1997). Moreover, even assuming that evaluations of stimuli occur automatically, this does not necessarily mean that affective reactions are the basis for these evaluations. In fact, further research from
12
R. S. WYER, Jr., ET AL.
Bargh’s laboratory (Garcia, 1996; see also Bargh & Alvarez, in press) suggests a somewhat different interpretation of his findings. In these studies, participants were exposed to unfamiliar priming stimuli that had been normatively scaled as favorable or unfavorable on the basis of pretesting, but which they had personally never seen before. These novel priming stimuli influenced responses to subsequent target stimuli in much the same way as familiar primes. This means that the influence of priming stimuli on the processing of subsequent ones is not a result of affective reactions that had become associated with the primes through prior learning. Rather, the priming stimuli, although unfamiliar, spontaneously activated an evaluation process (i.e., a cognitive procedure, or production, as conceptualized earlier), and the results of this process either facilitated or interfered with the processing of subsequent stimuli that were also evaluated spontaneously. In fact, the phenomena identified by Bargh and his colleagues may not bear on the automatic elicitation of affect at all. Rather, they may reflect an automatic categorization process that is not affect-eliciting in and of itself. That is, persons may spontaneously categorize stimuli as either “good” or “bad.” (This automatic categorization process could reflect an unlearned, evolution-based disposition to identify stimuli as potentially harmful or not harmful in order to survive.) However, affective reactions might not provide the basis for this categorization. Rather, these reactions might only be elicited in the course of a postcategorization cognitive process that requires an appraisal of the stimulus and its implications. If this two-stage evaluation process exists, it could help to resolve a controversy between Bargh et al.’s claim that evaluative responses are automatically elicited by stimuli independently of attitude intensity, and Fazio, Sanbonmatsu, Powell, and Kardes’s (1986) conclusion that automatic attitude activation increases with attitude strength. The arguments that surround this debate are subtle and complex (cf. Chaiken & Bargh, 1993;Fazio, 1993). The above conceptualization, however, suggests that the two parties to the debate might be focusing on different stages of a two-stage process, and that their positions are not, in fact, in~ompatible.~ Conclusions drawn by Winkielman, Zajonc, and Schwarz (1997) are also worth considering in this context. In this research, participants were sublimiBargh et al. (1992) find that the effects observed by Fazio et at. (1986) are more likely to occur when participants’ attitudes toward the stimuli presented are reported before the priming task in the same experimental sessions, and that the contingency of priming effects on attitude strength disappears when a greater delay is imposed between attitude measurement and the test situation. This contingency also suggests that Fazio et al. may be tapping an influence of affect that occurs over and above the effect observed by Bargh et al., and that requires a recent activation of the affective reactions associated with the priming stimuli in order to occur.
AFFECT AND INFORMATION PROCESSING
13
nally exposed to happy or angry faces. Immediately following each face, they evaluated a relatively neutral Chinese ideograph. Although participants typically did not report experiencing affective reactions to the ideographs, they nevertheless evaluated them more favorably when the ideographs had been preceded by a happy face then when they had been preceded by an angry one. The authors concluded that participants experienced affective reactions to the facial expressions of which they were unaware, and that they used these reactions as an informational basis for their judgments of the ideographs (Schwarz & Clore, 1983; a more detailed discussion of the use of affect as information is provided later in this chapter). However, it seems equally plausible to assume that the faces did not elicit affective reactions per se, but rather, were spontaneously categorized as “good” or “bad.” Then, in the absence of other clear judgmental criteria, the category activated by each face was applied to the neutral stimulus that followed it, and the evaluation of the stimulus was based on this categorization. To this extent, Winkielman et al.3 results are quite consistent with our interpretation of Bargh et al.’s findings.
B. COGNITIVELY MEDIATED AFFECT (PATHS 5 AND 6) The concepts that individuals use to interpret a stimulus event can potentially be influenced by their information-processing objectives, the perspective from which they consider the event, transitory situational factors, and the chronic accessibility of different subsets of relevant concepts and knowledge. Their interpretation of the event, in turn, can influence the emotions they experience (Ortony et al., 1988; Roseman, 1991). On the other hand, affect can also be elicited by a more global cognitive appraisal of a situation that does not involve the identification of any specific feature (Lazarus, 1982, 1991). The effects of cognition on affect are recognized in theories concerning the correlates of depression (Beck, 1982; Nolen-Hoeksema & Morrow, 1993) and ruminative thought (Martin & Tesser, 1996). Because the features of a previous experience that are retrieved from memory typically decrease in number over time,5 the affect that is elicited is likely to decrease correspondingly. On the other hand, one’s processing objectives can sometimes
’
This decrease in detail does not necessarily reflect a general decay of the features that compose this representation. Rather, new, less detailed representations may be formed of the experience in the course of processing that occurs subsequent to its construction (e.g.. communicating about the experience to others). and these more recently formed representations become relatively more accessible in memory than the originals (for a theoretical treatment of the cognitive mechanisms that underlie such effects, see Wyer & Srull, 1986, 1989).
14
R. S. WYER, Jr., ET AL.
require the retrieval of affect-eliciting concepts that were not applied to an event at the time it occurred. In such cases, the emotional reactions experienced when thinking about the event could be even stronger than they were at the time the event actually took place. C. INDIVIDUAL DIFFERENCES IN AFFECT ELICITATION (PATH 4) Chronic individual differences can exist in the intensity and duration of affect elicited by an event (Diener, Colvin, Pavot, & Allman, 1991; Diener, Sandvik, & Pavot, 1991; Larson & Diener, 1987; Seidlitz, Wyer, & Diener, 1997). Seidlitz et al. (1997), for example, found that people with chronically high subjective well-being perceived positive life experiences as more intense (but not more enduring), and perceived negative experiences as less enduring (but not less intense), than persons with low subjective well-being. Individual differences can also exist in the importance attached to the feelings one experiences, the ability to distinguish between different types of affective reactions, and the tendency to ruminate about these reactions (Bagby, Parker, & Taylor, 1984; Emmons, 1992; Gohm & Clore, 1999; Salovey, Mayer, Goldman, & Palfei, 1994; Wyer, 1996). D. SUMMARY Affective reactions can be elicited by proprioceptive stimulation (Path 2), by cognitions about the immediate situation in which one finds oneself (Path 5), and by thoughts about past and future situations (Path 5 ) . These cognitions may be either well articulated or may consist of an undifferentiated appraisal of a situation that can occur at a low level before and does not require awareness (e.g., LeDoux, 1996). External stimuli may elicit affect through the mediating influence of these cognitions. However, these stimuli are unlikely to have a direct impact on affective reactions that are not mediated by some processing of their significance. Therefore, Paths 1 and 3 are unlikely to exist.
IV. Cognitive Consequences of Affect: Memory and Encoding Processes Affect has been postulated to influence information processing at a number of different stages. Different theories concern the impact of affect on
AFFECT A N D INFORMATION PROCESSING
15
the retrieval of previously acquired concepts and knowledge from memory (Path 7), on the interpretation of new information (Path S), and on the cognitive operations that are performed in processing this information (Path 9). Affect has also been postulated to have direct informational effects on both the responses that are made to a stimulus (Path 10) and on the evaluation of these responses (Path 11). According to the formulation we propose, however, Paths 7 and 8 may not exist. Moreover, many influences of affect on information processing may be mediated by the impact of affect on response evaluation (Path 11) and may not reflect a direct influence of affect on the motivation or ability to think extensively about the information provided (Path 9). These conclusions follow from our assumption that affective reactions can be responses to the concepts and cognitions that compose declarative knowledge, and can also be referents of this knowledge, but are not themselves part of the mental representations that compose this knowledge. As we will indicate, many experiments that have claimed support for the consequences of affect described by Paths 7-11 are subject to alternative interpretation. The existence of an alternative interpretation of research findings does not, of course, invalidate the original one. However, the conceptualization we propose is quite compatible with much of the research performed to date, and can account for results that other formulations cannot. A. THE INFLUENCE OF AFFECT ON THE ACCESSIBILITY OF DECLARATIVE KNOWLEDGE
If affective reactions are features of the cognitions that compose declarative knowledge, they should have the same impact on information processing as other, semantic features. That is, the experience of these reactions should increase the accessibility in memory of concepts and knowledge representations with which they are associated. Thus, for example, happy persons should tend to recall and use positively valenced concepts and knowledge, whereas sad individuals should be more disposed to recall and use negatively valenced cognitions (Bower, 1981). Our conceptualization does not make this prediction. The physiological reactions that exemplify a particular affect-related concept (e.g., “happy,” “angry,” etc.) may sometimes be interpreted in terms of this concept and, therefore, may activate the concept in much the same way as exposure to a specific external stimulus (e.g., a chair) activates a semantic concept it exemplifies. However, this process may not occur unless the reactions being experienced (a) are above some minimal level of intensity, and (b) clearly
16
R. S . WYER, Jr., ET AL.
exemplify a single, preexisting concept. The following postulate summarizes this contention: Postulate 1 Affective reactions may activate a specificaffect-relevant concept that can be used to interpret them. However, affective reactions do not spontaneously influence the accessibility in memory of other similarly valenced semantic concepts and declarative knowledge.
The conditions in which affective reactions are and are not likely to be associated with the recall and use of declarative knowledge can be conceptualized by considering this postulate in the context of our previous discussion of the conditions in which affective reactions are elicited. Three general conditions were identified. 1. Affect can be elicited by proprioceptive cues. This affect, however, is typically of low intensity. Therefore, it may not normally receive conscious attention, and may not be spontaneously interpreted in terms of an affectrelated semantic concept. Consequently, it is unlikely to influence the accessibility of declarative knowledge. 2. Affect can be elicited by a global cognitive appraisal of a situation that does not require attention to its specific features (Lazarus, 1982). If the affect that results from such an appraisal is of very low intensity, a spontaneous comprehension of it in terms of a specific affect-related concept it exemplifies may not occur. If the affect is moderately intense, it might be comprehended in terms of an applicable semantic concept but may inspire little effort to understand the conditions that gave rise to it. Consequently, this affect is also unlikely to affect the retrieval of more general concepts and knowledge. However, if the affect that is elicited by one’s initial appraisal of a situation is very intense, it may stimulate an attempt to explain its occurrence. As a result, the concepts and knowledge that are involved in this cognitive activity may become temporarily more accessible in memory. 3. Affect can be elicited by cognitions about specific events that occur in a new situation or by thoughts that occur in the course of complying with external situational demands. For example, people who receive feedback about their performance on an achievement test may spontaneously think about their competence and intelligence, and these thoughts, in turn, may elicit affect. Or, the events that occur in a movie might require the activation and use of concepts and knowledge with which affect is associated. Thoughts about a past experience can also elicit affective reactions
AFFECT AND INFORMATION PROCESSING
17
in the course of describing the experience to others. In all of these cases, the affect that people experience is accompanied by previously acquired semantic concepts and cognitions. Once these concepts are activated, they could cue the retrieval of declarative knowledge that is descriptively related to them. Moreover, this could occur independently of the affective reactions that happen to accompany the concepts. Figure 2 summarizes the implications of this analysis. Dashed lines indicate the direct influence of affect on the retrieval and use of semantic and episodic knowledge postulated by Bower (1981) and others (Forgas, 1995; Isen et al., 1978). Solid lines indicate the relations implied by Postulate 1. That is, affective reactions do not directly influence the retrieval of previously acquired declarative knowledge. Their influence, if any, is indirect, being mediated either by semantic concepts that are applicable for interpreting the reactions or by thoughts that are generated in an attempt to explain their occurrence. On the other hand, cognitions about features of the situation in which affective reactions occur can have an impact on the accessibility and use of previously formed cognitions independent of the affective reactions that the situation elicits. Three bodies of research bear on these possibilities. The first concerns the assumption that affect increases the accessibility of the specific affectrelated semantic concepts it exemplifies, but does not spontaneously influence the accessibility of other valenced concepts that are irrelevant to its interpretation. The second concerns the influence of affective reactions on
-
Mood induction procedures
Retrieval I,acquired ofpreviously knowledge
0
Affective L reactions
Thoughts stimulated by affective reactions
Accessibility
+ OF semantic concepts and declarative
Interpretation of new information
Fig. 2. Hypothetical effects of mood-induction procedures on information retrieval and the encoding of new information. Solid lines indicate paths postulated by proposed conceptualization. Dashed lines denote paths implied by the assumption that affective reactions function as concepts in memory.
18
R. S. WYER, Jr., ET AL.
the interpretation and selective encoding of new information into memory (Path 8). The third body of research focuses on the impact of affect on the recall of previously acquired episodic knowledge (Path 7). In much of the research, effects of affect that participants experience in the situation being investigated cannot easily be distinguished from the effects of concepts and knowledge that have been activated by the procedures used to induce this affect. Nevertheless, some tentative conclusions can be drawn. 1. The Influence of Affect on Semantic Concept Accessibility According to Postulate 1,affect spontaneously influences the accessibility of only those semantic concepts that can be used to interpret the specific type of affective reactions being experienced. Two sets of studies potentially bear on this hypothesis. First, Bargh and his colleagues (Bargh, 1997) showed that exposure to stimulus words increases the ease of recognizing other words that are similar in valence, but decreases the ease of recognizing words that differ in valence, As we have already pointed out, however, Bargh’s findings may reflect a categorization process that is automatically activated by exposure to a stimulus, and do not necessarily indicate that affect per se influences semantic concept accessibility. Experiments by Niedenthal and her colleagues (Niedenthal & Setterlund, 1994; Niedenthal, Halberstadt, & Setterlund, 1997) provide direct support for the conceptualization we propose. Participants were induced to feel either happy or sad as a result of listening to classical music that was either upbeat (e.g., Eine Kleine Nachtmusik) or gloomy (e.g., Rachmaninoff ’s Piano Concerto no. 2). The music was determined on the basis of normative data to induce happiness or sadness in particular rather than other types of positive and negative feelings (Niedenthal et al., 1997). However, participants were told that the study was intended to investigate the relation between auditory and visual perception, and were given no indication of the affective reactions the music was intended to elicit. After listening to the music for 12-15 minutes, participants performed a word recognition task. Some words referred to feelings and behavior similar to those that participants were experiencing (e.g., delight, joy, weep, despair, etc.). Others, however, referred to concepts that were positively or negatively valenced but either referred to different emotions than those participants were experiencing (pride, fear, anger, etc.; see Niedenthal et al., 1997) or did not refer to emotions at all (e.g., “wisdom,” “luck,” “crime,” “injury,” etc.; see Niedenthal & Setterlund, 1994). Participants identified affect-related words more quickly if the words described feelings similar to those they were experiencing at the time. In other words, semantic concepts denoted by these words were made more accessible in memory
AFFECT AND INFORMATION PROCESSING
19
as a result of the feelings that participants were experiencing. In contrast, the accessibility of words that described other positive and negative emotions, or that described positively and negatively valenced persons, objects, and events that were unrelated to emotions, was not influenced at all by the affect that participants were experiencing. This pattern of results, which was replicated in several different studies, is quite consistent with Postulate 1. Two further implications of the results are of particular importance. First, the results support our contention that not all positively valenced and negatively valenced stimuli elicit affect. That is, evaluations can be based on other, nonaffective criteria as well. In the absence of this assumption, the contingency identified by Niedenthal and her colleagues would be difficult to reconcile with the automatic evaluation effects reported by Bargh (1997), which generalize over all types of stimuli. Second, if people’s affective reactions typically influence the accessibility of only those semantic concepts that are exemplified by the specific reactions being experienced, the effects of these reactions on the interpretation and recall of information should be similarly restricted. Put another way, any more general influence of induced affect on the selective encoding, interpretation, and retrieval of valenced information are likely to occur for reasons that have little to do with affect per se. 2. The Influences of Affect on the Interpretation and Selective Encoding of New Information a. Selective Encoding. If the affect that people experience increases the accessibility of affect-related semantic concepts, information that is easily interpretable in terms of these concepts should be more likely to receive attention than other information and, therefore, should be more likely to be retained in memory. Although several studies bear on this possibility, their interpretation is equivocal. In a study by Forgas and Bower (1987), participants were induced to feel either happy or unhappy by giving them positive or negative feedback about their performance on a test of social adjustment. Then, they were asked to form impressions of hypothetical individuals on the basis of behaviors that varied in favorableness along dimensions of competence, intelligence, likableness, and happiness. Participants spent more time thinking about the descriptions, and consequently had better recall of these descriptions later, if the descriptions were evaluatively congruent with their moods. However, these findings must be considered in the context of Neidenthal and Setterlund’s (1994) evidence that affective reactions only increase the accessibility of semantic concepts that are descriptively related to the feel-
20
R. S. WYER, Jr., ET AL.
ings that participants are experiencing. If this is true, these reactions should not influence the attention paid to positively and negatively valenced information in general. This means that the effects observed by Forgas and Bower, which pertain to information that is not directly mood-related, must be due to other factors. It seems reasonable to suppose that the performance feedback that Forgas and Bower used to induce affect also elicited thoughts about competence, likableness and other attributes implied by the behaviors that participants read subsequently. Semantic concepts activated by these thoughts, rather than the affective reactions accompanying them, could easily have influenced the attention paid to these behaviors and the likelihood of encoding them into memory. An earlier series of studies, by Bower, Gilligan, and Montiero (1981), is also noteworthy. Participants were induced through hypnosis to feel either happy or sad by recalling a past experience and were then given a posthypnotic suggestion to maintain this affective state while they read a story. In some conditions, the story concerned two persons, one of whom had very favorable experiences and the other of whom had a series of misfortunes. Participants had better recall of the information about the character whose experiences were congruent in valence with the affect they were experiencing. When participants’ moods were induced after they had read the story rather than before, however, mood had no influence on the type of information they recalled. Thus, the effects were due to the selective attention to and encoding of affect-congruent information at the time this information was received. Some of this information described feelings of the characters that were similar to those that participants were themselves experiencing. Therefore, this selective encoding could be attributable in part to the greater accessibility of semantic concepts that referred to these feelings. On the other hand, the mood-induction procedures employed by Bower et al. explicitly mentioned the moods that participants should experience. The concepts activated by these procedures are likely to have remained accessible in memory after participants were dehypnotized.6 Therefore, they may have affected the attention paid to aspects of the information that could be interpreted in terms of them and the recall of these aspects later. Our reinterpretation of Bower et al.’s findings is consistent with conclusions drawn by Mayer, Gayle, Meehan, and Haarman (1990). They argued that if affective reactions per se were the determinant of mood-congruent effects on recall, these effects should increase with the intensity of the In this and other studies cited in this chapter, affect was induced under hypnosis. However, although people are unable to recall information they have been told to forget under hypnosis, the semantic concepts that were activated in the course of responding to this information remain highly accessible and influence responses once persons are no longer hypnotized (cf. Kihlstrom, 1980).
AFFECT AND INFORMATION PROCESSING
21
affect that participants experienced. In a study of the effects of induced mood on the recall of positively and negatively valenced words, however, they found no evidence that this was the case (see also Blaney, 1986). As the authors conjectured, it seems likely that the impact of the mood-induction procedures they used was primarily due to the cueing properties of the cognitions that accompanied the moods that participants experienced and not to their affective reactions per se.’ b. Interpretation of Ambiguous Information. If affective reactions spontaneously activate semantic concepts that concern these reactions, they should increase the likelihood of recalling and using the concepts to interpret information to which they apply (for summaries of the effect of concept accessibility on the interpretation of information, see Bargh, 1994; Higgins & King, 1981; Wyer & Srull, 1981, 1989). Several studies summarized by Forgas (1995) investigated this possibility. Bower (1981) reported that persons who were induced to experience either positive or negative affect interpreted the situations portrayed on Thematic Apperception Test (TAT) cards in a manner that was affectively congruent with their induced mood state. However, these differences could result from a tendency for persons to construct fantasies that are affectively consistent with their mood independently of the features of the TAT cards they were shown. Thus, the results do not necessarily indicate that persons actually interpreted the features of the cards they saw in terms of affect-congruent concepts. In an experiment by Forgas, Bower, and Krantz (1984), participants were induced under hypnosis to feel happy or sad as a result of writing about a past experience. They then watched a videotape of a previously recorded interaction between themselves and another person while they performed an achievement task. Happy participants identified more skilled behaviors, and fewer unskilled behaviors, than sad participants did. Note, however, that the behaviors that participants observed were not relevant to affect per se. Thus, there is no reason to believe that concepts activated by the participants’ affective states per se would influence the interpretation or selective encoding of the behaviors they observed (Niedenthal & Setterlund, 1994). It seems more reasonable, therefore, to assume that semantic concepts that mediated the effects obtained by Forgas et al. were activated by descriptive aspects of the procedures used to induce affect and not by the affect itself (see footnote 6).
’
Results reported by Erber (1991) might appear inconsistent with this conclusion. He found that affect induced through an unobtrusive manipulation of participants’facial expressions (cf. Strack et al., 1988) increased participants’ recall of trait terms that were congruent in valence with affect they were experiencing. However, recall was assessed following a task in which the trait terms may have been differentially processed in the course of making judgments to which they were relevant. Therefore, differences in the later recall of these traits may have been the result of this factor rather than affect per se.
22
R. S. WYER, Jr., ET AL.
Although an unpublished study by Kelly and Wyer (cited in Wyer & Srull, 1989) avoided this ambiguity, the interpretation of their results is also equivocal. Briefly, persons who felt either happy or sad as a result of recalling a life experience read a story about someone whose affective reactions could be interpreted as exemplifying both of those emotions. (In one story, for example, a boy’s grandmother died and left him enough money to attend the college of his choice.) Participants tended to interpret the protagonist’s feelings as similar to those they were personally experiencing. These results could indicate that affect-relevant concepts associated with participants’ own affective state influenced their interpretation of the protagonist’s. However, the mood-induction procedure employed in this study explicitly asked participants to recall either a “happy” or “sad” experience. Thus, the semantic concepts activated by these instructions might have influenced participants’ interpretation of the story rather than their affective reactions per se.
3. The Influence of Affect on the Retrieval of Episodic Knowledge Numerous studies have attempted to demonstrate that experimentallyinduced mood states cue the retrieval of previously-acquired knowledge that is evaluatively congruent with these states (for summary, see Forgas, 1995). In most of this research, however, the conditions that are necessary to demonstrate the cueing properties of affect per se have not been met. In the study reported by Bower (1981), for example, participants who had been induced under hypnosis to feel “happy” were later more likely to recall happy childhood experiences, and less likely to recall unhappy ones, than individuals who had been induced to feel “sad.” However, the moodinducing instructions explicitly referred to the affective states that participants were induced to experience. It seems likely that these instructions activated concepts of the affective states they described, and that these concepts cued the retrieval of memory representations that contained features of the concepts (see footnote 6). This may have occurred independently of the affect that participants were experiencing. In fact, the support for the hypothesis that experimentally induced mood states cue the retrieval of mood-congruent information appears to be weak and inconsistent (Blaney, 1986; Bower, 1991; Bower & Mayer, 1985). In the previously mentioned study by Bower et al. (1981), participants read a passage containing both positively valenced and negatively valenced descriptions of persons and the events they experienced. When differences in participants’ mood were induced before the passage was read, descriptions were better recalled when their valence was congruent with the affect participants were experiencing than when it was incongruent. However,
AFFECT AND INFORMATION PROCESSING
23
when the affect that participants experienced was not induced until after they had read the passage, it had no influence on the type of information recalled. In other words, any effects of participants’ feelings on their recall of the information presented in this study resulted from processes that occurred at the time the information was first received and comprehended rather than at the time of retrieval. Bower and Mayer (1985) suggested that experimentally induced mood manipulations are often too weak to permit their effects to be detected. This contingency would of course be consistent with Postulate 1. That is, weak mood inductions are unlikely to stimulate participants to think spontaneously about their affective reactions and to interpret them in terms of affect-related concepts and knowledge. Without this mediating cognitive activity, an influence of mood on information retrieval should not occur. As we have noted, a direct test of the proposed conceptualization requires the use of affect-induction procedures that do not themselves activate semantic concepts relevant to the material to be recalled. The few studies that have employed such procedures support this conceptualization. In research by Riskind, Rholes, and their colleagues (Rholes, Riskind, & Lanke, 1987; Riskind, Rholes, & Eggars, 1982), some participants were induced to experience depression by asking them to express a series of self-referent statements (Velten, 1968). In some cases, the statements were self-deprecating (e.g., “I am worthless”), whereas in other cases, they referred to unpleasant somatic conditions (e.g., “I am tired”). The two procedures were equally effective in inducing depression. However, only the first procedure influenced participants’ recall of mood-congruent personal experiences. This indicates that affect per se was not sufficient to account for the differential recall of participants’ past experiences in the two conditions. A series of studies by Parrott and Sabini (1990) are particularly provocative. In one study, participants listened to either uplifting or depressing music under explicit instructions to use the music to get into a “happy” or “sad” mood. These participants subsequently recalled experiences in high school that were affectively congruent with the mood they were in, replicating Bower’s (1981) results. In other conditions, however, participants listened to the same music for the ostensible purpose of judging its aesthetic quality, and no mention was made of the mood state being manipulated. In this case, participants recalled high school experiences that were affectively incongruent with the mood they were in. Naturally occurring mood states (resulting from sunny or rainy weather, or from good or bad performance on an exam) also led to the recall of mood-congruent past experiences.
24
R. S. WYER, Jr., ET AL.
The reason for the recall advantage of mood-incongruent knowledge when participants’ mood was not explicitly labeled is somewhat speculative. However, results obtained in the aforementioned study by Strack et al. (1988) suggest an explanation. Specifically, an unobtrusive manipulation of the positive affect that participants were experiencing as they read a series of cartoons increased their estimates of how amused they were by the cartoons. In other words, they used their affect as a basis for these estimates. However, when participants were asked to judge cartoons they had seen at an earlier point in time, they used the affect they were experiencing at the time of judgment as a standard of comparison in construing the implications of the feelings they had had to the cartoons they had seen earlier. Thus, they reported being relatively less amused when they were experiencing positive affect than when they were not. A similar process might underlie the results obtained by Parrott and Sabini. That is, when people are asked to recall an experience they had in high school, the concept “high school” may spontaneously elicit affective reactions. These reactions are likely to be more negative, on average, than the feelings that happy individuals were experiencing at the time of judgments, and may consequently elicit thoughts that (e.g.), “I felt less happy in high school than I do now.” Correspondingly, “high school” may elicit more positive feelings than unhappy persons are experiencing and may lead them to think that “I felt happier in high school than I do now.” These thoughts may then cue the recall of experiences that exemplify them, producing the mood-incongruent retrieval effect that Parrot and Sabini observed. When participants were explicitly told to get into a happy or sad mood, however, the semantic concepts activated by these instructions may have cued the retrieval of knowledge that exemplified them, and these effects may have offset the effects of thought processes that would otherwise have occurred. In any event, Parrott and Sabini’s results strongly suggest that the recall of previously acquired knowledge is not cued by participants’ affective reactions per se. Thus, they are quite consistent with Postulate 1.# The assumption that affective reactions generally cue memories of events that elicited these reactions is also compromised by the apparent asymmetry of their effects. That is, happy moods facilitate the recall of happy events, The rnood-incongruent retrieval effects obtained by Parrott and Sabini were restricted to the first experience that participants recalled; the second one they recalled was typically mood-congruent. However, this reversal can also be explained in terms of the mediating impact of the thoughts that participants were likely to have in the situations being investigated. For example, happy persons who were stimulated to think that “I did not feel as good in high school as I do now,” and to recall an experience that confirmed this thought, may then have been stimulated to think that “it wasn’t all [hut bad,” and to recall a favorable (moodcongruent) experience that confirmed this thought. Unhappy person’s cognitive responses might have been analogous. (For an elaboration of this speculation, see Wyer & Srull, 1989.)
AFFECT AND INFORMATION PROCESSING
25
but sad moods do not facilitate the recall of unhappy ones (Natale & Hantas, 1982; Salovey & Singer, 1985). This asymmetry might be attributed to motivational factors (Isen, 1984; Singer & Salovey, 1988). That is, happy persons may be motivated to recall happy events in order to maintain their mood, whereas unhappy individuals are motivated to avoid recalling unhappy events. This interpretation also argues against the assumption that the retrieval of affect-congruent concepts and knowledge is stimulated by affective reactions per se. The research described above concerned the effect of experimentally induced affect on information retrieval. Many other studies have looked at the impact of chronic affective states (e.g., depression) on memory (for a review, see Blaney, 1986). However, the results of this research are somewhat inconsistent, and their relevance to Postulate 1 is hard to evaluate. The difficulty is similar to that encountered in laboratory studies. For example, depressives and nondepressives are likely to differ not only in the affect they are experiencing but also in the semantic concepts and episodic knowledge that are chronically accessible in memory. Therefore, differences in recall could easily be due to the cuing properties of these cognitions rather than to affect per se. Summary. The results reported by Parrott and Sabini (1990), and Bower et al.’s (1981) failure to obtain postinformation mood effects on information recall, directly contradict the proposition that affective reactions per se influence the retrieval of declarative knowledge from memory. Other studies have identified effects of mood-induction procedures on the interpretation, encoding, and retrieval of information that is similar in valence to participants’ moods but is not semantically related to affective states. However, Niedenthal and Setterlund’s (1994) findings imply that these effects are probably due to factors other than affective reactions per se. In light of these considerations and the generally mixed empirical support for moodcongruent effects on retrieval (for reviews; see Blaney, 1986; Bower, 1991; Mayer et al., 1990; Riskind, 1989), it seems reasonable to consider Postulate 1 to be viable. B. THE ROLE OF AFFECT IN ACTIVATING PROCEDURAL KNOWLEDGE (PATH 9) Affective reactions are likely to play a more important role in the activation of procedural knowledge than in the accessibility of declarative knowledge. As we noted earlier, this knowledge can be conceptualized metaphorically as having the form of “IF [XI, THEN [Y]” productions, where [Y] denotes a sequence of cognitive or motor acts and [XIdenotes a configura-
26
R. S. WYER, Jr., ET AL.
tion of internal stimulus features that, in combination, serve as a precondition for eliciting this sequence. The precondition could be a mixture of perceptual codings of external stimuli, internally generated thoughts or concepts, and bodily sensations. When the conditions specified in the configuration are met, the sequence of activities associated with them is automatically initiated. Some productions are situation- and response-specific (e.g., “IF a person returns a lost wallet and one’s goal is to form an impression of this person, THEN interpret the behavior as honest”). Others are more general (Smith, 1990). For example, the kind of undifferentiated appraisal of a situation that Lazarus (1982) postulated might stimulate a general (nonsituationspecific) response strategy. Procedures can be acquired in the course of deliberative goal-directed cognitive activity. Over time and repetition, however, they may be activated and performed with a minimum of conscious cognition mediation (E. Smith, 1990, 1994). Affective reactions, like perceptions, thoughts, and other internally generated stimuli, can be parts of a configuration of stimuli that in combination serve as a precondition for a production.9 On the other hand, affective reactions are unlikely to be the only precondition. Cognitions about the specific situation at hand or task to be performed are also included. Without knowing the nature of these situation-specific cognitions, the role of affect in procedural knowledge can therefore be hard to predict. Nevertheless, the following postulate seems viable: Postulate 2 The affective reactions that one experiences, along with concepts and cognitions that are activated by an appraisal of a stimulus situation, can compose a precondition for a previously learned sequence of responses (i.e., a cognitive production). This response sequence may be applied spontaneously, with little conscious cognitive mediation, under conditions in which the precondition is met. Despite the vagueness of this postulate, some general predictions seem tenable. Bless et al. (1996) postulate that positive affect is associated with
’
Although the features of a precondition can sometimes correspond to representations that are stored in memory as part of declarative knowledge, their role is quite different. Moreover, they can function independently of this knowledge. Bargh, Chen, and Burrows (1996, Experiment I), for example, found that the activation of semantic concepts can influence situation-specific behavior without affecting the use of these concepts to interpret information to which they are relevant. In the present context, affective reactions could likewise be part of a precondition that elicits behavior independently of its association with concepts that compose declarative knowledge.
AFFECT A N D INFORMATION PROCESSING
27
a disposition to use broad concepts and general knowledge structures to perform experimental tasks, whereas negative affect is associated with attention to situational details. They assume that these differences result from a previously learned association of positive affect with conditions in which success has not required a careful scrutiny of the situation or task to be performed. In contrast, negative affect is often associated with failure and negative behavioral consequences that can only be avoided by attention to situational details (cf. Schwarz, 1990). To this extent, positive affect (along with other situation-related cognitions) may often be part of a learned precondition for the use of general (e.g., categorical) criteria in responding to an achievement task, whereas negative affect may be a precondition for a more detailed analysis of the task requirements and the information bearing on them. It seems reasonable to suppose that these alternative productions are elicited spontaneously when the preconditions for them are met and are performed with little if any conscious cognitive mediation. Note that the different productions activated by positive and negative affect require the use of different types of knowledge. If the procedures activated by positive affect involve the use of general concepts and knowledge, they may stimulate a search of memory for representations of this type. In contrast, the procedures activated by negative affect may require more specific concepts that can be used to interpret informational details. At least two studies provide evidence that this is so. Bless et al. (1996) reasoned that if people who experience positive affect spontaneously activate general knowledge structures to use in processing information, they should be able to process this information more efficiently. Moreover, their use of these knowledge structures should free cognitive resources for use in other, unrelated cognitive activity. To evaluate this possibility, participants who were induced experimentally to feel happy or unhappy listened to two tape-recorded stories about daily life events (e.g.. making a telephone call from a public phone booth) while they performed a clerical task. Both recognition memory for the events mentioned in the audiotapes and accuracy in performing the clerical task were recorded. As expected, happy participants were more likely than unhappy ones to use a prototypic script (Schank & Abelson, 1977) to comprehend the taped information. That is, happy participants were more likely than unhappy ones to report prototype-consistent events as having been mentioned, but were less likely than unhappy participants to report atypical events. To the extent that participants apply a general prototype in comprehending the tape-recorded information, however, they should be able to devote more cognitive resources to the clerical task. This was also the case. Happy
28
R. S. WYER, Jr., ET AL.
participants performed better on the secondary, clerical task than unhappy participants did. A quite different study by Dienes (1996) has similar implications. In this study, happy and unhappy participants were asked to learn a list of words that exemplified either easily identifiable concepts (animals, countries, etc.) or obscure ones (e.g., things made of wood). Learning the list could be facilitated by attending to the details of the individual words, by mentally organizing the words in terms of the concepts they exemplified, or both. (For evidence that the organization of information items into categories improves learning, see Hunt & Einstein, 1981). The author reasoned that when the concepts exemplified by the stimulus words were obvious, both happy and unhappy participants would spontaneously identify and use these concepts to organize the stimuli. When the concepts were more obscure, however, only happy participants should spontaneously identify them, so only these participants’ learning of the stimulus words should be facilitated. Consistent with this reasoning, happy participants had better memory than unhappy participants €or the hard-to-categorize words but not for the easyto-categorize words. The conclusions that persons who experience positive affect bring categorical and general knowledge structures to bear on tasks they perform does not imply that affective reactions per se cue the retrieval of these knowledge structures. Rather, this retrieval is mediated by the impact of affect on the activation of cognitive productions that require different types of declarative knowledge. Note that the activation of a production also depends on the presence of other, situational features that make up the precondition for eliciting it. The cognitive productions that are activated and applied in some situations do not always require the types of knowledge of concern in the studies by Diener and Bless et al. In other cases, affective reactions might not be a feature of the preconditions that elicit goal-relevant productions and, therefore, might have no influence at all. In this regard, Murray, Sujan, Hirt, and Sujan (1990) found that although happy participants used fewer categories to classify stimuli than control participants when they were told t o focus on similarities, they used relatively more categories to classify the stimuli when they were told to focus on differences. These latter findings qualify the conclusion that happy persons typically use broader categories to classify stimuli than unhappy persons. Further experiments by Murray et al. indicated that differences in categorization observed in both task conditions were likely to result from a more general tendency for positive affect to increase the number of attribute dimensions that participants brought to bear on the particular task they were asked to perform. Thus, it increased their ability to make distinctions among the stimuli they were considering as well as to identify similarities.
AFFECT AND INFORMATION PROCESSING
29
As Murray et al. suggest and we elaborate presently, these results could reflect mood-induced differences in the motivation to perform the task at hand rather than differences in the cognitive productions that participants spontaneously apply under the conditions being investigated. The cognitive procedures we have postulated in this section are presumably activated and applied spontaneously, without conscious deliberation. Other formulations (e.g., Isen & Shalker, 1982; Mackie & Worth, 1989) assume that affect influences the strategies that are employed through its mediating impact on the motivation or ability to engage in concentrated cognitive activity. These influences could also be reflected in Path 9, as described in Figure 1. However, the conceptualization we propose assumes that these influences are actually mediated by the impact of affect on response evaluation (Path 11).These alternative interpretations are evaluated in later sections of this chapter.
C. CONCLUSIONS The results summarized in this section are compatible with our contention that except under very restricted conditions (cf. Niedenthal et al., 1997), affective reactions per se do not influence either the retrieval of affectcongruent semantic concept and declarative knowledge (Path 7) or the selective attention to and interpretation of new information (Path 8). Affective reactions can nevertheless influence the accessibility and use of declarative knowledge indirectly through its mediating impact on the cognitive procedures that persons spontaneously activate and apply in the course of attaining a particular cognitive objective (Path 9). That is, these reactions in combination with features of the task to be performed can spontaneously activate a procedure that requires a certain type of declarative knowledge (Bless et al., 1996; Dienes, 1996). However, the procedure that is activated, and the type of knowledge it requires, is likely to depend on the type of task being performed and the situational conditions that surround it. Consequently, general conclusions concerning the influence of affective reactions on the accessibility of different types of knowledge should be treated with caution.
V. The Use of Affect as Information in Making Judgments (Path 10) The outputs of information processing are often judgments. These judgments, which could either be reported overtly or remain covert, could
30
R. S. WYER, Jr., ET AL.
concern oneself, another person, an event, the consequences of one’s behavior, or an abstract concept. The affective reactions that are elicited by a referent can sometimes be a basis for judging it. However, when and how this is done depends on the type of judgment to be made. Earlier in this chapter, we distinguished between affect-based evaluations and descriptively based evaluations. Whereas evaluations of the first type are based primarily on one’s feelings toward the target being judged, evaluations of the second type are only based on affect when more directly relevant criteria are either unavailable or difficult to apply. This difference has sometimes been ignored in formulations of the role of affect in inference processes. Forgas (1992, 1995), for example, assumes that affect functions as a heuristic that is applied only when people are either unable or unmotivated to employ other, nonaffective criteria. Schwarz and Clore (1988) also refer to a “how-do-I-feel-about-it?’’ heuristic in conceptualizing the informational influence of affect. Although circumstances arise in which such a heuristic is applied, we believe that affective reactions are often the primary basis for many judgments and are not simply used when other criteria are unavailable. People’s use of their affective reactions as a basis for judgments is complicated by the fact that the feelings they experience at the time of judgment can be elicited by stimuli other than the ones they are judging. These feelings could result from judgment-irrelevant thoughts that occurred before the target stimulus was considered. Unfortunately, people cannot clearly distinguish between the affect that is elicited by one source and the affect that is elicited by another. Consequently, the affect they experience for reasons that have little to do with the object they are evaluating can often have an impact on the judgments they report. When people’s attention is called to the fact that the affect they are experiencing could be influenced by judgment-irrelevant factors, they may attempt to correct for this influence. As Schwarz (1990) points out, this correction can occur in two ways. In some cases, persons may simply discount their feelings and employ alternative criteria. When alternative criteria are unfavorable or difficult to apply, however, people may be forced to estimate the magnitude of the affect they are experiencing for judgmentirrelevant reasons and adjust their judgments to compensate for it. These estimates are not always accurate. If persons do not adjust enough, extraneous sources of affect will continue to influence their judgments. If, on the other hand, they adjust too much, this extraneous affect will have a negative, contrast effect on their judgments. The conditions in which adjustments and discounting processes are most likely to occur are conveyed in the following two postulates.
AFFECT AND INFORMATION PROCESSING
31
Postulate 3 Persons use the affect they are experiencing to make judgments of their feelings about themselves, other persons, objects, or events, or to make other judgments for which these feelings are the primary criterion. If they perceive that the affect they are experiencing is due in part to judgment-irrelevant sources, they will adjust the judgments they report to compensate for the influence of this extraneous affect. Postulate 4 People may use the affect they are experiencing as a heuristic basis for evaluative judgments that are not typically based on affective reactions when they are either unable or unmotivated to use other, more directly relevant criteria. This will not be true, however, if they perceive that the affect they are experiencing is due in part to judgment-irrelevant factors and if nonheuristic bases for judgment are available. Then, they will discount their affect entirely as a source of information and base their judgments on other, descriptive criteria. Methodological Considerations. The use of affect as an informational basis for judgments has typically been investigated using procedures originally employed by Clore and his colleagues (Clore & Byrne, 1974; Gouaux, 1971; Griffitt & Veitch, 1971; and refined by Schwarz and Clore (1983). Specifically, suppose that before people are asked to make a judgment to which affect is relevant, they are induced to feel either happy or unhappy for reasons that are unrelated to this judgment. Then, because they cannot easily segregate the affect from this extraneous source from the affect elicited by their thoughts about the object being judged, this irrelevant affect should have an impact on the judgment they make. On the other hand, suppose persons do not consider affect to be a basis for their judgment. Then, neither their affective reactions to the object they are judging nor the affect they are experiencing for other reasons should influence the judgments they make. In short, the extent to which extraneous sources of affect influence judgments provides a basis for inferring whether affect in general is considered to be a relevant judgmental criterion. A. AFFECT-RELEVANT JUDGMENTS The assumption that affective reactions provide the basis for judgments is hardly of recent vintage. Thurstone’s (1959) attitude scaling procedures
32
R. S . WYER. Jr.. ET AL.
were intended to assess persons’ affective reactions to a concept independently of the content of their beliefs about it. The assumption that persons base their liking for someone on their affective reactions to the individual also underlies Byrne’s (1971; see also Clore & Byrne, 1974) similarityattraction hypothesis. Despite several earlier attempts (e.g, Gouaux, 1971; Griffitt & Veitch, 1971), it remained for Schwarz and Clore (1983) to demonstrate convincingly that affect can have informational impact on judgments independently of the situational and informational conditions that produce it. In these studies, persons who were experiencing positive or negative affect as a result of either an experimental manipulation (e.g., writing about a happy or sad event) or the weather were asked to judge their life satisfaction. Happy participants reported greater satisfaction than unhappy ones. However, when participants’ attention was called to a judgmentirrelevant reason for the affect they were experiencing, this difference disappeared. Increasing participants’ awareness of this alternative explanation for the affect they were experiencing would have had little effect unless they were otherwise disposed to use this affect as a basis for their judgments. Further evidence that affective reactions are used as a basis for judgments to which they are relevant was obtained by Strack, Schwarz, and Gschneidinger (1985). They found that when persons recalled pleasant and unpleasant past experiences in ways that did not elicit affective reactions, they used these experiences as standards of comparison in evaluating their present life satisfaction. Thus, they reported greater satisfaction after recalling an unpleasant experience than after recalling a pleasant one. However, when the experience that participants recalled elicited affect, it had a positive influence on their judgments similar to that observed by Schwarz and Clore (1983). In other words, the influence of the affect elicited by the experience offset the contrast effects that resulted from a consideration of its descriptive implications. A particularly impressive demonstration of the influence of affect on judgments was performed by Strack et al. (1988) in an experiment described earlier. Briefly, participants were asked to estimate how amused they were by a number of cartoons while holding a felt-tip pen either between their teeth (a procedure that activated muscles involved in smiling or between the lips (which activated muscles involved in frowning). This proprioceptive stimulation was expected to elicit positive or negative affective reactions, respectively (see Path 2, Figure 1). Although participants were totally unaware that their facial expressions were being manipulated, they reported being more amused by the cartoons in the first condition than the second. Supplementary data provided evidence that participants considered their
AFFECT AND INFORMATION PROCESSING
33
affective reactions t o be the primarily reievant basis for their judgments. If they were simply being used as a heuristic, these reactions should have influenced all evaluative judgments that participants were asked to make. In fact, facial feedback only influenced judgments of amusement (an estimate of positive feelings toward the cartoons) it did not influence judgments of how funny the cartoons were in general (estimates that may have been based in part on normative considerations). Therefore, the data argue against Forgas’s (1992, 1995) assumption that the only direct influences of affect on judgments are heuristic. 1. Corrections for Bias in Making Affect-Based Judgments Postulate 3 asserts that when affect is the primary basis for judgment, people make these corrections by first computing a judgment on the basis of the affect they are experiencing and then adjusting this estimate to compensate for the bias they perceive to exist. Schwarz and Clore (1983) provide evidence for these adjustment processes. In one experiment, affect was induced by having participants write about a happy or sad experience in a rather odd-looking sound-proof booth. Some participants were forewarned that the booth might make them feel rather depressed; others were told that it might make them feel somewhat elated, and others were given no information at all about the booth’s effects. Later, participants reported their life satisfaction. Participants who were given no information about the booth’s effects reported less satisfaction with their life after writing about an unhappy experience than after writing about a happy one. Moreover, telling unhappy participants that the booth might make them unhappy (thus producing an judgment-irrelevant explanation for their feelings) increased their ratings of life satisfaction to a level similar to that of happy participants, whereas telling them that the booth would make them elated decreased their estimates of life satisfaction even further. Thus, these participants not only based their judgments on the affect they were experiencing but adjusted for the impact of affect they considered to be irrelevant. In contrast, happy participants judged themselves to be equally satisfied with their lives in all conditions, regardless of how the booth was characterized. The nonparallel effects of positive and negative affect could reflect a more general difference in the assumptions that people make about how much they need to correct for the bias produced by different types of affect (for evidence of differences in persons’ implicit theories of correction in other judgments domains, see Petty & Wegener, 1993; Wegener & Petty, 1995,1997). Although no evidence bears directly on this possibility, a study by Wyer and Budesheim (1987) is suggestive. They found that participants
34
R. S. WYER, Jr., ET AL.
who thought they might be influenced by judgment-irrelevant information about a person adjusted their judgments to a lesser extent when the implications of the biasing information were favorable than when they were unfavorable. People apparently expect others to have favorable attributes and, therefore, perceive information that confirms these expectations to be less biasing. An analogous difference could exist in the situations considered by Schwarz and Clore. That is, suppose people normally expect to have positive feelings about a person they are judging (themselves or someone else). If this is so, they may perceive judgment-irrelevant positive affect to be less biasing than judgment-irrelevant negative affect, and may adjust their judgments less to compensate for it. Findings reported by Gasper and Clore (1998) may qualify this conclusion, however. They found that chronically anxious persons do not adjust for the influence of transitory situational factors on the negative affect they are experiencing. Apparently, calling these individuals’ attention to extraneous sources of negative affect makes salient to them that their feelings are not due to transitory factors alone. Consequently, it increases the influence of the negative affect they are experiencing rather than decreasing it. 2. Effects of Motivation and Ability in Adjustment for Bias Persons’ adjustment for the biasing influence of extraneous affect may depend on not only their perceptions of how much adjustment is necessary, but also their ability and motivation to make this adjustment. In fact, an alternative explanation of Schwarz and Clore’s findings is that persons are simply not motivated to adjust for biasing influences of information that has favorable implications for themselves. Support for this contention was obtained by Arkin, Gleason, and Johnston (1976). They found that although persons took responsibility for failure only if situational factors that could account for this outcome were unavailable, they took responsibility for success regardless of whether alternative explanations for the outcome were provided. Adjustments for bias may also depend on the ability to estimate the amount of correction that is necessary. In a study by Albarracin and Wyer (1999), some participants were distracted as they read an article describing the consequences of instituting comprehensive examinations at their university. These participants could not accurately assess their cognitive reactions to the message content and, therefore, confused it with the feelings they were experiencing because of their mood. As a result, they reporting more favorable attitudes toward the position advocated when they were happy than when they were not. Nondistracted participants, on the other hand,
AFFECT AND INFORMATION PROCESSING
35
were better able to assess the affective implications of the message content and, at the same time, to adjust (in fact, to overadjust) for the biasing influence of the message-irrelevant affect they were experiencing. These participants reported less favorable attitudes toward the position when they were happy than when they were not. Further evidence of the effects of motivation and ability on adjustment processes was obtained by Isbell and her colleagues (Isbell & Wyer, 1999; Ottati & Isbell, 1996). In these studies, participants were asked to estimate their liking for a political candidate on the basis of his stands on a number of social issues. If participants had either little knowledge about politics or little motivation to make an accurate judgment, they evaluated the candidate more favorably when they were happy than when they were not. However, if participants were high in political expertise or were motivated to be accurate, they reported less favorable evaluations of the candidate when they were happy. Thus, participants’ motivation and ability to think carefully about the evaluative implications of the information they received stimulated them to correct for the influence of judgment-relevant affect. In doing so, however, they adjusted too much, producing a contrast effect of this affect on the evaluations they reported. Further evidence indicated that these adjustments occurred “on-line” as each piece of information was presented and evaluated rather than at the time participants reported their judgments. Thus, their effects persisted long after the mood-related affect that participants experienced at the time they received the information had dissipated. In fact, on-line influences of affective reactions may be more the rule than the exception. In responding to persuasive communications that contain several different arguments, recipients may construe their affective reactions to the implications of each argument separately at the time they are exposed to it, and the overall attitude they report may reflect a composite of these component reactions (Fishbein, 1963). Similarly, persons who make a purchasing decision may experience both positive and negative affective reactions to specific features of the product at the time they learn about them, and their final decision may reflect a composite of these reactions (Adaval, 1996). This possibility is considered in more detail presently. Note that our conceptualization assumes that the use of affect as information and corrections for its influence are conscious processes. A contrary position has been taken by Winkielman et al. (1997) in a study described earlier (see also Zajonc, 1980). Briefly, Winkielman et al. showed that subliminally priming happy and angry faces influenced participants’ evaluations of Chinese ideographs even though the participants did not report experiencing affective reactions at the time they made these evaluations. Moreover, making salient alternative explanations for the affect they were
36
R. S. WYER, Jr., ET AL.
experiencing had no impact on the influence of the priming stimuli. Winkielman et al. concluded that participants were unconsciously using the affect elicited by the priming stimuli as a basis for their judgments of the ideographs and thus were unable to correct for its influence. As we pointed out earlier, however, these findings could reflect a spontaneous categorization process of the sort we postulated to underlie the automatic evaluation effects identified by Bargh (1997), and might not have anything to do with the use of affect per se as an informational basis for judgments.
B. JUDGMENTS TO WHICH AFFECT IS NOT DIRECTLY RELEVANT According to Postulate 4, affective reactions are most likely to be used to make descriptivejudgments when either (a) relevant descriptive information is unavailable or difficult to retrieve from memory, or (b) the implications of this information are hard to compute. In such conditions, affect may provide a heuristic basis for judgment as postulated by Schwarz and Clore (1988). When more directly relevant judgmental criteria are easily accessible in memory, however, these processes may not occur. Levine, Wyer, and Schwarz (1994) obtained support for this possibility. In this study, some participants first wrote about either a happy or an unhappy achievement experience, whereas others wrote about either a happy or unhappy social experience. Then, all participants judged their competence in both the domain to which the recalled experience was relevant and the domain to which it was not. Participants were expected to base their judgments on the descriptive implications of a relevant past experience when such an experience came to mind easily, but to use their affect as a basis for judgment when such an experience was not immediately accessible. Content analyses indicated that participants took personal responsibility for the achievement experiences they reported regardless of their favorableness. However, they attributed the social experiences they reported either to other persons or to external circumstances. Thus, the achievement experiences that participants recalled had descriptive implications for their achievement competence, whereas the social experience they recalled had little relevance to their social competence. Consistent with this finding, participants’ judgments of their achievement competence were influenced by the achievement experience they recalled, but their estimates of social experience were not af€ected by the social experience they recalled. In contrast, both types of recalled experience (or, more accurately, the affect elicited by them) influenced participants’ judgments of their competence
AFFECT AND INFORMATION PROCESSING
37
in the domain to which their recalled experience did not pertain. In other words, participants based their competence judgments on the descriptive implications of a judgment-relevant past experience when such an experience had been made salient to them as a result of the story they wrote. When a relevant past experience did not immediately come to mind, however, they based their judgments on the affect they experienced as a result of writing a judgment-irrelevant story, independently of the story’s descriptive implications. Other studies indicate that when people can easily recall information that has direct implications for their judgments, affective reactions that are only indirectly relevant to these judgments have little effect. Schwarz, Strack, Kommer, and Wagner (1987), for example, found that although participants’ affective reactions to a clean or dirty experimental laboratory room influenced their estimates of general life satisfaction, these reactions did not affect their evaluation of their personal living quarters. In the latter case, participants used the experimental room as a comparative standard in judging their place of residence independently of the affect it elicited. Thus, these latter judgments were apparently based on descriptive criteria rather than affect. The use of affect as a heuristic basis for judgment could underlie the effects of wishful thinking (McGuire, 1960; see also McGuire & McGuire, 1991). That is, people believe that things they like are going to occur and that things they dislike are improbable. Consistent with this tendency, Albarracin and Wyer (1997) found that when persons were distracted from thinking about the implications of a behavior-relevant persuasive message, the message-irrelevant affect they were experiencing influenced both their attitudes toward the behavior and their beliefs that the behavior would have desirable consequences. When participants could carefully evaluate the content of the message, however, they based their beliefs in the behavior’s consequences on descriptive implications of its content, and the affect they were experiencing had less influence. Thus, as in the studies by Schwarz, Levine, and others, participants were inclined to use judgmentirrelevant affect as a basis for beliefs only when the implications of the descriptively relevant information were inadequate. In summary, people appear to use their affective reactions as a basis for judgments in a wide variety of situations. This could occur not only when people are explicitly requested to make judgments but also spontaneously, when they make a decision that requires an assessment of their feelings toward a behavior or an object toward which the behavior is directed. To this extent, affect can influence performance on achievement tasks that require an evaluation of task-related activity or its consequences. We elaborate this possibility presently.
38
R. S. WYER, Jr., ET AL
VI. Effects of Affect on Motivation and Ability (Path 9) The cognitive procedures that are elicited by positive and negative affect are theoretically activated and applied automatically (Postulate 2). In contrast, other theories implicitly or explicitly assume that the impact of affect on the use of different information-processing strategies is more deliberative, and is mediated by its influence on people’s ability or motivation to think carefully about the information they receive. These theories purport to explain the tendency for people to process information less extensively when they are happy than when they are not. As we will indicate, there is some question regarding the generalizability of this tendency and, in fact, whether it even exists. Before discussing the implications of our own conceptualization, alternative theories and the evidence bearing on them are worth reviewing briefly.
A. THEORETICAL PERSPECTIVES 1. The Influence of Affect on Cognitive Capacity Mackie and Worth (1989; see also Isen, 1987; Worth & Mackie, 1987) postulate that positive affect is associated in memory with a large amount of previously acquired knowledge and that it activates this knowledge spontaneously whenever it is experienced. Once activated, the knowledge takes up capacity and, therefore, decreases the cognitive resources favorable for other, goal-directed information processing. As a result, the ability to evaluate the implications of new information is diminished. In support of this contention, Mackie and Worth found evidence that when the opportunity to think about a persuasive message was limited, happy participants were less influenced by the strength of arguments contained in the message than others were. When participants were given as much time as they wanted to think about the message, however, this difference was eliminated. 2. The Impact of Affect on Motivation
The affect that people experience could also influence their motivation to think extensively about the information they receive. Isen and Shalker (1982), for example, assumed that because persons who feel happy are motivated to maintain these pleasant feelings, they avoid engaging in demanding cognitive activity that is likely to bring them down. In contrast, unhappy persons are motivated to get rid of their unpleasant feelings (Isen, 1984). Consequently, they often try to distract themselves from thinking
AFFECT AND INFORMATION PROCESSING
39
about the events that caused their unhappiness. For example, they might watch a movie, engage in physical exercise or, in a laboratory experiment, concentrate more fully on the tasks they are asked to perform. In combination, these considerations also imply that people are less inclined to evaluate carefully the implications of information they receive when they are happy than when they are unhappy. This should be particularly true when they expect the cognitive activity required to process the information to be aversive. If this reasoning is correct, these processing differences should be eliminated or even reversed by leading happy persons to believe that thinking about the information they receive will be enjoyable and, therefore, will maintain or enhance their positive mood state. Wegener, Petty, and Smith (1995) showed that this was indeed the case. Participants listened to one of two communications. One message advocated a position with which participants were inclined to disagree, and was introduced with instructions that it was likely to make them unhappy. The content of this message had less impact on participants’ attitudes when they were happy than when they were not. The other communication advocated a position that participants typically favored, and was introduced with a comment that reading it was likely to be enjoyable. In this case, the message content had greater influence on participants’ attitudes when they were happy. Wegener et al.3 (1995) results therefore qualify the conclusion that persons who experience positive affect are generally unmotivated to engage in extensive cognitive processing. There are also qualifications on the assumption that people who experience negative affect are motivated to process new information extensively. For one thing, the distraction provided by this cognitive activity may only help to eliminate negative affect if the affect is mild and if nothing can be done about the events that produced it. Extremely unhappy individuals are likely to ruminate about the events that created their unhappiness instead of distracting themselves by engaging in irrelevant activity (Ellis & Ashbrook, 1988; Klinger, 1975; Martin & Tesser, 1996). To this extent, they may be just as insensitive to details of the affect-irrelevant information they receive as happy persons are. 3. ~otivafional Consequences of Using Affect as Informat~on
A somewhat different conceptualization of the impact of affect on information processing was proposed by Schwarz and his colleagues (Schwarz, 1990; Schwarz & Bless, 1991; Schwarz, Bless, & Bohner, 1991). They note that the experience of positive affect is typically associated with pleasant life events and success in goal-directed activity, whereas the experience of negative affect often accompanies unpleasant events and failures. As a
40
R. S. WYER, Jr., ET AL.
result of these past associations, people often interpret the positive affect they experience in a new situation as an indication that the situation does not require careful scrutiny, but interpret the negative affect they experience as a sign that the situation is potentially problematic and requires attention. Consequently, they may engage in less extensive processing of situationrelevant information in the first case than in the second. This hypothesis, therefore, provides yet another explanation of the people’s tendency to be less sensitive to the strength of arguments conveyed in a persuasive communication when they are happy than when they are not. B. EMPIRICAL AMBIGUITIES In principle, the factors postulated by the three theories outlined above could contribute independently to the overall influence of affect on information processing. Nevertheless, some evidence appears to contradict the assumptions that underlie each of the formulations. Particularly questionable is the hypothesis that affect influences cognitive capacity. For one thing, this hypothesis assumes that affect influences the accessibility of a diverse amount of declarative knowledge. This assumption appears to be invalid for reasons identified by Niedenthal and Setterlund (1994) and noted earlier. The hypothesis is also incompatible with evidence that persons who are asked to perform two tasks simultaneously often do better on these tasks if they are happy than if they are unhappy (Bless et al., 1996); this would not be true if happy individuals’ cognitive resources were limited. Wegner et al.’s (1996) finding that happy persons are more sensitive to the content of a persuasive message when they expect it to be enjoyable is also problematic for this hypothesis. Finally, cognitive theory and research raises more general questions concerning the validity of the assumption that processing differences of the sort that Mackie and Worth consider reflect differences in the cognitive resources that participants have available (for a review, see Logan, 1997). On the other hand, evidence that affect influences the motivation to evaluate details on the information presented is also questionable. In the aforementioned study by Bless et al. (1990), happy participants’ attitudes were less affected than sad participants’ attitudes by the strength of arguments contained in a persuasive message. However, thought-listing data indicated that happy and unhappy participants did not differ in the total number of message-related thoughts they generated in response to the message. Rather, they only differed in the type of thoughts they reported; sad participants elaborated the positive implications of strong arguments, whereas happy participants attempted to refute their validity. Thus, both
AFFECT AND INFORMATION PROCESSING
41
groups of participants were equally motivated to think about the message content but differed in how they thought about it. Two studies of impression formation by Isbell, Clore, and Wyer (1998), described later in this chapter, have similar implications.
C. SUMMARY The fact that affective reactions influence the impact of information on judgments and decisions is well established. However, general differences in either the capacity or the motivation to engage in analytic information processing cannot explain why and when the influence occurs. A somewhat different version of the affect-as-information hypothesis proposed by Schwarz and Clore (1983,1988) provides a viable account of the effects of affective reactions on goal-directed information processing and has additional implications as well. This application is discussed in the next section.
VII. The Influence of Affect on Response Evaluation (Path 11): A Performance-Feedback Model of Affect as Information According to the feedback conceptualization of goal-directed information processing we propose (see Figure l), people spontaneously monitor their behavior and the outputs it generates, to determine if they have attained the goal they are pursuing, or, if not, whether the strategy they are using to pursue the goal is appropriate. If they deem the results of their activity to be sufficient, they terminate their goal-directed activity. Otherwise, they continue processing, using the same or a different strategy, until they generate an output that they consider to be satisfactory. When objective criteria exist for deciding whether a goal has been attained, the decision to engage in a goal-directed activity is obviously based on these criteria. More often than not, however, the criteria for goal attainment are ambiguous. A woman who is writing a term paper, for example, is likely to have little objective basis for deciding whether the amount and quality of her output is sufficient. Therefore, her decision to stop or continue writing may be based in part on how she feels about the work she has done at the time these feelings are assessed. Similarly, people’s decisions to continue using a given strategy or to try a different approach could also be based on their feelings about the adequacy of their initial attack on the problem at hand.
42
R. S. WYER, Jr., ET AL.
There is an important contingency in the validity of this hypothesis, however. That is, people sometimes engage in behavior because it is intrinsically enjoyable, independently of its instrumentality in attaining an external goal. In this case, the affect they are experiencing may have different implications than those described above. If the goal of the student in our earlier example is to write a paper of high quality, she may implicitly ask herself at some point if she feels it has met this criterion. Then, she is more likely to answer her self-generated question affirmatively, and therefore to stop writing sooner, if she is experiencing positive affect than if she is not. However, suppose she focuses her attention on whether she is having fun working on the paper rather than the quality of her output, and asks herself if she feels that this is the case. Then, she may be more inclined to answer her question affirmatively, and continue writing longer, if she is experiencing positive affect than if she is not. These examples suggest the possibility that people use the affect they are experiencing at the time they are engaging in goal-directed activity as information in evaluating the effectiveness of their goal-directed activity. This evaluation can bear on two aspects of this activity. First, it can concern whether the outputs that have been generated up to that time are sufficient for attaining the goal. Second, it could concern whether the strategy being used to generate outputs is likely to be effective in attaining their objective or whether an alternative strategy should be adopted. The way in which affect is brought to bear on these concerns depends on the nature of the goal being pursued and the type of activities that are involved in its pursuit. Certain of these contingencies were first identified by Martin, Ward, Achee, and Wyer (1993). However, Martin et al. focused on only the use of affect as a basis for deciding whether to continue engaging in goaldirected activity or to terminate it. Our conceptualization also takes into account its use as information about the appropriateness of the strategy one has chosen to attain a particular objective. As a result, it has additional implications that are not addressed by the original theory. The present formulation is embodied in two postulates: Postulate 5 When people engage in behavior for the purpose of attaining an external goal and objective criteria for inferring the attainment of this goal do not exist, they use the affect they are experiencing to infer whether they feel that the strategy they have used in pursuit of this goal, and the outputs they have generated by employing it, are appropriate. Therefore, they are more likely to infer that they have met the criterion for satisfactory performance if they are experiencing positive affect than if they are not.
AFFECT AND INFORMATION PROCESSING
43
Postulate 6 When people’s primary reason for engaging in behavior is enjoyment, they use the affect they are experiencing as a basis for inferring this enjoyment. Therefore, they will continue the activity longer if they are experiencing positive affect than if they are not. In combination, Postulates 5 and 6 suggest that the affect persons experience will have different influences on their decision to continue performing an activity in which they are engaged, depending on whether their motivation for engaging in the activity is extrinsic or intrinsic. Two studies by Martin et al. (1993) demonstrated these differences empirically. In one study, happy or sad participants were asked to form an impression of someone on the basis of a series of behaviors. Each behavior was printed on a separate card of a deck that participants were given. Participants under performance-criterion conditions were told to continue turning the cards and reading the behaviors until they had enough information to form an impression. In contrast, participants under enjoyment-criterion conditions were told to continue as long as they were enjoying what they were doing. The first group of participants should implicitly ask themselves as they read the behaviors if they feel they have enough information. Therefore, they should be more likely to answer this question affirmatively, and to stop reading sooner, if they are happy than if they are sad. The second group of participants, however, should ask themselves if they are enjoying the task, again responding more affirmatively if they are happy than if they are sad. In this case, therefore, happy participants should continue reading behaviors longer than sad participants. Results confirmed these predictions. That is, happy participants read fewer behaviors than sad participants in the first condition, but more behaviors than sad participants in the second. A second study replicated these findings using a different task. In this case, participants were asked to generate a list of birds. Some participants were told to continue until they thought it was a good time to stop, whereas others were told to continue as long as they were enjoying themselves. Happy participants persevered less long on the task than unhappy participants when they were told to use a performance criterion, but longer than unhappy participants when they were encouraged to employ an enjoyment criterion. In Martin et al.’s studies, the criterion for stopping or continuing was explicitly imposed by the experimenter. When the criterion is more implicit, the relative emphasis on performance and enjoyment is likely to vary with the nature of the task and the context in which it is performed. In laboratory studies, participants are often given a particular task objective by the experi-
44
R. S. WYER, Jr., ET AL.
menter. Under these conditions, participants may spontaneously employ a performance criterion. That is, they may base their decision to persevere on whether they feel that their responses fulfill these externally imposed task requirements. This hypothesis was supported in a third condition of Martin et al.’s second experiment in which participants were simply told to stop generating birds when “they wanted to.” The results in this condition were very similar to those obtained when participants were given an explicit performance criterion; that is, happy participants stopped sooner than unhappy ones. These results confirm the speculation that participants in many experimental situations spontaneously focus their attention on whether they have fulfilled the implicit task objectives they are given. Nevertheless, when an experimental task is particularly interesting, or when achievement goals are deemphasized, an enjoyment criterion might be applied as well. Some evidence in support of the latter possibility was reported by Murray et al. (1990) in a study mentioned earlier in this article. Some participants were asked to list similarities between two popular television programs, whereas others were asked to list differences between the programs. The task was intrinsically interesting to college-age participants and, therefore, was likely to stimulate them to use an enjoyment criterion in deciding how much effort they should expend. To this extent, participants should use the affect they were experiencing as a basis for inferring their enjoyment. Consistent with this assumption, happy participants reported enjoying the task more than control participants did. Correspondingly, happy participants listed more similarities between the television programs when they were asked to focus on similarities, and listed more differences between the programs when they were focused on differences, than control participants did. Note that these results would be hard to explain on the basis on either the assumption that positive affect generally increases the accessibility of global concepts in memory or the assumption that positive affect reduces cognitive resources. It seems more reasonable that happy participants inferred from the affect they were experiencing that they were enjoying the task, and therefore devoted more effort to an identification of both similarities and differences in the stimuli they were judging than other participants did. Murray et al.’s finding that happy participants reported greater enjoyment of the task than other participants raises an additional consideration. That is, when several goals are available in a situation, the affect that persons happen to be experiencing could influence the goal they choose to pursue as well as their inference that they have or have not attained it (for a discussion of this possibility, see Schwarz & Clore, 1996). For example, happy individuals may often be more likely to adopt an enjoyment objective than unhappy individuals (Isen, 1987). If this is so, happy persons may
AFFECT AND INFORMATION PROCESSING
45
persist on a task because they infer that they are enjoying it, whereas unhappy persons, who adopt a performance criterion, may persist because they feel that they have not yet attained an adequate level of achievement. To this extent, both happy and unhappy persons might persist to an equal extent, albeit for different reasons. Despite this ambiguity, the influence of affect on behavior in a large number of situations both in and outside the laboratory can be conceptualized in terms of its mediating influence on the answers to questions that persons implicitly ask themselves about their behavior and its outcomes. Moreover, this conceptualization calls attention to several contingencies in the impact of affect that have not been investigated in the research performed to date. Several areas of research are worth reviewing from this perspective. A. CREATIVITY A creativity task that is often employed in the laboratory requires participants to generate instances of a concept. This task is similar to that constructed by Martin et al. (1993, Experiment 2) except that the concept involved has unclear boundaries, thus allowing participants to generate novel responses. Thus, for example, participants might be asked to generate uses of a brick or things they could use to sit on. In this situation, like that constructed by Martin et al., participants might assume they should stop when they have generated enough good responses to fulfill the demands of the task. However, the task itself might often be intrinsically enjoyable, as it requires a degree of imaginativeness. Thus, many participants might spontaneously apply an enjoyment criterion as well. Consistent with this speculation, Isen and Daubman (1984) found that happy participants generated more responses when performing such a task than control participants did. This suggests that an enjoyment criterion predominated. A somewhat different interpretation of these results, however, is suggested by the fact that happy participants also tend to generate more novel responses than other participants. When the criterion for an appropriate response is unclear, participants may inhibit reporting responses that they feel are poor exemplars of the concept they are considering. Specifically, they may subjectively generate a response (e.g., a potential use for a brick), ask themselves if they feel their response is a good one, and only report it if they answer this question affirmatively. Thus, they are more likely to write down their response on the list they are compiling for the experimenter if they are happy than if they are not. The tendency for happy persons to perform better then other persons on the Remote Associates Test (Isen,
46
R. S. WYER, Jr., E T AL.
Daubman, & Nowicki, 1987) could also be consistent with this interpretation. Finally, the interpretation might also explain the difference between the influence of affect on responses in Isen and Daubman’s (1984) study and its effects in control conditions of Martin et al.’s (1993) experiment. That is, the criterion for membership in the category that participants considered in Martin et al.’s experiment (birds) was clearly defined, and so the use of affect to evaluate the appropriateness of individual exemplars was less likely. Individual differences may exist in the relative emphasis that is placed on performance objectives and enjoyment objectives. Persons with high need for cognition (Cacioppo & Petty, 1982), for example, typically find intellectual activity to be intrinsically enjoyable. They may therefore be likely to adopt an enjoyment criterion, whereas persons with low need for cognition may be guided to a greater extent by performance criteria. If this is so, the experience of positive affect might increase the persistence of high need-for-cognition individuals on response-generation tasks, but might decrease the persistence of low need-for-cognition individuals on such tasks. In fact, data obtained by Martin et al. (1993) confirm this prediction. B. IMPRESSION FORMATION
In impression formation research (for a review, see Wyer & Carlston, 1994), participants typically receive information about several behaviors a person has performed with instructions to form an impression of the person. A large amount of information is typically presented (e.g., 25-30 items; see Wyer & Martin, 1986), and so it is difficult to form an impression that takes all of it into account. Consequently, participants in these experiments may often ask themselves at various points if the information they have received is a sufficient basis for a concept of what the individual is like. If they answer this self-generated question affirmatively, they may base their concept of the person on this information without taking into account the implications of the information they receive later. If this is so, and if happy participants are more likely than unhappy participants to respond affirmatively to such a question, they should use relatively less information to form their impressions. Moreover, the information they use should typically occur at the beginning of the sequence rather than at the end. Martin et al.’s (1993, Experiment 1) results confirm the assumption that happy participants with an impression formation objective use less information to form an impression than sad participants do. The prediction that happy persons are more likely than unhappy ones to base their impression on the first information they receive was confirmed in research by Isbell
AFFECT AND INFORMATION PROCESSING
47
et al. (1998) mentioned earlier. Participants who were feeling either happy or sad were asked to form an impression of someone who was described at the outset as either extraverted or introverted. This general description was followed by a series of behaviors, an equal number of which exemplified each of these traits. After receiving the information, participants judged the target’s extraversion and then recalled the behaviors they had received. We expected that the general description of the target that participants received at the outset would activate a concept of either an extravert or an introvert, and that participants would implicitly ask themselves if they felt that this concept was an appropriate basis for their impression. Happy participants, who should answer this question affirmatively, should be more inclined than unhappy participants to apply the concept to the target and to base the judgments they report later on this concept. This was in fact the case, as indicated earlier. That is, the initial target description had a substantial effect on happy participants’ judgments of the target’s extraversion but had very little impact on unhappy participants’ judgments. The fact that happy participants used the concept activated by the initial description of the target as a basis for their judgments does not mean that they failed to think about the behavioral information they received later. To the contrary, persons who apply this concept to the target should think about the behavioral information they receive with reference to it (Srull & Wyer, 1989). Moreover, they should think more extensively about behaviors that are inconsistent with the concept in order to reconcile their occurrence (Hastie, 1980; Srull & Wyer, 1989; Wyer & Srull, 1989). As implied by this reasoning, these participants recalled more inconsistent behaviors than consistent ones. Unhappy participants, who did not apply the concept activated by the initial trait description to the target, did not engage in this inconsistency resolution. Consequently, they recalled behaviors that were consistent and inconsistent with this description equally well. The phenomena identified by Isbell et al. are not restricted to laboratory situations in which affect is induced experimentally. In a study conducted over the telephone, participants’ self-reported feelings of happiness were used to infer their affective state. The impact of affect on judgments observed in this study, shown in Figure 3A, was identical to that observed in the first experiment. In additional conditions of this study, however, before engaging in the impression formation task participants were asked to report the possible reasons for their mood, thus calling their attention to the actual source of the feelings they were experiencing. In these conditions, participants appeared to reject the implications of their feelings when deciding whether the initial target description was a sufficient basis for their impressions. Consequently, as shown in Figure 3B, impact of participants’
48
R. S. WYER, Jr., ET AL.
A 10
Sales Rep
8 6 v)
.-8 4
’1
c
E *
2
s o
\
Librarian
-2
6
8
.c 4
5
2*
B 0
r:
-2
-4 -6
-8 I
I
I
Unhappy
Mood
Happy
Fig. 3. Effects of trait-descriptionof a target person on judgments by happy and unhappy persons for whom the source of their affective reactions (A) was not and (B) was made salient. (Adapted from Isbell, Clore, & Wyer, 1998.)
affective state on their use of the initial target description was exactly the opposite of that observed when the true source of their feelings was not salient to them. That is, the initial target description had a substantial effect on unhappy participants’ judgments, whereas its influence on happy participants’ judgments was negligible.
AFFECT AND INFORMATION PROCESSING
49
More general implications of our conceptualization should be noted. In many cases, the initial impressions that people form of someone are based largely on their feelings toward the person. In these cases, the affect they experience at the time they judge the person could potentially have a dual function. First, it may convey judgment-relevant information about the individual being judged (Schwarz & Clore, 1983). Second, it could be used as an indication of whether this affect is a sufficient basis for making the judgment. Thus, people who experience positive affect at the time they are asked to judge someone might not only infer that they feel favorably toward the target, but also that these feelings are a sufficient basis for the judgment they are asked to make. In contrast, people who experience negative affect may infer that they feel negatively toward the target but also may conclude that these feelings are not a sufficient basis for the judgment. This implies that the effects of positive affect on impression judgments may often be greater than the effects of negative effect. Our previous speculation, that people are generally less likely to correct for the biasing influence of positive affect than for the biasing influence of negative affect (cf. Schwarz & Clore, 1983), is consistent with this analysis.
C. STEREOTYPING Our interpretation of Isbell et al.’s findings also has implications for the effects of affect on the use of stereotypes. In a study by Bodenhausen (1993; Bodenhausen et al., 1994), for example, happy and sad participants were asked to judge the culpability of a defendant on the basis of information about the defendant’s background and the conditions surrounding the crime. The description of the crime was preceded by the target’s name and hometown, which either identified him as Hispanic or provided no clue as to his ethnicity. It seems reasonable to suppose that the defendant’s ethnicity spontaneously activated a stereotype, and that participants implicitly asked themselves if they felt this was an appropriate basis for their judgment, answering affirmatively if they were happy and negatively if they were sad. Consistent with this reasoning, the stereotype had more impact on happy participants’ judgments than on unhappy participants’. Note that if the stereotype had not been activated until after the detailed crime-related information was presented, participants would not be in a position to ask themselves at the outset whether the stereotype was a sufficient basis for judgment. If anything, they should ask themselves if they felt the crimerelated information was sufficient. To this latter extent, happy participants in this condition might be more influenced by the crime-related information,
50
R. S. WYER, Jr., ET AL.
and less influenced by the stereotype, than sad participants are. As far as we know, this possibility has not been examined.
D. COMMUNICATION AND PERSUASION People who receive a communication that advocates a certain position are likely to have two goals in mind. One is to decide if the position being advocated is valid. A second objective, which is often not incompatible with the first, is simply enjoyment. Individuals with the first objective are likely to form an initial impression of the likelihood that the position advocated is valid before evaluating carefully the communication’s contents. This impression could be based on the source of the message, on surface features of the communication itself (e.g., writing style), or on a previously formed attitude toward the position in question. When these a priori criteria do not exist, people may tentatively accept the position advocated as correct, and only later may attempt to discredit it. In this regard, Gilbert (1991) reported evidence that persons must tentatively accept an assertion as true in order to comprehend it, and that a later stage of processing is required to refute its validity. In the present context, this suggests that people are initially disposed to accept the arguments contained in a message as true, and that additional processing must be performed in order to refute them. According to the present conceptualization, persons who make an initial appraisal of a message ask themselves if this appraisal is an appropriate basis for accepting the position advocated, and are more inclined to answer this question affirmatively if they are happy than if they are sad. This has two implications. First, happy persons are more likely than unhappy persons to rely on the source of a message as a basis for accepting the position advocated. Second, happy persons are less inclined than sad persons to make a careful evaluation of the arguments contained in a message. Consequently, their acceptance of the position advocated is less likely to be influenced by the quality of these arguments. Support for the first hypothesis has been obtained by Roselli (1995), and the second hypothesis is consistent with results of several studies described earlier (Bless et al., 1993; Mackie & Worth, 1989). The conclusion that happy participants with a performance objective are less likely than unhappy participants to be influenced by the quality of arguments contained in a message might be conceptualized in terms of Petty and Cacioppo’s (1986) elaboration-likelihood model. That is, happy participants appear more inclined to engage in peripheral processing, whereas unhappy persons are more disposed to engage in central processing. However, the fact that happy participants’ attitudes do not depend on the
AFFECT AND INFORMATION PROCESSING
51
strength of arguments presented in the message does not imply that they ignore these arguments. Rather, they may evaluate them in terms of a position they have adopted on the basis of other criteria. Bless et al.’s (1990) study bears directly on this possibility. Happy and sad participants received a communication advocating a fee increase at their university, All participants may have spontaneously formed an initial impression of the policy’s desirability based on message-unrelated criteria that came to mind at the time (e.g., the money they would have to pay). Happy participants, however, may have decided that this impression was a sufficient basis for opposing the policy and, having adopted this attitude, thought more extensively about aspects of the message that called its validity into question. Therefore, as Bless et al. found, they counterargued the arguments contained in the message regardless of whether they were weak or strong, and the attitudes they reported did not depend on the cogency of these arguments. In contrast, unhappy participants may have considered their initial impression to be an insufficient basis for their attitude. Therefore, they evaluated the message content more dispassionately, elaborating the position implications of strong arguments but counterarguing weak ones, and basing their attitudes on the results of this cognitive activity. Bless et al.’s findings confirm this interpretation as well. Thus, both happy and unhappy participants engaged in similar amounts of cognitive activity, but the nature of this activity differed. To this extent, these results confirm implications of Isbell et al.’s (1998) study in a different domain. Two contingencies on this analysis are suggested by our conceptualization. First, the different cognitive activities performed by happy and unhappy persons should only occur if these persons perceive their feelings to have implications for the sufficiency of their initial impression as a basis for their attitude. If participants attribute their feelings to other, attitudeirrelevant factors, the difference in processing should theoretically not occur. Results reported by Sinclair, Mark, and Clore (1994) confirm this prediction. They found that the effect of argument strength on participant’s acceptance of a persuasive message was less on sunny days (when participants were happy) than on rainy days (when they were unhappy). However, calling participant’s attention to the weather, thereby making salient the true source of their feelings, eliminated this difference. Second, our analysis of Bless et a1.k (1990) study assumes that recipients’ objective at the time they read the message is to evaluate the validity of the position being advocated. In contrast, suppose persons have an enjoyment objective. Then, precisely the opposite influence of affect might occur. The aforementioned study by Wegener et al. (1995), suggesting that happy persons are more influenced by the quality of arguments in a message than other participants when they are led to believe that the message will be
52
R. S. WYER, Jr., ET AL.
enjoyable, confirms this implication of the performance-feedback conceptualization we propose. E. ACCURACY The conceptualization we have outlined could apply to judgments of physical stimuli as well as social ones. People who are called upon to judge a stimulus may spontaneously make a holistic appraisal of the stimulus at the time they are first exposed to it before they scrutinize its individual features. They may then ask themselves if this appraisal provides a sufficient basis for judgments. To this extent, happy individuals should be more likely than unhappy ones to base their judgments on this initial appraisal alone. One implication of this is that when accurate judgments are likely to require a careful analysis of the individual stimulus features happy participants are less likely to be accurate than unhappy ones. Evidence reported by Sinclair and Mark (1992), showing that happy participants were less accurate than sad participants in estimating the correlation between two variables on the basis of scatter plots, confirms this speculation.
F. BEHAVIOR IN NONLABORATORY SITUATIONS Many situations arise outside the laboratory in which persons’ behavioral decisions are likely to be mediated by their feelings about the behavior and its implications. Students who prepare for an exam must decide if they have studied long enough. A person who wants to ask someone for a date must decide if it is the right time to do so. Or, a person who wishes to buy a car must decide if he or she has enough information about it to make a decision. In such instances, people should stop studying earlier, ask persons for a date sooner, and make a purchasing decision more quickly, if they feel happy than if they do not. For similar reasons, happy persons may be more inclined to behave impulsively, or in ways that upon reflection they might consider undesirable. A series of studies by Forgas (1998) are interesting to consider in this context. In three studies, participants were asked to indicate how they would phrase a verbal request in a hypothetical social interaction. It seems reasonable to suppose that people who want another person to do sornething (e.g., to repay a loan) will spontaneously generate a verbal phrasing of the request that corresponds directly to the desired action (e.g., “Give me the money you owe me”) and may implicitly ask themselves if it is
AFFECT AND INFORMATION PROCESSING
53
appropriate to express the request in these terms. Only if they answer this question negatively may they alter the phrasing in a way that corresponds to normative standards of politeness. If this is so, the performance-feedback conceptualization we propose suggest that happy persons are more likely to express their requests impolitely than unhappy persons are. This, in fact, is what Forgas found. Although these studies were performed in the laboratory, they seem likely to generalize to other situations as well. However, the conclusion that happy persons are less likely to engage in extensive information processing in situations outside the laboratory is somewhat questionable. These differences theoretically exist only when individuals have an implicit performance goal and base their decisions to process information on whether they feel they have attained this goal. In nonlaboratory situations, information processing is often performed for enjoyment and not for the purpose of attaining an external objective. In these circumstances, the performance-feedback conceptualization suggested by Martin et al. (1993) and elaborated here implies that persons who experience positive affect are likely to persevere longer, and to process information more analytically, than persons who experience negative affect. For example, happy individuals may work harder on crossword puzzles, and may read news articles in greater depth, than sad individuals do. Moreover, they might make less premature decisions about other persons on the basis of the information they receive about them, in contrast to results observed in the laboratory (Isbell et al., 1998). The possibilities, which are not implied by other conceptualizations of the influence of affect on performance, should be kept in mind when drawing conclusions on the basis of experimental research in which performance objectives are implicitly emphasized. G. A FINAL COMMENT
It is important to bear in mind that predictions on the basis of the performance-feedback model we propose require a priori knowledge of the goal that persons are pursuing and the alternative processing strategies they are likely to employ. In many instances, these criteria and strategies are self-evident. In other cases, however, they are less clear. Moreover, as Schwarz and Clore (1996) point out, the affect that persons experience may have an impact on the goal that persons pursue as well as their strategy for attaining it. Nevertheless, the performance-feedback model potentially accounts for a wide range of phenomena and makes predictions that other conceptualizations cannot. Its implications are therefore worth further consideration.
54
R. S. WYER, Jr., ET AL.
VIII. Multiple Influences of Affect on Judgment A. THEORETICAL CONSIDERATIONS As we noted earlier, few attempts have been made to circumscribe either theoretically or empirically the conditions in which each of the various influences of affect on information processing is likely to predominate. The only previous conceptualization of the role of affect in cognitive processing that speaks directly to this matter was proposed by Forgas (1992, 1994, 1995). He identifies four different strategies that people could use to judge a target stimulus. In two of these strategies, affect comes into the picture, but in different ways. One, heuristic-processing strategy is used when individuals are unmotivated to think extensively about the judgment to be made, but a previously formed judgment of the target is unavailable. In this case, judges theoretically use their affective reactions as a heuristic basis for judgment without considering other judgment-relevant knowledge they may have. The second, substantive-processing strategy is employed when judges are motivated to make an accurate judgment and, to d o so, they must select and interpret new information about the target in relation to their prior knowledge. In this case, affect is postulated to influence judgments through its moderating impact on the concepts and knowledge that persons retrieve and use to interpret the new judgment-relevant information. The present conceptualization therefore differs from Forgas’s in two important respects. First, although we postulate that persons sometimes use affect as a heuristic basis for judgment (Postulate 4), we also assume that it can be the primary basis for certain types of judgments (Postulate 3). Second, in contrast to the implications of a substantive-processing strategy, we argue that affective reactions per se do not influence the retrieval of previously acquired knowledge. Support for our claims has already been reviewed and need not be reiterated. However, two sets of findings are worth reiterating briefly in this context. 1. Schwarz and Clore (1983) found that persons reported greater life satisfaction on sunny days than on rainy ones. This was presumably because they felt relatively happier on sunny days, and used these feelings as a basis for their judgments. Personally relevant judgments might be expected on a priori grounds to stimulate substantive processing. Thus, they should be based on the retrieval of mood-congruent knowledge. However, Parrott and Sabini (1990) found that persons were more likely to recall unhappy life experiences on sunny days than on rainy ones. Why persons recall mood-incongruent self-relevant knowledge under the same conditions in
AFFECT AND INFORMATION PROCESSING
55
which they make mood-congruent self-judgments is hard to explain in terms of Forgas’s theory. 2. Perhaps the most compelling evidence obtained in support of Forgas’s conceptualization was reported by Petty, Schumann, Richman, and Strathman (1983). They found that participants reported generally more favorable attitudes toward the position advocated in a message when they were happy than when they were not. However, the cognitive processes underlying these effects (as inferred from the content of participants’ thought listings) appeared to depend on participants’ motivation to engage in cognitive elaboration of the message content. Specifically, the effects of induced affect on unmotivated participants’ attitudes were independent of the thoughts they had about the message content, suggesting the use of a heuristic processing strategy. In contrast, the effects of induced affect on motivated participants’ attitudes were mediated by the favorableness of the thoughts they listed, suggesting the use of a substantive-processing strategy. Although these data are provocative, the causal direction of the effects obtained by Petty et al. is not completely clear. Perhaps both motivated and unmotivated participants based their attitudes in part on their feelings. However, motivated participants, unlike unmotivated ones, tended to think more extensively about aspects of the issues that were consistent with this attitude in order to confirm its validity. To this extent, these attitudes may have influenced the thoughts they listed subsequently rather than the reverse. In summary, we clearly share Forgas’s (1992, 1995) belief that affect can play multiple roles in information processing. However, we are skeptical of his specific assumptions and of his interpretation of the results he uses to support them.
B. EMPIRICAL SEPARATION O F THE INFLUENCE O F AFFECT According to the conceptualization we propose, affect influences judgments as a result of (a) its activation of a processing strategy that stimulates the access and use of categorical knowledge (Postulate 2) and (b) its use as information about one’s feelings about the object being judged (Postulates 3 and 4). The question is how to separate these and other possible influences of affect on judgments and to assess their relative contributions. One strategy for attaining this objective was used successfully by Adaval (1996) to examine the role of affect in consumer judgment. Based on N. Anderson’s (1971,1981) information integration theory. Adaval postulated that evaluations of a commercial product ( J ) are a weighted average of
56
R. S. WYER, Jr., ET AL
the evaluative implications of the pieces of information about its individual attributes, specifically,
where Zo is the scale value (in units of favorableness) of an initial impression of the product that is formed prior to a scrutiny of its individual attributes, siis the scale value of the ithpiece of information about the product, and wi is the subjective weight that is attached to this information. The affective reactions to a given piece of information should be reflected in the scale value of this information; the attention paid to the information should be reflected in its weight, and the use of affective reactions as information about one’s feelings toward the product as a whole (independently of one’s reactions to the individual pieces of product information) should be reflected in the initial impression. To evaluate these possibilities, participants who had been induced experimentally to feel happy or sad were asked to judge a series of products that varied in terms of both brand name (favorable or unfavorable) and (b) price (high or low). Additional judgments were made of products described by each attribute in isolation. Curve-fitting procedures were then used to estimate the weight and scale value associated with each level of price, each brand name, and the initial impression. This was done for each participant separately at each level of induced affect, in each of three product domains, and for each of four different judgments (estimates of the product’s quality, the sacrifice incurred by purchasing it, the value of the product, and the willingness to purchase it). Results were generally consistent with several hypotheses described in the previous sections concerning the impact of affect on judgments. However, some unexpected findings occurred as well, specifically: 1. If persons engage in substantive processing of the sort postulated by Forgas (1995), they should attend to pieces of information that are evaluatively congruent with their mood. In terms of Equation 1,this should be reflected in the assignment of a greater weight (wl) to mood-congruent pieces than to mood-incongruent ones. In fact, although Adaval’s data showed a slight tendency for this to be true, the effect was small and nonsignificant. Therefore, the mood congruence of the information was not a major contributor to its influence, consistent with Postulate 1 and contrary to Forgas’s (1995) model. 2. According to Postulate 2, positive affect elicits an information-processing strategy that results in greater attention to categorical bases for judgment. Consistent with this postulate, brand information had more influence
AFFECT AND INFORMATION PROCESSING
57
on judgments when participants were happy than when they were not, and this information received greater weight (wi) in the former condition. Unexpectedly, however, affect also influenced perceptions of the evaluative implications of brand information, as reflected in its scale value (xi). Specifically, happy participants interpreted favorable brand names as more favorable, but unfavorable brand names as less favorable, than did unhappy participants. This latter difference may be the result of processes similar to those observed by Herr (1986; Herr, Sherman, & Fazio, 1983) in a different content domain. That is, when people’s affective reaction to the brand information they receive is similar to the affect they are already experiencing, they cannot easily distinguish between the two sources of affect. Consequently, affect from both sources is likely to influence their inferences of how they feel about the referent they are judging. However, when recipients’ affective reactions to the brand information differ from the feelings they are experiencing for other reasons, the two sources of affect can be more clearly separated. In this case, recipients may use the irrelevant affect they are experiencing as a comparative standard in construing their feelings about the brand they are considering, and this may lead them to judge these feelings as having less favorable implications for their evaluations than they would otherwise. 3. Participants’ initial impressions of the product’s quality, and also of the sacrifice that would be incurred when purchasing it, both increased with the positive affect they were experiencing. The influence of affect on perceptions of sacrifice (a negative attribute) might seem surprising at first glance. However, quality and sacrifice are positively correlated in the real world; the higher the quality of a product, the more one must typically pay for it. It is therefore conceivable that participants used their affective reactions as a basis for their initial impressions of quality, and then based their estimates of sacrifice on these impressions. In summary, Adaval’s research provides evidence that affect can have several of the effects suggested by our conceptualization. The applicability of Anderson’s information integration model in isolating these effects should not be overgeneralized, however. The parameter estimation procedure involved in testing it requires the assumption that the absolute weight and scale value of each component piece of information is independent of the information accompanying it. This assumption is reasonable in the domain investigated by Adaval. That is, the different attributes of a given product may indeed be evaluated independently, and an overall judgment may be weighted function of these evaluations. As Wyer and Carlston (1979; see also Wyer & Srull, 1989) point out, however, this assumption is
58
R. S. WYER, Jr., ET AL.
unlikely to hold in many social judgment situations in which the pieces of information presented are responded to configurally. In forming person impressions, for example, the interpretation of a person’s traits is likely to depend on their context (Asch, 1946; Wyer & Carlston, 1979). In such cases, the procedure used by Adaval to test implications of the Anderson model is inapplicable. In fact, a study by Abele and Petzold (1994) indirectly supports this speculation. In their study, happy and unhappy participants formed impressions of a person described by sets of trait adjectives that varied systematically in their favorableness. In this study, only the scale value of participants’ initial impressions was systematically influenced by the affect that participants experienced. The failure for affect to influence the weights and scale values of the individual pieces of information presented could simply indicate that the information was not processed in a way that made these parameters psychologically meaningful.
IX. Task Analyses of the Impact of Affect on Information Processing The performance-feedback model we propose makes salient the fact that a clear understanding of how affective reactions influence goal-directed cognitive processing will ultimately require a specification of the requirements of the task to be performed. In the absence of such a conceptualization, any conclusions that are drawn are likely to have unclear generalizability. To exemplify the approach that we consider necessary, we analyze two areas of research that we have discussed more generally earlier in this article: communication and persuasion and self-evaluation processes. In doing so, we do not attempt to review all of the theoretical and empirical work performed in these areas that is relevant to the issues we raise. Rather, our purpose is simply to show that an analysis along the lines we propose can lead to new hypotheses concerning the way in which affective reactions come into play. A. COMMUNICATION AND PERSUASION Positive affect can often decrease the effect of argument quality on recipients’ acceptance of the position advocated. (For a review and analyses of the impact of affect on responses to a persuasive communication, see Schwarz et al., 1991). In evaluating the generalizability of this effect, however, it is important to distinguish between two types of persuasive mes-
AFFECT AND INFORMATION PROCESSING
59
sages. One type is designed to change a belief (that is, the likelihood that a particular statement is true, or that an event will occur). The other is intended to change an attitude, or one’s affective reactions to a person, object, event, or concept. The types of arguments that are typically contained in these communications and, therefore, the influence of affect on reactions to these arguments, are somewhat different. 1. Belief-Focused Communications Belief-focused communications bear on the likelihood of a past, present, or future state of affairs (e.g., the likelihood that abortion will be legalized, or that the CIA is a major importer of heroin into the United States). The arguments contained in these messages typically focus on antecedents, or conditions that imply the existence of the state of affairs in question. Suppose, for example, that a communication is intended to increase the belief that abortion will be legalized. Such a communication might assert that feminists are actively lobbying for the legalization of abortion and, if this is so, the event will occur. As noted by McGuire (1960; McGuire & McGuire, 1991; see also Wyer, 1974), these arguments often compose the premises of a syllogism with the form “A; if A, then B; B” (e.g., feminists are lobbying for abortion; if feminists are lobbying, abortion will be legalized; therefore, abortion will be legalized), the conclusion of which describes the state of affairs being asserted. In this type of communication, a strong argument is one that logically implies the conclusion and can easily be recognized as such. That is, it bears on premises of the form “A; if A, then B,” or their equivalents.‘O In addition, however, the argument bearing on premise A must be plausible; that is, it must increase the recipient’s belief that A is true. 2. Attitude-Focused Communications An attitude-focused persuasive message argues that a particular referent is more or less desirable. This referent could be a person, an event, a social policy, or a course of action (behavior). The cognitions on which an attitudefocused message focuses are captured by the model of attitude formation proposed by Fishbein (1967; see also Fishbein & Ajzen, 1975). According ‘I’ Premises of the form “A; if not B, then not A” (e.g., “feminists are actively lobbying for abortion rights; if abortion weren’t ultimately going to be legalized, they wouldn’t be lobbying”) would be logically equivalent to the premises “A; if A. then B”, and therefore would imply the conclusion B. However, premises of the form “A: if not A, then not B” (for example, “feminists are actively lobbying; if feminists were not actively lobbying, abortion would not be legalized”) would not logically imply this conclusion.
60
R. S. WYER, Jr., ET AL.
to this formulation, attitudes toward an event o r behavior are an additive function of the desirability of its possible consequences, each weighted by the belief that the consequence will occur. Given this conceptualization, a strong argument is one that either (a) increases beliefs that the target event will have a consequence the recipient already considers desirable, or (b) increases perceptions of the desirability of a consequence that the recipient already believes would occur. In most research on the effects of attitude-focused communications, manipulations of argument strength have not explicitly distinguished between these two effects. 3. The Role of Affect
The way in which affective reactions influence the impact of a persuasive communication is likely to depend substantially on whether the communication is belief-focused or attitude-focused. The conclusion of a belief-focused message is evaluated on the basis of descriptive criteria. Therefore, according to the conceptualization we have proposed, affect is unlikely to have a direct, informational influence on the acceptance of this conclusion. Nor is affect likely to influence the implications of knowledge that is retrieved from memory and brought to bear on evaluations of the conclusion. Rather, any impact that affect is likely to have should be mediated by its influence on either (a) participants’ assessment of the logical coherence of the arguments contained in the message, or (b) the amount of knowledge that is retrieved and used to assess the plausibility of these arguments. (For evidence of the impact on affect on syllogistic reasoning, see Fiedler, 1988). In contrast, an attitude-focused communication is effective because it leads the recipient to experience positive or negative affective reactions, and these reactions are used as a basis for evaluating the desirability of the behavior or state of affairs specified in the conclusion (Path 9, Figure 1). Suppose the strength of arguments contained in the communication is primarily a function of the desirability of the consequences it describes and, therefore, the affect that results from thinking about these consequences. Then, the influence of argument strength on the impact of the communication is likely to depend on whether recipients (a) think about the consequences described in the message in sufficient detail for the affect associated with them to be elicited, (b) combine the affect elicited by different consequences into a single affective response to the referent as a whole (Fishbein, 1967; Fishbein & Ajzen, 1975), and (c) distinguish this affect from the affect they are experiencing for other reasons. Note that if the latter distinction is not clearly made, message-irrelevant sources of affect are likely to have a direct informational influence on recipients’ perceptions of how they feel about the conclusion being advocated and,
AFFECT A N D INFORMATION PROCESSING
61
therefore, their acceptance of this conclusion. This direct, informational influence of affect should not occur in the case of a belief-focused communication. The influence of affect on responses to each type of communication may depend on the processing objective that recipients have at the time they read it. As implied by the performance-feedback model we propose, recipients with the objective of making an accurate assessment of the validity of a conclusion may process the information less extensively if they are experiencing positive affect and use these feelings as an indication that they have thought sufficiently about the message content to make a judgment. Thus, in the case of a belief-focused communication, happy individuals may tend less to identify logical inconsistencies in the arguments presented (Fiedler, 1988), and be less inclined to retrieve sufficient prior knowledge to assess their plausibility. Alternatively, if the message is attitude-focused, they may be less inclined to assess the likelihood and desirability of the individual consequences described in the message. In contrast, participants who read the message for enjoyment may be more likely to engage in these activities than unhappy recipients. On the other hand, persons should use the affect they are experiencing as a direct basis for judgments (Path 9, Figure 1) when they are considering an attitudefocused message, but not when they are considering a belief-focused message, and this might be true regardless of their processing objective.
B. SELF-EVALUATION PROCESS 1. General Considerations Judgments of oneself can pertain to many things, including specific traits, general self-worth, life satisfaction, and the likelihood of future life events. These judgments are based on different types of self-knowledge and require different amounts of cognitive computation. Self-knowledge can include episodic memories of past behaviors or experiences, self-descriptions that one has generated in the course of previous information processing, or general concepts of oneself and one’s behavior in certain types of situations. As Klein and Loftus 1993 have found, these types of self-knowledge may be stored in memory and retrieved independently of one another. Many self-relevant concepts and knowledge representations are likely to elicit affective reactions. However, whether these reactions are used as a basis for judgments depends in part on their relevance. Judgments of self-esteem or life satisfaction may be fundamentally estimates of one’s own feelings about oneself or about one’s life as a whole. Trait judgments, on the other hand, are more likely to be based on descriptive criteria.
62
R. S. WYER, Jr., ET AL.
In conceptualizing conditions in which different criteria for selfjudgments are applied, it is useful to apply a “race” model of procedural knowledge similar to that suggested by Logan (1988). Specifically, when a self-judgment is required, a number of different cognitive procedures for generating the judgment may be activated simultaneously. These processes could include (a) a search for a previously formed judgment-relevant trait description, (b) the identification of a judgment-related past experience and a construal of its implications for the judgment to be made, and (c) if relevant, the assessment of the implications of the affective reactions one is experiencing. Situational or individual difference factors that make one process easier to use than another will increase the likelihood that this process “wins.” An additional consideration, however, is implied by the feedback conceptualization of performance evaluation described earlier. That is, participants who have generated a response on the basis of the “winning” procedure may spontaneously ask themselves if they feel that this response is a sufficient basis for judgment. Participants who are experiencing positive affect may answer this question affirmatively, whereas those who are experiencing negative affect may not. Therefore, the “winning” criterion is more likely to be employed by happy persons than by unhappy ones. The influence of affect on four different types of self-judgments, judgments of global life satisfaction, self-worth, specific trait judgments, and estimates of personal risk, are worth considering from this perspective. 2. Judgments of Life Satisfaction A request to judge one’s life satisfaction might activate three different processes. First, respondents might search for a previously formed concept of their life as a whole that is directly applicable to the judgments. Second, they might recall one or more previous life events that are relevant to the judgment and base their response on the affect that is elicited by these memories. Finally, they might assess the affect that they happen to be experiencing at the time and use this as an indication of their feelings about life in general. These processes, which might occur simultaneously, are summarized in Figure 4. On a priori grounds, it is unclear which of these processes will predominate. For example, if persons in an experiment have a previously formed concept of their life as a whole, a request to estimate their life satisfaction might spontaneously activate this concept and so a judgment can quickly be computed on the basis of the affect it elicits. On the other hand, persons are infrequently called upon to make this type of judgment in daily life. Consequently, they may often not have a previously formed concept of their life as a whole stored in memory, and must compute a judgment at
63
AFFECT AND INFORMATION PROCESSING
A formed concept of life
I Judgment
life event
1
Transitory
B
I
I Retrieve previously formed trait concept
past experience
Construe implications for trait
Retrieve general
I
C
I
Judgment Compute implications of affective reaction
Fig. 4. Processing stages involved in making (A) global judgments of life satisfaction and self-esteem,(B) specific trait judgments, and (C) estimates of the likelihood of future life events.
64
R. S. WYER, Jr., ET AL.
the time they are asked to report it. In this case, it would normally take less time to compute a judgment on the basis of the transitory affect one is experiencing at the time of judgment than to do so on the basis of other, less easily accessible criteria, and so judgments are more likely to be based on the former computation. Evidence that transitory affect is used as a basis for life satisfaction judgments independently of any specific selfknowledge they have available (Schwarz & Clore, 1983; Strack et al., 1985) is consistent with this hypothesis. However, the performance-feedback model we have proposed raises a further consideration. People who experience positive affect for reasons that are objectively unrelated to their life situation might not only use these feelings as a basis for judgment but also consider this criterion to be sufficient. In contrast, persons who experience negative affect might be disposed to question the sufficiency of this criterion. Therefore, they might either search for other judgment-relevant information or, if no alternative criteria are easily available, might adjust for the bias that results from their use of their feelings as information. Schwarz and Clore’s (1983, Experiment 1) findings support this conjecture. In this study, happy persons appeared to use their feelings as a basis for judging their life satisfaction regardless of situational factors that had implications for their validity as a basis for judgment. In contrast, unhappy participants adjusted their judgments to compensate for the effects of these factors. Both happy and unhappy participants may have spontaneously considered the affect they were experiencing as a potential basis for judgment. However, whereas happy participants concluded that these feelings were a sufficient basis for judgment without thinking about situational factors that might influence them, unhappy participants considered that an assessment of their feelings was insufficient and took situational factors into account as well.
3. Judgments of Self-worth Estimates of self-worth, like estimates of life satisfaction, may typically be based on feelings about oneself. If this is so, these feelings are likely to be estimated by either (a) retrieving a previously formed concept of oneself with which affective reactions have become associated through learning; (b) recalling descriptive knowledge about oneself to which affective reactions have become conditioned and combining the implications of these reactions to arrive at an overall estimate of one’s feelings; and (c) assessing the implications of the transitory affect that one happens to be experiencing for reasons that may or may not actually be relevant. If a previously formed concept of “self” exists and is activated by the request to evaluate one’s
AFFECT AND 1NFORMATlON PROCESSING
65
self-worth, the affective reactions elicited by this concept might be used as the basis for judgment. If such a concept has not previously been formed, however, and if the affective reactions based on descriptive self-knowledge take time to compute, the transitory affect one happens to be experiencing for other reasons is likely to win the “race.” The aforementioned study by Levine et al. (1994) bears on this possibility. In this study, the descriptive implications of participants’ recalled experiences only affected their judgments of competence in the domain to which the experiences were relevant. However, participants’ estimates of their global self-esteem were influenced by the affect elicited by the life event they had recalled regardless of its descriptive implications. In terms of our conceptualization, this suggests that participants in this study did not have a previously formed concept of themselves easily accessible in memory. Therefore, they based their judgments on the affect they were experiencing at the time, regardless of whether or not the event that elicited the affect was self-relevant. 4. Trait Judgments
Judgments of the specific traits one possesses are more likely to be based on descriptive rather than affective criteria. The processes that underlie these judgments are summarized in Figure 4B. Specifically, persons who are asked if they have a specific trait may recall and report a previously formed concept of themselves with respect to this trait. Alternatively, they may infer the trait from the descriptive implications of trait-relevant selfknowledge that they have acquired (e.g., behaviors that exemplify the trait, descriptions of oneself by others, etc.). For affective reactions to have an informational influence on trait judgments, mediating computations must be performed in order to transform these reactions into descriptive terms. For example, people who experience negative affect might judge themselves as incompetent on the basis of reasoning: “I feel badly about myself; incompetence makes people feel badly; therefore, I am incompetent.” The syllogism implied by this reasoning (A; B implies A; therefore, B) is not logically valid. The computation of trait judgments based on this reasoning is likely to take longer than computations based on descriptive criteria. This is particularly true if a traitrelevant experience is easily accessible in memory at the time of judgment. Levine et al.3 (1994) results are consistent with this hypothesis. That is, the descriptive implications of participants’ recalled life experiences were used as a basis for judgments of attributes to which they were relevant. The affective implications of recalled experiences only influenced judgments of attributes in a domain to which these experiences were unrelated.
66
R. S . WYER, Jr.. ET AL.
Implications of the performance-feedback model are also of interest in this context. To the extent that trait judgments are likely to be computed more quickly on the basis of descriptive criteria than on the basis of affective criteria, persons who experience positive affect should be more likely to infer that these criteria are sufficient, whereas persons who experience negative affect are more likely to consider additional criteria. This implies in contrast to judgments of life satisfaction and self-worth, that trait inferences are more likely to be influenced by negative affect than by positive affect. 5. Likelihood Estimates
To predict the future, an individual could in principle conduct an exhaustive search of memory for similar events that have occurred previously, and use the incidence of such events to estimate the probability that they will occur. It seems more likely, however, that persons will employ an availability heuristic (Tversky & Kahneman, 1973) and base their judgment on the ease with which exemplars of the type of event being considered come to mind. Alternatively, as suggested by Johnson and Tversky (1983; see also Costans & Mathews, 1993; Gasper & Clore, 1998), persons might use the affective reactions elicited by a concept of the event as an indication of likelihood. That is, if people experience negative affective reactions to an event, they may infer that they are worried about it. Therefore, they may estimate its likelihood to be greater than persons who do not have these reactions." Correspondingly, persons may infer that the likelihood of a positive event is greater if they experience positive affect while thinking about it than if they do not. Figure 4C summarizes these possibilities. Assuming that each step of processing takes about the same amount of time, this figure implies that estimates are normally made more quickly on the basis of affective reactions than on the basis of an availability heuristic. It is again interesting to consider the implications of the performance-feedback model of affect we have proposed. That is, suppose persons are more likely to perceive the first basis for judgments they consider to be sufficient if they are experiencing positive affect than if they are experiencing negative affect. Then, they should be more likely to base their estimates of positive life events (which elicit positive affect) on their affective reactions to the events than to base their estimates of negative life events on this criterion. In principle, negative affect could be interpreted as an indication that one perceives the event being considered to be particularly aversive rather than as being likely to occur. However, Johnson and Tversky's (1983) results suggest that the latter inferences predominate.
AFFECT AND INFORMATION PROCESSING
67
6. Summary
It is clearly necessary to distinguish between the different types of selfjudgments that might be made, and the processes that underlie them, in order to understand how affective reactions are likely t o contribute to these judgments. Of particular importance is whether affective reactions provide the primary basis for the type of judgment being made or come into play only when more directly relevant criteria are difficult to apply. Without distinguishing between these two general types of judgments, general conclusions concerning the effect of affective reactions on self-judgments are difficult to draw.
X. Concluding Remarks
The conceptualization of affect proposed in this chapter distinguishes between the affective reactions that persons subjectively experience and the cognitions that elicit these reactions. An analysis of the effects of these reactions at different stages of processing reveals that they have their primary influence as a result of the information they provide about both (a) persons and objects to which the reactions are directed and (b) the likelihood that one’s processing objective has been satisfactorily attained. The performance-feedback conceptualization suggested by Martin et al.3 (1993) findings can account for many effects of affective reactions on task performance and on behavior more generally. There is nevertheless a need to evaluate more precisely the cognitive activities that are involved in pursuing a particular processing goal in order to predict the impact of affective reactions on goal-relevant judgments and behavior. Our analyses of the conditions in which affect is likely to influence responses to a persuasive message, and to have an impact on self-evaluations, exemplify this need. The theoretical framework we have proposed has numerous specific implications for the conditions in which affect is likely to influence information processing and the nature of this influence. However, the integrative utility of the formulation should not be overlooked. This integration is necessary for continued progress in understanding the interface of affect and cognition. The many specific theories that have emerged postulate processes that are typically not incompatible. However, these processes come into play at different times and in different conditions. The conceptualization we have outlined provides a framework within which these contingencies can potentially be identified and their relevance to a more general
68
R. S. WYER, Jr., ET AL.
understanding of the role of affect in information processing can be understood.
Acknowledgments The writing of this chapter and much of the research reported therein was supported by grants from the National Institute of Mental Health to the first author and from the National Science Foundation and the National Institute of Mental Health to the second. Many of the ideas underlying the conceptualization presented were stimulated by discussions with Karen Gasper, Carol Gohm, and other members of the Affect and Cognition Workshop, and the Social Cognition Group of the University of Illinois.
References Abele, A., & Petzold, P. (1994). How does mood operate in an impression formation task? An information integration approach. European Journal of Social Psychology, 24, 173-187. Abelson, R. P. (1983). Whatever became of consistency theory? Personality and Social Psychology Bulletin, 9, 37-54. Abelson, R. P., Kinder, D. R., Peters, M. D., & Fiske, S. T. (1982). Affective and semantic components in political person perception. Joiirnal of Personality and Social Psychology, 42, 619-630. Adaval, R. (1996). The role of affect (mood) states in the perception of quality, sacrifice and value. Unpublished Ph.D. dissertation, University of Illinois, Champaign-Urbana. Albarracin, D., & Wyer, R. S. (1997). Elaborative and nonelaborativeprocessing of a behaviorreLated persuasive communication: The effects of argument strength, affect,and situational distraction. Unpublished manuscript, University of Illinois at Urbana-Champaign. Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press. Anderson, N. H. (1971). Integration theory and attitude change. Psychological Review, 78, 171-206. Anderson, N. H. (1981). Foundations of information integration theory. San Diego, CA: Academic Press. Arkin, R. M., Gleason, .I. M., & Johnston, S. (1976). Effect of perceived choice, expected outcome, and observed outcome of an action on the causal attributions of actors. Journal of Experimental Social Psychology, 12, 151-158. Asch, S. E. (1946). Forming impressions of personality. Journal of Abnormal and Social Psychology, 41, 1230-1240. Bagby, R. M., Parker, J. D., & Taylor, G. J. (1994). The twenty-item Toronto Alexithymia Scale. Item selection and cross-validation of the factor structure. Journal of Psychosomatic Research, 38, 23-32. Bargh, J. A. (1984). Automatic and conscious processing of social information. In R. S. Wyer & T. K. Srull (Eds.), Handbook ofsocial cognition (Vol. 3, pp. 1-43). Hillsdale, NJ: Erlbaum. Bargh, J. A. (1994). The four horsemen of automaticity: Awareness, intention, efficiency and control in social cognition. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (2nd ed., Vol. 1, pp. 1-40). Hillsdale, NJ: Erlbaum.
AFFECT AND INFORMATION PROCESSING
69
Bargh, J. A. (1997). The automaticity of everyday life. In R. S. Wyer (Ed.), Advances in social cognition (Vol. 10, pp. 1-61). Mahwah, NJ: Erlbaum. Bargh, J. A., & Alvarez, J. (in press). The automatic evaluation of everything. In F. Strack & J. Laird (Eds.), Feeling and knowing. New York: Cambridge University Press. Bargh. J. A.. Bond, R. N., Lombardi, W. J.. & Tota, M. A. (1986). The additive nature of chronic and temporary sources of construct accessibility. Journal of Personality and Social Psychology, 50, 869-878. Bargh, J. A,, Chaiken, S., Raymond, P., & Hymes, C. (1996). The automatic evaluation effect. Unconditionally automatic attitude activation in a pronunciation task. Journal of Experimental Social Psychology, 32. 104- 120. Bargh. J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality d; Social Psychology, 71, 230-244. Bargh, J. A,, Lombardi, W., & Higgins, E. T. (1988). Automaticity of Person X Situation effects on impression formation: It’s just a matter of time. Journal of Personality and Social Psychology, 55, 599-605. Beck, A. I. (1982). Depression: Clinical, experimental and theoretical aspects. New York: Harper & Row. Birnbaum, M. H. (1981). Thinking and feeling: A skeptical review. American Psychologist, 36, 99-101. Birnbaum, M. H., & Mellers, B. A. (1979a). One-mediator model of exposure effects is still viable. Journal of Personality and Social Psychology, 37, 1090-1096. Birnbaum, M. H., & Mellers, B. A. (1979b). Stimulus recognition may mediate exposure effects. Journal of Personality and Social Psychology, 37, 391-394. Blaney, P. H. (1986). Affect and memory: A review. Psychological Bulletin, 99, 229-246. Bless, H., Bohner, G., Schwarz, H., & Strack, F. (1990). Mood and persuasion: A cognitive response analysis. Personality and Social Psychology Bulletin, 16, 33 1-345. Bless, H., Clore, G . L., Schwarz, N., Golisano, V., Rabe, C., & Woelke, M. (1996). Mood and the use of scripts: Does being in a happy mood really lead to mindlessness? Journal of Personality and Social Psychology, 71. 665-679. Bodenhausen, G. V. (1993). Emotions, arousal, and stereotypic judgments: A heuristic model of affect and stereotyping. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition and stereotyping: Interactive processes in group perception (pp. 13-37). San Diego: CA: Academic Press. Bodenhausen, G. V.. Sheppard, L. A., & Kramer, G. P. (1994). Negative affect and social judgment: The differential impact of anger and sadness. European Journal of Social Psychology, 24, 45-62. Bower, G. H. (1981). Emotional mood and memory. American Psychologist, 36, 129-148. Bower, G. H. (1991). Mood congruity of social judgments. In J. P. Forgas (Ed.), Emotion and social judgments (pp. 31-53). Oxford, UK: Pergamon. Bower, G. H., Gilligan, S. G., & Monteiro, K. P. (1981). Selectivity of learning caused by affective states. Journal of Experimental Psychology: General, 110, 451 -473. Bower, G. H., & Mayer, J. D. (1985). Failure to replicate mood-dependent retrieval. Bulletin of the Psychonomic Society, 23, 39-42. Byrne, D. (1971). The attraction paradigm. New York: Academic Press. Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and Social Psychology, 42, 116-131. Carver, C. S., & Scheier, M. F. (1981). Aiiention and self-regulation: A control theory approach to human behavior. New York: Springer-Verlag.
70
R. S. WYER, Jr., ET AL
Carver, C. S., & Scheier, M. F. (1990a). Principles of self-regulation action and emotion. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivarion and cognition (Vol. 2, pp. 3-52). New York: Guilford. Carver, C. S., & Scheier, M. F. (1990b). Origins and functions of positive and negative affect: A control-process view. Psychological Review, 97, 19-35. Chaiken, S., & Bargh, J. A. (1993). Occurrence versus moderation of the automatic attitude effect: Reply to Fazio. Journal of Personality and Social Psychology, 64, 159-165. Clore, G. L. (1992). Cognitive phenomenology: Feelings and the construction of judgment. In L. L. Martin & A. Tesser (Eds.), The construction of social judgment (pp. 133-164). Hillsdale, NJ: Erlbaum. Clore, G. L. (1994). Why emotions are never unconscious. In P. Ekman & R. J. Davidson (Eds.), The nature of emotion: Fundamental questions (pp. 285-290). New York: Oxford University Press. Clore, G. L., & Byrne, D. (1974). A reinforcement affect model of attraction. In T. L. Huston (Ed.), Foundations of interpersonal attraction (pp. 143-170). New York: Academic Press. Clore, G. L., & Ortony, A. (in press). Cognitive in emotion: Never, sometimes, or always? In L. Nadel & R. Lane (Eds.) The cognitive neuroscience of emotion. New York: Oxford University Press. Clore, G. L., Schwarz, N., & Conway, M. (1994). Affective causes and consequences of social information processing. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition, 2nd ed. (Vol. 1, pp. 323-417). Hillsdale, NJ: Erlbaum. Costans, J. I., & Mathews, A. M. (1993). Mood and the subjective risk of future events. Cognition and Emotion, 7, 545-560. Diener, E., Colvin, C. R., Pavot, W. G., & Allman, A. (1991). The psychic costs of intense positive affect. Journal of Personality and Social Psychology, 61, 492-503. Diener, E., Sandvik, E., & Pavot, W. (1991). Happiness is the frequency, not the intensity, of positive versus negative affect. In F. Strack, M. Argyle, & N. Schwarz (Eds.), Subjective well-being: An interdisciplinary perspective. Oxford, UK: Pergamon Press. Dienes, B. P. A. (1996). Mood as information: Affective cues for cognitive processing styles. Unpublished Ph.D. dissertation, University of Illinois at Urbana-Champaign. Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. New York: Harcourt Brace Jovanovich. Ellis, H. C., & Ashbrook, P. W. (1988). Resource allocation model of the effects of depressed mood states on memory. In K. Fiedler & J. Forgas (Eds.), Affect, cognition, and social behavior (pp. 25-43). Toronto: C. J. Hogrefe. Emmons, R. A. (1992). Abstract versus concrete goals: Personal striving level, physical illness, and psychological well-being. Journal of Personality & Social Psychology, 62, 292-300. Erber, R. (1991). Affective and semantic priming. Effects of mood on category accessibility and inference. Journal of Experimental Social Psychology, 27, 480-498. Fazio, R. H. (1986). Attitudes as object-evaluation associations: Determinants, consequences, and correlates of attitude accessibility. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences. Hillsdale, NJ: Erlbaum. Fazio, R. H. (1993). Variability in the likelihood of automatic attitude activation: Data reanalysis and commentary on Bargh, Chaiken, Govender, and Pratto (1992). Journal of Personality and Social Psychology, 64, 753-758. Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986). On the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229-238. Fiedler, K. (1988). Emotional mood, cognitive style, and behavior regulation. In K. Fiedler & J. Forgas (Eds.), Affect, cognition, and social behavior (pp. 100-119). Toronto: Hogrefe International.
AFFECT AND INFORMATION PROCESSING
71
Fishbein, M. (1963). An investigation of the relationship between beliefs about an object and the attitude toward that object. Human Relations, 16, 233-40. Fishbein, M. (1967). A behavior theory approach to the relations between beliefs about an object and the attitude toward the object. In M. Fishbein (Ed.), Readings in attitude theory and measurement (pp. 389-400). New York: Wiley. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior. Reading, MA: Addison-Wesley. Fiske, S. T., & Pavelchak, M. A. (1986). Category-based versus piecemeal-based affective responses: Developments in schema-triggered affect. In R. M. Sorrentino & T. T. Higgins (Eds.), The handbook of motivation and cognition: Foundations of social behavior (pp. 167-203). New York: Guilford Press. Fiske, S. T., & Taylor, S. E. (1991). Social cognition. New York: McGraw-Hill. Forgas, J. P. (1992). Affect in social judgments and decisions: A multiprocess model. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (Vol. 25, pp. 227-275). San Diego, CA: Academic Press. Forgas, J. P. (1994). The role of emotion in social judgments: An introductory review and an Affect Infusion Model (AIM). European Journal of Social Psychology, 24, 1-24. Forgas, J. P. (1995). Mood and judgment: The affect infusion model (AIM). Psychological Bulletin, I1 7, 39-66. Forgas, J. P. (1998). Asking nicely? The effects of mood on responding to more or less polite requests. Per.yonality & Social Psychology Bulletin, 24, 173-185. Forgas, J. P., & Bower, G. H. (1987). Mood effects on person perception judgments. Journal of Personality and Social Psychology. 53, 53-60. Forgas, J. P., Bower, G. H., & Krantz, S. (1984). The influence of mood on perceptions of social interactions. Journal of Experimental Social Psychology, 20, 497-513. Frijda, N. Ortony, A,, Sonnemans, J., & Clore, G. (1992). The complexity of intensity: Issues concerning the structure of emotion intensity. In M. S. Clark (Ed.), Review of Personality and Social Psychology (Vol. 11. pp. 60-89). Beverly Hills, CA: Sage. Garcia, M. (1996). The automatic evaluation of novel stimuli. Unpublished Ph.D. dissertation, New York University. Gasper, K., & Clore, G . L. (1998). The persistent use of negative affect by anxious individuals to estimate risk. Journal of Personality and Social Psychology, 74, 1350-1363. Gilbert, D. T. (1991). How mental systems believe. American Psychologist, 46, 107-119. Gohm, C. L., & Clore, G. L. (1999). Individual differences in emotional experience: A review of scales. Unpublished manuscript, University of Illinois at Urbana-Champaign, Gouaux, C. (1971). Induced affective states and interpersonal attraction. Journalof Personality and Social Psychology, 20, 37-43. Griffitt, W., & Veitch, R. (1971). Hot and crowded: Influences on population density and temperature on interpersonal affective behavior. Journal of Personality and Social Psychology, 17, 92-98. Hastie, R. (1980). Memory for behavioral information that confirms or contradicts a personality impression. In R. Hastie, T. Ostrom, E. Ebbensen, R. Wyer, D. Hamilton, & D. Carlston (Eds.), Person memory: The cognitive basis of socialperception. Hillsdale: N.J.: Lawrence Erlbaurn Associates. Herr, P. M. (1986). Consequences of priming: Judgment and behavior. Journal of Personality and Social Psychology, 51, 1106-1 115. Herr, P. M., Sherman, S. J., & Fazio, R. (1983). On the consequences of priming: Assimilation and contrast effects. Journal of Experimental Social Psychology, 19, 323-340. Higgins. E. T., Bargh, J. A., & Lombardi. W. (1985). The nature of priming effects on categorization. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 59-69.
72
R. S. WYER, Jr., ET AL.
Higgins, E. T., & King, G . (1981). Accessibility of social constructs: Information processing consequences of individual and contextual variability. In N. Cantor & J. F. Kihlstrom (Eds.), Personality, cognition and social interaction. Hillsdale, NJ: Erlbaum. Higgins, E. T., & Sorrentino, R. M. (Eds.) (1990). Handbook of motivation and behavior (Vol. 2). New York: Guilford. Hunt, R. R., & Einstein, G. 0. (1981). Relational and item-specific information in memory. Journal of Verbal Learning and Verbal Behavior, 20, 497-514. Ingram, R. E., Smith, T. W., & Brehm, S. S. (1983). Depression and information processing: Self-schemata and the encoding of self-referent information. Journal of Personality and Social Psychology, 45, 412-420. Isbell, L. M., Clore, G. L., & Wyer, R. S . (1998). Mood-mediated uses of stereotyping in impression formation. Unpublished manuscript, University of Illinois at UrbanaChampaign. Isbell, L. M., & Wyer, R. S. (1999). Correcting for mood-induced bias in impression formation: The roles of chronic and situational-induced motivation. Personality and Social Psychology Bulletin, 25, 237-249. Isen, A. M. (1984). Toward understanding the role of affect in cognition. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (Vol. 3, pp. 179-236). Hillsdale, NJ: Erlbaum. Isen, A. M. (1987). Positive affect, cognitive processes, and social behavior. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 20, pp. 203-253). New York: Academic Press. Isen, A. M., & Daubman, L. A. (1984). The influence of affect on categorization. Journal of Personality and Social Psychology, 47, 1206-1217. Isen, A. M., Daubman, L. A., & Nowicki, G. P. (1987). Positive affect facilities creative problem solving. Journal of Personality and Social Psychology, 52, 1122-1 131. Isen, A. M., Johnson, M., Mertz, E., & Robinson, G. (1985). Positive affect and the uniqueness of word association. Journal of Personality and Social Psychology, 48, 1413-1426. Isen, A. M., & Shalker, T. E. (1982). The effect of feeling state on evaluation of positive, neutral, and negative stimuli: When you “accentuate the positive,” do you “eliminate the negative”? Social Psychology Quarterly, 45, 58-63. Isen, A. M., Shalker, T. E., Clark, M. S., & Karp, L. (1978). Affect, accessibility of material in memory and behavior: A cognitive loop? Journal of Personality and Social Psychology, 36, 1-12. Johnson, E., & Tversky, A. (1983). Affect, generalization, and the perception of risk. Journal of Personality and Social Psychology, 45, 20-31. Kashima, Y.,& Kerekes, A. R. Z. (1994). A distributed memory model of averaging phenomena in person impression formation. Journal of Experimental Social Psychology, 30, 407-455. Kihlstrom, J. F. (1980). Posthypnotic amnesia for recently learned material Interactions with “episodic” and “semantic” memory. Cognitive Psychology, 12, 227-251. Klein, S. B., & Loftus, J. (1993). The mental representation of trait and autobiographical knowledge about the self. In T. K. Srull& R. S . Wyer (Eds.), Advances in social cognition (Vol. 5 , pp. 1-50). Hillsdale, NJ: Erlbaum. Klinger, E. (1975). Consequences to commitment to and disengagement from incentives. Psychological Bulletin, 82, 223-231. Kunst-Wilson, W. R., & Zajonc, R. B. (1980). Affective discrimination of stimuli that cannot be recognized. Science, 207, 557-558. Laird, J. D. (1974). Self-attribution of emotion: The effects of expressive behavior on the quality of emotional experience. Journal of Personality and Social Psychology, 29,475-486.
AFFECT AND INFORMATION PROCESSING
73
Larson, R. J., & Diener. E. (1987). Affect intensity as an individual difference characteristic: A review. Journal of Research in Personality, 21, 1-39. Lazarus, R. S. (1982). Thoughts on the relations between emotion and cognition. American Psychologist, 37, 1019- 1024. Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press. LeDoux, J. E. (1996). The enzotional brain. New York: Simon & Schuster. Levine, S. R., Wyer, R. S., & Schwarz, N. (1994). Are you what you feel? The affective and cognitive determinants of self-judgments. Experimental Journal of Social Psychology, 24, 63-77. Logan, G . D. (1988). Toward an instance theory of automatization. Psychological Review, 95* 492-527. Logan, G . D. (1997). The automaticity of academic life: Unconscious applications of an implicit theory. In R. S. Wyer (Ed.), Automaficify in everyday life: Advances in social cognition (Vol. 10, pp. 157-179). Mahwah, NJ: Erlbaum. Lott, B. E., & Lott, A. J. (1985). Learning theory in contemporary social psychology. In G . Lindzey & E. Aronson (Eds.), Handbook ofsociulpsychology (31d ed., Vol. 1, pp. 109-136) New York: Random House. Mackie, D. M., & Worth, L. T. (1989). Cognitive deficits and the mediation of positive affect in persuasion. Journal of Personality and Social Psychology, 57, 27-40. Martin, L. L., & Tesser, A. (1996). Some ruminative thoughts. In R. S . Wyer (Ed.), Ruminarive thoughts: Advances in social cognition (Vol. 9, pp. 1-48). Mahwah, NJ: Erlbaum. Martin, L. L., Ward, D. W., Achee, J. W., & Wyer, R. S. (1993). Mood as input: People have to interpret the motivational implications of their moods. Journal of Personality and Social Psychology, 64, 317-326. Mayer, J. D., Gayle, M., Meehan, M. E., & Haarman, A. (1990). Toward better specification of the mood-congruency effect in recall. Journal of Experimental Social Psychology, 26, 465-480. McGuire, W. J. (1960). A syllogistic analysis of cognitive relationships. In C. I. Hovland & M. J. Rosenberg (Eds.), Attitude organization and change (pp. 65-111). New Haven, CT: Yale University Press. McGuire. W. J., & McGuire. C. V. (1991). The content, structure and operation of thought systems. In R. S. Wyer, Jr., & T. K. Srull (Eds.). Advances in social cognition (Vol. 4, pp. 1-78). Hillsdale. NJ: Erlbaum. Miller, G . A,, Galanter, E., & Pribram, K. (1960). Plans and the structure of behavior. New York: Holt, Rinehart & Winston. Moreland, R. L., & Zajonc, R. B. (1979). Exposure effects may not depend on stimulus recognition. Journal of Personality and Social Psychology, 37, 1085-1089. Murphy, S . , & Zajonc, R. B. (1993). Affect, cognition, and awareness: Affective priming with optimal and suboptimal stimulus exposures. Journal of Personaliry and Social Psychology, 64, 723-739. Murray, N., Sujan, H., Hirt, E. R., & Syjan, M. (1990). The influence of mood on categorization: A cognitive flexibility interpretation. Journal of Personality and Social Psychology, 59, 411-425. Natale, M., & Hantas, M. (1982). Effect of temporary mood states on selective memory about the self. Journal of Personality and Social Psychology, 42, 927-934. Niedenthal, P. M.. Halberstadt. J. B., & Setterlund, M. B. (1997). Being happy and seeing “happy”: Emotional state mediates visual word recognition. Cognition and Emotion, I I , 403-432. Niedenthal, P. M., & Setterlund, M. B. (1994). Emotion congruence in perception. Personality and Social Psychology Bulletin, 20, 401-41 1.
74
R. S. WYER, Jr., ET AL.
Nolen-Hoeksema, S., & Morrow, J. (1993). The effects of rumination and distraction on naturally occurring depressed moods. Cognition and emotion, 7, 561-570. Ortony, A., Clore, G. L., & Collins, A. (1988). The cognitive structure of emotions. New York: Cambridge University Press. Ottati, V. C., & Isbell, L. M. (1996). Effects of mood during exposure to target information on subsequently reported judgments: As on-line model of assimilation and contrast. Journal of Personality and Social Psychology, 71, 39-53. Ottati, V. C., & Wyer, R. S. (1992). Affect and political judgment. In S. Iyengar & W. J. McGuire (Eds.), Explorations in political psychology (pp. 296-320). Durham, NC: Duke University Press. Parrott, G., & Sabini, J. (1990). Mood and memory under natural conditions: Evidence for mood and incongruent recall. Journal of Personality and Social Psychology, 59, 321-336. Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral routes to attitude change. New York: Springer-Verlag. Petty, R. E., Priester, J. R., & Wegener, D. T. (1994). Cognition processes in attitude change. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition, 2nd ed. (Vol. 2, pp. 69-142). Hillsdale, NJ: Erlbaum. Petty, R. E., Schumann, D. W., Richman, S. A., & Strathman, A. J. (1993). Positive mood and persuasion: Different roles for affect under high and low elaboration conditions. Journal of Personality and Social Psychology, 64, 5-20. Petty, R. E., & Wegener, D. T. (1993). Flexible correction processes in social judgment: Correcting for context-induced contrast. Journal of Experimental Social Psychology, 29, 137-165. Rholes, W. S., Riskind, J. H., & Lane, J. (1987). Emotional mood states and memory biases: The effects of cognitive priming and mood. Journal of Personality and Social Psychology, 52, 91-99. Riskind, J. H. (1989). The mediating mechanisms in mood and memory: A cognitive-priming formulation. Journal of Social Behavior and Personality, 4, 173-184. Riskind, J. H., Rholes, W. S., & Eggars, J. (1982). The Velten Mood Induction Procedure: Effects on mood and memory. Journal of Consulting and Clinical Psychology, 50,146-147. Roseman, I. J. (1984). Cognitive determinants of emotion: A structural theory. In P. Shaver (Ed.), Review of personality and social psychology: Emotions, relationships and health (Vol. 5, pp. 11-36). Beverly Hills, CA: Sage. Roseman, I. J. (1991). Appraisal determinants of discrete emotions. Cognition and Emotion, 5, 161-200. Rosenberg, M. J., & Hovland, C. I. (1960). Cognitive, affective and behavioral components of attitudes. In C. I. Hovland & M. J. Rosenberg (Eds.), Attitude organization and change: An analysis of consistency among attitude components (pp. 1-14). New Haven, CT:Yale University Press. Roseli, F. (1995). Positive affect and heuristic cues in persuasion. Poster presented at the American Psychological Society Convention, New York. Salovey, P., Mayer, J. D., Goldman, S. L., & Palfai, T. P. (1994). Emotional attention, clarity and repair: Exploring emotional intelligence using the trait meta-mood scale. In J. Pennebaker (Ed.), Emotion, disclosure and health. Washington, D C American Psychological Association. Salovey, P., & Singer, J. A. (1985, August). The effects of mood on the recall of childhood and recent autobiographic memories. Paper presented at the annual convention of the American Psychological Association, Los Angeles, CA. Schwarz, N. (1990). Feelings as information: Informational and motivational functions of affective states. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and cognition: Foundations of social behavior (Vol. 2, pp. 527-561). New York: Guilford.
AFFECT AND INFORMATION PROCESSING
75
Schwarz, N., & Bless, H. (1991). Happy and mindless, but sad and smart? The impact of affective states on the analytic reasoning. In J. Forgas (Ed.), Emotion and social judgment (pp. 55-71). Oxford: Pergamon. Schwarz, N., Bless, H., & Bohner, G. (1991). Mood and persuasion. Affective states influence the processing of persuasive communications. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 24, pp. 161-199). San Diego, CA: Academic Press. Schwarz, N.. & Clore, G . L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45, 513-523. Schwarz, N., & Clore, G. L. (1988). How do I feel about it? Informative functions of affective states. In K. Fiedler & J. Forgas (Eds.), Affecf, cognition, and social behavior (pp. 44-62). Toronto: Hofgrefe International. Schwarz, N., & Clore, G. L. (1996). Feelings and phenomenal experiences. In E. T. Higgins & A. Kruglanski (Eds.), Socialpsychologyc A handbook of basicprinciples. New York: Guilford. Schwarz. N., Strack. F.. Kommer, D., & Wagner, D. (1987). Soccer, rooms and the quality of your life: Mood effects on judgments of satisfaction with life in general and with specific life-domains. European Journal of Social Psychology, 17, 69-79. Seidlitz, L., Wyer. R. S., & Diener, E. (1997). Cognitive correlates of subjective well-being: The processing of valued life events by happy and unhappy persons. Journal of Research in Personality, 31, 240-256. Shank, R. C.. & Abelson, R. (1977). Scripts, plans, goals, and understanding. Hillsdale, NJ: Erlbaum. Simon, H. A. (1967). Motivational and emotional controls of cognition. Psychological Review, 74, 29-39. Sinclair, R. C., & Mark, M. M. (1992). The influence of mood state on judgment and action. Effects on persuasion, categorization, social justice, person perception, and judgmental accuracy. In A. Tesser & L. L. Martin (Eds.), The construction of social judgn2ent.s (pp. 165-194). Hillsdale, NJ: Erlbaum. Sinclair, R. C., Mark, M. M., & Clore, G. L. (1994). Mood-related persuasion depends on (mis)attributions. Social Cognition, 12, 309-326. Singer, J. A,. & Salovey, P. (1988). Mood and memory: Evaluating the network theory of affect. Clinical Psychology Review, 8, 21 1-251. Smith, C. A,, & Ellsworth, P. C. (1985). Patterns of cognitive appraisal. Journal of Personality and Social Psychology, 48, 813-838. Smith, E. R. (1984). Model of social inference processes. Psychological Review, 91, 392-413. Smith, E. R. (1990). Content and process specificity in the effects of prior experiences. In T. K. Srull & R. S. Wyer (Eds.), Advances in social cognition (Vol. 3, pp. 1-59). Hillsdale, NJ: Erlbaum. Smith, E. R. (1994). Procedural knowledge and processing strategies in social cognition. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (2nd ed., Vol. 1, pp. 99-151). Hillsdale, NJ: Erlbaum. Smith, E. R. (1996). What do connectionism and social psychology offer each other? Journal of Personality and Social Psychology, 70, 893-912. Sorrentino, R. M., & Higgins, E. T. (Eds.). (1986). Handbook of niotivafion and cognition: Foundations of social behavior. New York: Guilford. Srull, T. K., & Wyer. R. S. (1986). The role of chronic and temporary goals in social information processing. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 503-549). New York: Guilford Press. Srull, T. K., Rr Wyer, R. S. (1989). Person memory and judgment. Psychological Review, 96, 58-63.
76
R. S. WYER, Jr., ET AL.
Staats, A. W., & Staats, C. K. (1958). Attitudes established by classical conditioning. Journal of Abnormal and Social Psychology, 57, 37-40. Strack, F., Schwarz, N., & Gschneidinger, E. (1985). Happiness and reminiscing: The role of time perspective, affect, and mode of thinking. Journal of Personality and Social Psychology, 49, 1460-1469. Strack, F., Martin, L. L., & Stepper, S. (1988). Inhibiting and facilitating conditions of the human smile: A nonobtrusive test of the facial feedback hypothesis. Journal ofPersonality and Social Psychology, 54, 768-777. Thurstone, L. L. (1959). The measurement of values. Chicago: University of Chicago Press. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5, 207-232. Vallacher, R., & Nowak, A. (1994). Dynamical systems in social psychology. San Diego, CA: Academic Press. Velten, E. (1968). A laboratory task for the induction of mood states. Behavior Research and Therapy, 6, 473-482. Wegener, D. T., & Petty, R. E. (1995). Flexible correction processes in social judgments. The role of naive theories in correction for perceived bias. Journal of Personality and Social Psychology, 68, 36-51. Wegener, D. T., & Petty, R. E. (1997). The flexible correction model: The role of naive theories of bias in correction. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 29, pp. 142-208). San Diego, Academic Press. Wegener, D. T., Petty, R. E., & Smith, S. M. (1995). Positive mood can increase or decrease message scrutiny: The hedonic contingency view of mood and message processing. Journal of Personality and Social Psychology, 69, 5-15. Winkielman, P., Zajonc, R. B., & Schwarz, N. (1997). Subliminal affective priming resists attributional interventions. Cognition and Emotion, 11, 433-465. Worth, L. T., & Mackie, D. M. (1987). Cognitive mediation of positive mood in persuasion. Social Cognition, 5, 76-94. Wyer, R. S. (1974). Cognitive organization and change: An information-processing approach. Hillsdale, NJ: Erlbaum. Wyer, R. S. (1996). Ruminative thoughts. New Jersey: Lawrence Erlbaum Associates. Wyer, R. S., & Budesheim, T. L. (1987). Person memory and judgments. The impact of information that one is told to disregard. Journal of Personality and Social Psychology, 53, 14-29. Wyer, R. S., & Carlston, D. E. (1979). Social cognition inference and attribution. Hillsdale, NJ: Erlbaum. Wyer, R. S., & Carlston, D. E. (1994). The cognitive representation of persons and events. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition (2nd ed., Vol. 1, pp. 41-98). Hillsdale, NJ: Erlbaum. Wyer, R. S., & Martin, L. L. (1986). Person memory: The role of traits, group stereotypes, and specific behaviors in the cognitive representation of persons. Journal of Personality and Social Psychology, 50, 661-675, Wyer, R. S., & Radvansky, G. A. (in press). The comprehension and validation of social information: The role of situation models. Psychological Review. Wyer, R. S., & Srull, T. K. (1981). Category accessibility: Some theoretical and empirical issues concerning the processing of social stimulus information. In E. T. Higgins, C. P. Herman, & M. P. Zanna (Eds.), Social cognition: The Ontario Symposium (Vol. 1, pp. 161-198). Hillsdale, NJ: Erlbaum. Wyer, R. S., & Srull, T. K. (1984). Handbook of social cognition. Hillsdale, NJ: Erlbaum.
AFFECT AND INFORMATION PROCESSING
77
Wyer, R. S., & Srull, T. K. (1989). Memory and cognition in its social context. Hillsdale. NJ: Erlbaum. Wyer, R. S., & Srull, T. K. (1994). Handbook of social cognition (2nd ed.). Hillsdale. NJ: Erlbaum. Wyer, R. S., & Srull, T. K. (1986). Human cognition in its social context. Psychological Review, Y3, 322-359. Zajonc, R. B. (1980). Feeling and thinking. Preferences need no inferences. American Psychologist, 35, 151-175. Zajonc, R. B., & Markus, H. (1984). Affect and cognition: The hard interface. In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotion, cognition and behavior (pp. 73-102). Cambridge, UK: Cambridge University Press. Zajonc, R. B., Murray, S . T., & Inglehart, M. (1989). Feeling and facial efference: Implications of the vascular theory of emotion. Psychological Review, 96, 395-416. Zanna. M., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In d. Bar-Tal & A. Kruglanski (Eds.), The social psychology of knowledge (pp. 315-334). Cambridge, UK: Cambridge University Press.
This Page Intentionally Left Blank
LINGUISTIC INTERGROUP BIAS: STEREOTYPE PERPETUATION THROUGH LANGUAGE
Anne Maass
Prejudice and stereotyping have always been a prominent topic in social psychology, but until recently very little attention has been paid to the role of language in the transmission and maintenance of stereotypes (for exceptions see Giles & Johnson, 1987; Graumann & Wintermantel, 1989; Hamilton, Gibbons, Stroessner, & Sherman, 1992; Rusher, 1998; Van Dijk, 1984, 1987, 1988). This is surprising if one considers that language is undoubtedly the predominant means by which stereotypes are communicated through interpersonal discourse, by which they are transmitted from generation to generation, and by which the press and other mass media create social representations of social groups. Language also constitutes the predominant means by which sociologists and social-psychologists tend to measure stereotypes. Indeed, some authors have even taken a rather radical position by denying the mere existence of a-linguistic stereotypes (Mininni, 1982). Hence, if we want to understand how stereotypes function and how they become shared knowledge in a given society, we need to understand not only the cognitive and motivational processes by which they are driven but also the way in which they are transmitted. Over the past decade, we have investigated one aspect of language, namely language abstraction, that may play a subtle but potentially important role in stereotype transmission and maintenance. It is the goal of this chapter to describe this research program as it developed over the past years, to discuss its implications, and to propose extensions of the model to related areas. I. The Linguistic Intergroup Bias Model
The linguistic intergroup bias (LIB) model (Maass, Salvi, Arcuri, & Semin, 1989) describes a systematic bias in language use which may, in a ADVANCES IN EXPERIMENTAL SOCIAL PSYCHOLOGY. VOL II
19
Copyright 0 1999 by Academic Prev All rights of reproduction in any form reserved M)65-2601/99 $30(X)
T
subtle way, contribute to the perpetuation of stereotypes. The main prediction states that a positive behavior displayed by an in-group member will be described in relatively abstract terms, whereas the same behavior shown by an out-group member will be described in relatively concrete terms. For example, the in-group member offering help to a needy person may be described as “helpful” or “altruistic,” whereas an out-group member engaging in exactly the same behavior may be described as “helping.” On the contrary, negative behaviors will be described in concrete terms when displayed by an in-group member, but in abstract terms when displayed by an out-group member. In the case of aggressive behaviors, for instance, the in-group member may be described as “hurting somebody,” whereas the out-group member may be described as “aggressive.” The concept of language abstraction has been adopted from Semin and Fiedler’s (1988, 1991, 1992) linguistic category model which, in its original version, distinguishes four levels of abstraction in interpersonal terms. The most concrete terms are descriptive action verbs (DAV) such as “ A hits B” that provide an objective description of a specific, observable event. Interpretive action verbs (IAV) are slightly more abstract as they describe a larger class of behaviors (“A hurts B ” ) , although they maintain a clear reference to a specific behavior in a specific situation. The third level is represented by state verbs such as “A hates B,” which describe enduring psychological states that generalize beyond specific situations and behaviors but refer to a specific object (person B). Finally, the most abstract terms are adjectives describing a general disposition that apply across situations, behaviors, and objects (“A is aggressive” for extensions and applications of the linguistic category model see Semin & Fiedler, 1992). Importantly, interpretive implications vary systematically as one moves from the concrete to the abstract pole of the continuum. The more abstract the statement, the greater the amount of information provided about the protagonist, the greater the implicit temporal and cross-situational stability, and the more difficult it becomes to verify the veridicality of the statement and to imagine disconfirming instances (Semin & Fiedler, 1988). Interestingly, the assumptions of the linguistic category model seem to apply across a variety of languages including English, German, French, Italian, Dutch, Spanish, and Greek. Based on the linguistic category model, the LIB model therefore predicts that the differential language use for in-group and out-group behaviors has a number of important implications for the maintenance of stereotypic beliefs. In particular, describing an observed behavior in abstract terms such as “A hates B” or “ A is aggressive” implies that the act reflects some stable, underlying characteristic or behavior tendency of the actor. The use of abstract terms suggests inferences from the specific act to the actor’s
LINGUISTIC INTERGROUP BIAS
81
more general characteristics or states. Concrete descriptions ( A hit B ) , to the contrary, suggest that the behavioral episode is an isolated event that is not necessarily linked to the actor’s enduring characteristics or psychological states. Not surprisingly then, a hypothetical listener should form a very different impression of the same act and develop very different future expectations depending on the level of abstraction used by the speaker to describe the event. Thus, the LIB model makes two interrelated predictions that are schematically represented in Figure 1: On one hand, it predicts that desirable ingroup and undesirable out-group behaviors will be described in more abstract terms than undesirable in-group and desirable out-group behaviors. On the other hand, it predicts that this bias in language use has an impact on subsequent mental processes, including inferences and expectations such that the abstract statements used to describe desirable in-group and undesirable out-group behaviors induce the idea that the act represents an enduring tendency of the actor and has a high likelihood of being repeated in the future. It becomes easily evident how such differential language use may bolster existing stereotypes. In particular, one can imagine a self-perpetuating cycle in which stereotype-congruent negative behaviors of the out-group are described in abstract terms that reinforce the prior expectation. In contrast, unexpected positive behaviors of the out-group tend to be described in concrete terms, implying that they are isolated acts; such a concrete description of a desirable out-group behavior therefore leaves the generally negative conception of the out-group intact and prevents stereotypes from being easily invalidated by contradicting evidence. In the remaining part of this chapter, I will (a) describe the research paradigm most frequently employed in LIB studies, (b) review empirical
Fig. I. Main predictions of the Linguistic Intergroup Bias model.
82
ANNE MAASS
findings testing the main predictions, (c) address questions of external validity, (d) compare hypotheses about the underlying mechanisms of the LIB, (e) speculate about extensions of the model beyond intergroup relations, and (f) discuss a number of open problems that are potentially of interest. 11. Research Paradigm A yearly horse-race competition in the town of Ferrara in Northern Italy served as setting for the first test of the linguistic intergroup bias model which provided also the experimental paradigm for the majority of subsequent studies (Maass et al., 1989). Participants from competing teams (contrade) were approached and asked to view a series of cartoon scenes in which the protagonists (either belonging to their own team or to an opposing team) engaged in socially desirable or undesirable behaviors. An example for a negative episode is provided in Figure 2, in which a member of one team secretly drugs a horse of the competing team-an act not uncommon during the night before the race. An example for a positive episode is a scene in which a member of one team interrupts the race in order to help an injured member of the opposing team. Cartoon scenes such as these have the advantage that they are practically language-free and are therefore unlikely to bias the respondent’s verbal reactions.
Fig. 2. Example of cartoons used as stimulus material in the Linguistic Intergroup Bias paradigm.
83
LINGUISTIC INTERGROUP BIAS
4-
Desirable
3.5 --
0
Undesirable
3 .-
-2 --
2.5
1.5 .4
In-group
Out-group
Fig. 3. Mean abstraction as a function of valence and category membership in the Palio study (From Maass et al.. 1989. Copyright 0 1989 by the American Psychological Association. Adapted by permission of the publisher.)
The respondent’s reactions to each scene are measured in one of two ways: In the cued sentence completion procedure, subjects provide a brief written description of the act which is subsequently scored for language abstraction by independent raters.’ The subject is usually cued by “This person . . .” in order to assure that the protagonist appears in the subject position of the phrase. Alternatively, a multiple choice procedure is employed in which subjects are provided with four alternative descriptions of each scene corresponding to the linguistic category model. Response examples for the horse-drugging scene are: A member of contrada Sun Giacomo (a) drugs the horse of contrada San Giorgio, (b) damages the contrada Sun Giorgio, (c) detests the contrada Sun Giorgio, ( d ) is un,fair. The subjects’ task is simply to chose the response alternative that they feel describes the episode best.* The results of the Palio studies are presented in Figure 3. Regardless of the methodology employed, participants used relatively abstract terms when describing desirable in-group and undesirable out-group episode, but shifted their language somewhat towards the concrete end of the continuum when describing positive out-group and negative in-group behaviors.
’
The interrater reliability tends to be rather high in these studies (generally between .8S and .95). Both procedures require pretesting to assure that episodes are perceived as clearly positive or negative in valence regardless of group membership of protagonist. In addition, the multiplechoice procedure requires extensive pilot testing to assure that all response alternatives are pertinent to the scene and that no single response alternative attracts the majority of responses; otherwise there will not be sufficient variance in responses.
84
ANNE MAASS
111. Generalization across Intergroup Settings, Populations, and
Linguistic Groups Since the original study, the LIB has been tested in over 30 experiments conducted in a wide range of linguistic and cultural contexts. Language biases consistent with the LIB model have been obtained in a wide variety of intergroup settings, including political party affiliation (Karpinski & von Hippel, 1996; Rubini & Semin, 1994) competing sport teams (Franc0 & Maass, 1996b), rival towns (Lazzarato, 1989, Exp. 4), schools and nations (Arcuri, Maass, & Portelli, 1993), age groups (Ng & Chan, 1996), blackwhite race relations in the United States (Von Hippel, Vargas, & Sekaquaptewa, 1997), Southern versus Northern Italians (Maass, Ceccarelli, & Rudin, 1996; Maass, Milesi, Zabbini, & Stahlberg, 1995), competing interest groups such as hunters versus environmentalists (Maass et al., 1996) and, in some cases, among sexes (Fiedler, Semin, & Finkenauer, 1993; Guerin, 1994; but see Ng & Chan, 1996). A comparison across studies suggests that the bias is particularly pronounced where hostility between groups is high, such as between hunters and environmentalists (Maass et al., 1996-see Figure 4). It is important to point out that the above studies produced convergent results despite the fact that they focused on very different intergroup settings and were conducted in diverse cultural and linguistic contexts, including Italian, German, Dutch, American, and Chinese populations.
Desirable
0 Undesirable
In-group
Out-group
Fig. 4. Mean abstraction as a function of valence and category membership: hunter versus environmentalist (From Maass et al., 1996. Copyright 8 1996 by the American Psychological Association. Adapted by permission of the publisher.)
LINGUISTIC INTERGROUP BIAS
85
A. FAILURES OF REPLICATION At the same time, there are a few (for the most part unpublished) studies that failed to replicate the LIB and that may help define the boundary conditions of the effect. These studies fall into one of three groups: (a) those using very “benign” intergroup settings in which there are only minor conflicts between groups (such as students of different faculties, Lazzarato, 1989, or males vs. females, Ng & Chan, 1996), (b) those using the minimal group paradigm (see Harris, 1993) and (c) those investigating the LIB within the context of black-white race relations in Europe (see Beaton & Brown, 1996;Loda, 1995-96; Mikula & Graf, 1994; Palmer, 1997). Failures of replication in “benign” and minimal group settings suggest that the LIB requires some degree of intergroup conflict or prejudice to occur. Thus, the LIB does not seem to emerge in intergroup settings that lack either intergroup competition or clear stereotypic expectancies about the groups involved (I will come back to this issue in section VI). It is less clear why the LIB did not emerge in studies investigating black-white relations in Europe. European subjects (including British, Austrians, and Italians) do not seem to manifest the LIB toward blacks, whereas their American counterparts do show it (von Hippel, Sekaquaptewa, & Vargas, 1997). This may either be attributable to methodological or sampling differences or else it may reflect differences in race relations between the geographical areas, at least in the student population investigated in these studies.
IV. Generalization across Measurement Turning to the measurement issue, it is important to understand whether the LIB is limited to the original measurement procedures or whether it emerges also when different research tools are employed. As mentioned earlier, we had originally used two measurement procedures: (a) the cued sentence completion procedure that requires subjects to freely describe the scene while holding the sentence subject constant (e.g., Maass et al., 1989, study 2), and (b) the multiple choice procedure that requires subjects to select one of four or five predisposed response alternatives describing the same, visually presented cartoon scene (e.g., Maass et al., 1989, study 1). A more sensitive version of this latter procedure has been proposed by von Hippel and collaborators (Karpinski & von Hippel, 1996; von Hippel, Vargas, & Sekaquaptewa, 1997) in which the subject, rather than selecting
86
ANNE MAASS
a single description, rates the appropriateness of each response alternative on a 10-point scale. During the past years, different additional methods have been probed that have in general produced very similar results. In particular, different authors have elicited free verbalizations about in-group and out-group, thus allowing subjects to construct their own descriptions without reference to a specific context (Fiedler, Semin, & Finkenauer, 1993; Maass, Montalcini, & Biciotti, 1998; Ng & Chan, 1996). This method is profoundly different from the original procedure, as it requires subjects to provide summary statements about a group, which tend to be more abstract (and consequently of reduced range) compared to descriptions of specific acts. Despite this fact, negative out-group and positive in-group characteristics still tended to be described more abstractly than the remaining combinations, suggesting that the LIB model is not confined to the description of specific scenes but can be extended to generalized stereotypical beliefs about in-group and out-group. Yet another indirect way to assess the LIB is to analyze derivations of trait adjectives. Semin and Fiedler (1991) have suggested that adjectives may differ in abstractness depending on their derivation. Some adjectives, such as “pushy” are derived from descriptive action verbs (DAVs); others such as “aggressive” or “helpful” are derived from interpretive action verbs (IAVs); others, such as “hateful” and “adorable” from state verbs (SVs), whereas some are not derived (e.g., “introverted”). Applying this idea to the intergroup context, Franco and Maass (1996b) asked subjects to generate trait adjectives describing in-group and out-group that were subsequently rated for derivation. In some languages, such as Italian, trait derivation can be established in a rather objective fashion based on precise derivation rules, such as type of suffix (see Scalise, 1995, for classifying derivation of Italian verbs, nouns, and adjectives). Using such rules, adjectives were classified into those derived from descriptive or interpretative action verbs, from state verbs, and those without verb derivation. Franco and Maass (1996b) found that positive in-group and negative out-group traits tended to have more abstract derivations than negative in-group and positive out-group traits, suggesting that the linguistic intergroup bias is not limited to verbs but may also be encountered in trait derivation^.^ The striking consistency of results across the four measurement procedures (multiple choice, cued sentence completion, free verbalization, and Note, however, that Semin and Fiedler (1991) have argued that IAV-derived adjectives are more abstract than SV-derived adjectives, thus proposing an abstractness hierarchy for adjectives that differs from their original verb category model. In contrast, our data were analyzed according to the original model assuming a linear increase in abstractness from DAV, IAV, SV up to nonderived adjectives.
LINGUISTIC INTERGROUP BIAS
87
adjective derivation) suggests that the LIB does not depend on a specific methodological tool but can be revealed in many different ways, varying from highly structured and rather artificial methods to relatively unrestricted language production.
V. External Validity: Linguistic Intergroup Bias and Mass Media Yet, the above studies share one limitation: in all of these studies the participant’s verbal production or preferences were experimentally elicited and subjects were fully aware of their participation in a psychological experiment. Thus, it remained to be shown that the LIB would also occur in spontaneous language production. If the LIB is indeed such a robust phenomenon as it appears from experimental studies, then one can easily imagine an infinite number of communication settings in which it may play a subtle but potentially important role, including court cases, teaching, or mass communication. Mass communication is particularly intriguing considering that a single message may have a disproportionate impact as it reaches thousands of people simultaneously. To our knowledge, there are at least eight media studies analyzing the language abstraction in news stories involving in-group and out-group, four of which focus on sport reports (Maass, Corvino, & Arcuri, 1994, Exp. 1; Ng & Tait, 1994, Exp. 1 and 2; Schmid, 1995) and four on political events (Maass et al., 1994, Exp. 2 and 3; Dal Pos, 1994; Ng & Tait, 1994).
A. SPORTS REPORTS The existence of the linguistic intergroup bias in mass communication was first tested in an analysis of newspaper reports on soccer games in which the Italian national team had either won or lost (Maass et al., 1994, Exp. 1). All articles published the day after each game by two leading daily newspapers and by one daily sport paper were analyzed. For all sentences in which the in-group or the out-group appeared as the sentence subject, verbs or adjectives were rated for valence as well as for language abstraction using Semin and Fiedler’s (1988) coding scheme. In line with the LIB model, positive behaviors of the out-group were almost exclusively described in concrete terms (93%), whereas concrete terms constituted only about 75% in the remaining conditions. Apparently journalists shifted their language towards the concrete pole of the continuum when describing socially desirable out-group behaviors, implying that such behaviors were limited to a
88
ANNE MAASS
specific situation and moment. This bias, however, was observed only for those phrases that were indicative of high identification of the journalist with the national team (Italy vs. opposing country or we vs. they), but did not occur in those statements that implied less personal identification (statements about the team or single players). Although this study has produced results that are consistent with the LIB model, it has the shortcoming that-due to language differences-only communications from one side (Italian newspapers) were considered. This problem was resolved in a series of studies by Ng and Tait (1994) who analyzed articles that were published in different countries but shared the same language (Australian vs. New Zealand newspapers commenting on the World Cricket Series of 1981, and British vs. New Zealand newspapers commenting on an incident involving a New Zealand cameraman and an English cricket coach). Language was found to be largely unbiased for positive statements, but negative statements about the out-group tended to be much more abstract than negative statements about the in-group. Thus, not only were negative comments about the in-group rare, but when they occurred, they tended to be highly concrete, suggesting that the described events represented isolated facts unrelated to enduring properties of the in-group. A final content analysis regards 29 wrestling reports published in the World Wrestling magazine in which comments about face (good guy) and heel (bad guy) were compared (Schmid, 1995). Not only were faces described more positively than heels, but the analysis of language abstraction fully confirmed the LIB model: positive statements were more abstract when referred to the face than when referred to the heel, whereas the opposite occurred for negative statements. Taken together, the above analyses of sports reports suggest that journalists are not quite as objective as one may wish. Overall, concrete descriptions predominate in journalistic language, but this tendency appears particularly pronounced when describing positive out-group (Maass, et al., 1994, Exp. 1;Schmid, 1995) and negative in-group behaviors (Ng & Tait, 1994; Schmid, 1995). This confirms the LIB in journalistic language, but only within the specific realm of sport reports. Sport reports may differ from political reports in many ways; in particular, in-group favoritism may be considered more legitimate, and hence be expressed more readily, in this context. We therefore proceeded to test the LIB in socially more relevant political contexts. B. POLITICAL REPORTS The first analysis of mass communication on political events referred to an anti-Semitic episode in which neofascist groups in northern Italy exposed
89
LINGUISTIC INTERGROUP BIAS
banners depicting swastika and anti-Semitic slogans during a basketball game between an Italian and an Israeli team (Maass et al., 1994, Exp. 2). The main interest of this study was to compare how this episode was reported in Jewish and non-Jewish journals both published in Italian language. Again, all phrases referring to the aggressors (neofascist groups, hooligans) or to the victim (Israeli players, Jews) were coded for valence as well as for language abstraction. An interesting pattern emerged from the analysis of the two types of papers: regardless of the source (Jewish vs. non-Jewish), the aggressor was described as clearly negative. However, the level of abstraction chosen to express such statements varied considerably between Jewish and nonJewish papers (see Figure 5 ) , with abstract statements about the aggressor being five times as frequent in Jewish than in non-Jewish reports. Importantly, this is not attributable to a general difference in style, as Jewish and non-Jewish papers did not differ in their descriptions of the victim. Thus, both Jewish and non-Jewish journalists criticized the neofascist groups, but non-Jewish journalists described the event much more concretely than their Jewish colleagues, thereby avoiding any generalization beyond the specific event.
W Concrete 100
80 90
0 Abstract
I
I
Jewish
NonJewish Actor
Jewish
NonJewish Victim
Fig. 5. Percentage of concrete versus abstract descriptions of aggressor and victim by Jewish versus non-Jewish journalists. (Adapted from Maass et al. 1994.)
90
ANNE MAASS
Further evidence for biased language comes from a comparison of how student newspapers (prostudent) and community newspapers (propolice) commented on a student riot in New Zealand (Ng & Tait, 1994). Positive statements about the rioting students versus police were found to be largely unbiased, but negative statements varied greatly in abstraction depending on whether they were referred to in-group or out-group. Abstract formulations (30%) were quite common when commenting on the out-group, but completely absent when commenting on the in-group. Thus, again, journalists not only described the in-group less frequently in negative terms than the out-group, but when they did so, they used only concrete terms that imply little generalization beyond the specific event. A third example comes from an unpublished study by Dal Pos (1994) investigating how northern and southern Italian press reports on Mafiarelated crimes in Sicily (assassination of Falcone in May 1992 and of Borsellino in July 1992). Dal Pos observed no reliable differences between northern and southern publications for the first crime, but the repetition of a similar assassination after only few months induced a marked change in the style of northern reports, which became both more negative and more abstract. In contrast, the reports in Sicilian newspapers became somewhat more concrete. Interestingly, the greatest difference in style between northern and southern newspapers was observed for commentaries, whereas news stories remained consistently concrete throughout the two time periods, regardless of source. This suggests that language biases may be operating mainly in commentaries that have the explicit scope to interpret events, but less so in news stories that are presumed to report on events in a more objective way. Finally, one study has investigated language abstraction in broadcasting (Maass et al., 1994, Exp. 3). In this study, all eight o’clock news transmitted by Italian public television channels were analyzed during the first week of the Gulf War in 1991 as well as during the first week after the war. Looking separately at phrases referring to the in-group (allied forces or leaders) and to the out-group (Iraqi forces or leaders), a clear pattern emerges: The more negative the statements were about Iraq, the more abstract they tended to be (Y = .48). Interestingly, this bias disappeared quickly after the war. For the in-group, no reliable correlation between abstraction and positivityhegativity was observed either during or after the war. Taken together, the studies reported here indicate quite consistently that the LIB is operating in journalistic language, although it is rarely identifiable for the entire pattern of predictions; in most studies, bias is limited to one group or to one type of statement (either positive or negative), while it is
LINGUISTIC INTERGROUP BIAS
91
rarely simultaneously present in all combinations. Hence, the LIB may be somewhat more fragile in naturally occurring language production. At the same time, the overall pattern resulting from both controlled experiments and text analyses of media reports suggests that the LIB represents a widespread and robust phenomenon. Once we had established the robustness of the phenomenon, we turned to the more exciting question of which mechanisms sustain such differential language use in intergroup contexts.
M. Underlying Mechanisms: In-Group Protection or Differential Expectancy?
When we first started this research project (see Maass et al., 1989), we considered two possible mechanisms that may be responsible for the LIB, one motivational, the other cognitive in nature. The motivational account is loosely based on social identity theory (Tajfel & Turner, 1987; Taylor & Moghaddam, 1987) and assumes that the LIB serves to enhance or protect one’s social identity. Assuming that concrete descriptions dissociate the actor from the act, whereas abstract descriptions imply that the behavior reflects a stable and enduring property of the actor, one may argue that the LIB helps to portray the in-group in a favorable and the out-group in an unfavorable light. According to this account, the LIB represents one possible strategy of maintaining a positive image of one’s own group, very much like other well-known strategies, including biased reward or resource allocation or in-group-enhancing attributions and evaluations. An entirely different but equally plausible interpretation comes from a cognitive point of view and assumes that the LIB is driven by differential expectancies. A speaker may prefer abstract language whenever she or he observes behaviors that confirm prior expectancies, exactly because such expectancy-congruent behaviors are considered expressions of a stable and typical behavior tendency of the actor. If somebody expects Italians to be hedonistic then any behavior indicative of enjoyment of life should quite logically be described in abstract terms. In contrast, unexpected or surprising behaviors (such as sacrificing a pleasant evening with friends in order to finish up a paper) are considered atypical and as such ought to be described in concrete terms. According to this account, the LIB simply derives from the fact that, on the average, members of a given group usually expect more desirable and fewer undesirable acts from in-group than from out-group members (e.g., Howard & Rothbart, 1980).
92
ANNE MAASS
The main difference between the two explanations is that the expectancybased explanation predicts abstract language use for expected and concrete language use for unexpected behaviors, regardless of the valence of the behavior, whereas the in-group-protection explanation predicts abstract language use for positive in-group and negative out-group behaviors regardless of stereotypic expectancy. In many intergroup settings the predictions of the two approaches are likely to converge because people generally tend to expect positive behaviors from the in-group and negative behaviors from the out-group. So, why should we care to know whether the LIB is mainly driven by in-group-protection motives or by differential expectancies? Besides academic curiosity, there are at least two reasons why the identification of the underlying mechanisms seems important. First of all, there are a number of relevant intergroup settings in which the two approaches are likely to produce divergent predictions. Let’s consider the case of sex stereotypes, where men and women tend to hold surprisingly similar expectations about what constitutes typically male and typically female characteristics. For example, males may be expected to be independent but insensitive, whereas women may be expected to be sensitive to the needs of others, but lack initiative. If in-group-protection motivation is the driving force, then men and women ought to describe positive behaviors of their own group and negative behaviors of the out-group more abstractly than the remaining combinations, regardless of whether the behavior is typically masculine or typically feminine. In contrast, the differential expectancy view predicts that all subjects should describe role-congruent behaviors (masculine behaviors of male and feminine behaviors of female protagonists) in abstract, and out-of-role behaviors in concrete terms, regardless of valence. Thus, predictions of the two approaches will coincide only in those situations in which expectancy and valence are naturally confounded with people holding positive expectations about the in-group and negative expectations about the out-group. The two approaches will make distinct predictions in all those settings in which expectancy and valence are either orthogonal or at least partially independent. The second argument for why it is important to understand the underlying mechanisms regards the conditions that may enhance or reduce the bias. In particular, it seems difficult to create conditions that would undermine or reduce the LIB unless we know which mechanisms are driving it. B. EVIDENCE FOR DIFFERENTIAL EXPECTANCY Over the last few years, we have conducted a series of studies in which we have attempted to unconfound valence and stereotypic expectancy in
LINGUISTIC INTERGROUP BIAS
93
order to test the two explanations against each other. In a first unpublished study (Maass, Giordana, & Fontana, 1990), male and female subjects were presented with cartoons depicting male and female actors performing typically male or typically female behaviors of positive or negative valence (an example of typically male behavior of negative valence would be “to insult another driver,” whereas a typically female behavior of negative valence would be to “talk negatively about somebody behind his back”). Results show that regardless of sex of subjects, sex-role congruent behaviors were described more abstractly than out-of-role behaviors, although this effect was of very small magnitude. We therefore conducted a second experiment, again selecting an intergroup setting that would allow us to unconfound stereotypic expectancies and valence (Maass et al., 1995). In this experiment, northern vs. southern Italy was chosen as intergroup setting, as subjects from both regions held converging positive as well as negative expectations about the two groups. Southern Italians were seen as hospitable and warm but also sexist and intrusive, whereas northern Italians were believed to be industrious and emancipated but also materialistic and intolerant. Importantly, these stereotypes were shared by participants from northern and southern Italy. Subjects of northern or southern origin viewed cartoons in which southern or northern protagonists displayed these behaviors varying in valence as well as stereotypicality. They were asked to select one of four descriptions of each scene, corresponding to Semin and Fielder’s linguistic category model. The results showed absolutety no support for the in-groupprotection explanation, but they did show the predicted effect for typicality (see Figure 6, left portion, study 1). Behaviors that were typical of the protagonist’s group were described more abstractly than atypical or unexpected behaviors, and this occurred regardless of valence. This suggests that the LIB is driven by differential expectancies rather than in-groupprotective motivation or, at least, that differential expectancies are sufJicient to produce the LIB while motivational processes seem unnecessary. Since then, the idea that the LIB is an expectancy-guided phenomenon has received support from a number of additional studies involving different intergroup settings. Wigboldus (1998, Exp. 2) found that Dutch subjects described behaviors that were congruent with the protagonist’s membership more abstractIy than incongruent behaviors, but this was confirmed only for the out-group (Flemish), whereas no bias was observed for the ingroup (Dutch). Although these results provided only partial support for the expectancy bias, it should be noted that it is not rare in the LIB literature to find more robust effects for the out-group than for the in-group (Maass et al., 1989, 1994; Wigboldus, 1998, Exp. 1). Just as in the previous studies, valence was found to be irrelevant, suggesting that expectancy rather than
94
ANNE MAASS Congruent
3.5
Incongruent
T
Study 1
Study 2
Study 3
Fig. 6. Abstraction for expectancy-congruent versus expectancy-incongruent behaviors. (From Maass et al., 1995, Study 1,2, and 3 . Copyright 0 1995 by the American Psychological Association. Adapted by permission of the publisher.)
in-group-protective motivation was operating. In a somewhat different design, Rubini and Semin (1994) asked members of the Italian Communist Party and of the Catholic movement Comunione e Liberazione to provide explanations for why fellow in-group members may engage in expected or unexpected behaviors. They found that participants referred to more abstract concepts and, in particular, used reliably more adjectives when explaining party-congruent behaviors (such as attending the annual meetings of one’s own party) than when explaining incongruent behaviors (reading the newspaper of the opposing party). Interestingly, the tendency to describe expectancy-congruent behaviors abstractly is particularly pronounced under high need-for-closure. Webster, Kruglanski, and Pattison (1997) investigated the LIB under conditions in which people were inclined to reach closure either because of a general predisposition toward closure (Exp. 1) or because they were exposed to environmental conditions (aversive noise) that motivates closure (Exp. 2). In both cases, Webster et al. observed that individuals high (vs. low) in need-for-closure preferred high levels of abstraction to describe expected (positive in-group and negative out-group) behaviors, whereas need-forclosure was not found to affect unexpected behaviors. Following the interpretation of the authors, this pattern may be attributed to the drive for permanence that is generally associated with high need-for-closure. Under
LINGUISTIC INTERGROUP BIAS
95
high need-for-closure, people tend to hold on to past knowledge structures and safeguard them against disconfirmation, so it is not surprising that they ought to prefer highly abstract language when encountering information that is in line with their prior expectancies. If the LIB is, indeed, driven by expectancies, then an interesting implication emerges; namely that the bias can logically be extended to individuals for which we hold specific expectancies or to social groups that are irrelevant to our own social identity. If it is driven by differential expectancies rather than in-group-protective motives, then there is no reason to assume that the bias is limited to or specific of intergroup situations. Stereotypes are only one form of expectancies; indeed, most expectancies evolve around individuals or personality types (“she is intelligent, “he is sociable”). If language abstraction is so easily influenced by group-based stereotypic expectancies then there is little reason to believe that it should be insensitive to individual-based expectancies. In other words, one may hypothesize that the LIB is not confined to groups per se, but can be extended to the realm of individual behaviors unrelated to group membership. ”
C. BEYOND INTERGROUP RELATIONS: EXTENSION TO INDIVIDUAL BEHAVIORS In order to test this possibility, we investigated language use in a situation in which neither protagonist nor research participants belonged to a particular group (Maass et al., 1995, Exp. 2). As in the standard paradigm, participants viewed cartoons portraying positive or negative behaviors, but this time they were asked to imagine that their “best friend” or their “worst enemy” performed the behavior. Although category membership was presumably irrelevant here, a language bias equivalent to the LIB was observed: negative behaviors were described more abstractly when performed by a disliked other than when performed by a liked other, whereas the opposite trend emerged for positive behaviors (see Figure 6, middle, study 2). Almost identical results have been obtained by Wigboldus (1998, Exp. 1) on Dutch subjects, suggesting that the effect is not limited to the Italian setting. Unfortunately, however, in neither study can it be excluded with certainty that the liked protagonist did not belong to some (unknown) in-group and the disliked one to some (unknown) out-group; we therefore decided to submit the hypothesis to a stricter test by running an experiment in which the protagonist’s membership had no personal relevance at all to the subject and in which the expectancies were induced experimentally (Maass et al., 1995, Exp. 3).
96
ANNE MAASS
Following a procedure developed by Rothbart, Evans, and Fulero (1979), expectations were induced by telling participants that the target person had been described by friends, parents, and teachers as having a given characteristic, which, depending on the condition, was “sociable, ” “intelligent, ” “unsociable,” or “unintelligent. Subsequently, subjects received a cartoon showing the target person behaving in a way that either confirmed or disconfirmed the expectancy. Again, subjects had to select one of four descriptions varying in abstraction. The results show that regardless of the specific expectancy induced, expectancy-congruent behaviors were described more abstractly than expectancy-incongruent behaviors (see Figure 6, right portion, study 3). Importantly, this occurred in the complete absence of social category membership and/or personal involvement of our research participants. The design contained an additional factor, namely whether the expectancy was provided at the individual level (e.g., the target person was described as “sociable”) or, equivalent to a stereotype, at a group level (e.g., the target person was described as belonging to a “sociable” group). As expected, results were quite similar in the two conditions, but differences between congruent and incongruent behaviors were more pronounced when expectancies were individual-rather than group-based. This is quite in line with recent models on entitativity (e.g., Hamilton & Sherman, 1996), which assume that people expect great consistency and congruence within individuals but allow for some degree of variance within groups. It is therefore not surprising that the expectancy-based language bias was more pronounced for individuals: When a protagonist violated an individual expectancy that referred to his or her personal characteristics, people strongly shifted their language towards the concrete pole of the continuum, whereas this shift occurred to a lesser degree when the protagonist deviated from a group-level expectation. Taken together, these studies confirm the idea that the LIB model can be extended beyond intergroup settings and that expectancy-driven language biases may be found to the same or to an even greater degree when speakers hold beliefs about the characteristics of others that are unrelated to their group membership. Such expectancy-driven biases in language, which may either be group- or individual-based, have now been termed linguistic expectancy bias (LEB, Wigboldus, 1998). Hence, the LEB may play a significant role in belief maintenance in general as it constitutes a subtle form of maintaining not only group-related stereotypes but any belief. This suggests that the link between social beliefs and language abstraction is much more general than we had suspected when we originally formulated the LIB model. ”
LINGUISTIC INTERGROUP BIAS
97
D. A R E MOTIVATIONAL PROCESSES IRRELEVANT? The fact that the LIB can be produced by differential expectancies alone even in the absence of personal involvement of the speaker in a social group does not prove that in-group- and self-enhancing motivations are necessarily and always irrelevant. Indeed, the findings cited above do not exclude the possibility that in-group-protective motives may gain importance in more competitive or openly hostile intergroup settings. Some of the earlier studies, although not specifically designed to test in-groupprotective concerns, suggest that motivation may play a role in at least some circumstances. For example, the previously mentioned analysis of media reports on the Gulf War (Maass et al., 1994, Exp. 3) identified a strong language bias against the out-group during the war when intergroup relations were particularly tense, but this bias disappeared very quickly after the war. It seems difficult to explain this change on the basis of cognitive-expectancy processes alone. Additional indirect evidence that identity-protective mechanisms may be involved comes from Maass et al.’s (1994, Exp. 1) analysis of sport reports where reporters displayed a LIB on statements indicative of high in-group identification (e.g., we vs. rhey), but not on statements showing low personal involvement (e.g., talking about a single player). More recently, Maass et al. (1996, Exp. 1) have investigated the role of motivational processes in a more systematic way by focusing on highly competitive conditions. In the first study, we took advantage of an intergroup setting that we knew from previous research to be highly conflictual, namely hunters versus environmentalists. More importantly, we varied the degree of intergroup competition by either reinforcing or reducing the existing antagonism between the groups. In the former condition, participants received a statement in which the opposing group took a hostile stand against the in-group; environmentalists accused hunters of “endangering the natural patrimony of all,” whereas hunters accused environmentalists of “getting lost in endless political discussions.” In the latter condition, intergroup competition was reduced by information suggesting that the majority of the respective outgroup was displaying a cooperative attitude and had expressed the desire to “leave behind old attitudes of rivalry” and “to join forces” for the solution of increasingly severe environmental problems. The rationale behind this manipulation was that if in-groupprotective motivation is implicated, then the LIB should be considerably stronger under high than under low competition. As in previous research, we used cartoons in which hunters or environmentalists displayed positive or negative behaviors and assessed language abstraction. Results show that the LIB was considerably greater under competition than under cooperation (see Figure 7). These results were nicely paralleled
98
ANNE MAASS Cooperation
Desirable Undesirable
I
In-group
Competition
Out-group
Desirable
0 Undesirable
4f
3.5
1.5 I
In-group
Out-group
Fig. 7. Abstraction as a function of valence and category membership under competitive versus cooperative conditions. (From Maass et al., 1996. Copyright 0 1996 by the American Psychological Association. Adapted by permission of the publisher.)
by those obtained for a matrix measure where participants distributed public funds to hunting vs. environmentalist organizations. Thus, the same factor (competition) that increased in-group favoritism on the rewardallocation matrices (a measure generally employed in social identity research) also increased the magnitude of the LIB. However, the two strate-
LINGUISTIC INTERGROUP BIAS
99
gies of outgroup discrimination (reward allocation and LIB) were neither reliably correlated nor were they mediating each other, suggesting that intergroup competition exerts a similar but independent effect on both. Importantly, the LIB seems to be a successful strategy in protecting one’s personal and collective self-esteem, considering that participants who had displayed a greater LIB also showed better postexperimental self-esteem (which held true even when discrimination on the matrix measure was partialled out). This study suggests that motivational processes may, indeed, be driving language use when intergroup relations are particularly hostile or competitive. However, this study provides no information about the question whether the increase in motivational processes actually weakens the contribution of the cognitive, expectancy-based process or whether the two processes produce independent, additive effects. We therefore conducted a conceptual replication (Maass et al., 1996, Exp. 2) involving Northern vs. Southern Italians, both currently living in Switzerland, because the existence of positive and negative stereotypical traits for both groups allowed us to construct materials in which typicality and positivity were orthogonal. The advantage was that the simultaneous incorporation of typicality and social desirability into the design allowed us to estimate the relative strength of expectancy-based and motivationally driven processes under different degrees of in-group threat. A second design modification regarded the threat manipulation. In contrast to the previous study (Maass et al., 1996, Exp. l), all participants received an in-group-threatening message intended to trigger motivational processes, but in one condition hostility was induced between the two target groups (southern vs. northern Italians), in the other between a superordinate in-group (Italians) and a superordinate outgroup (Swiss). Thus, subjects in the north-south comparison condition received information about a large-scale survey that had revealed profound cultural and value differences between the two groups as well as frequent discriminatory reactions of outgroup members (e.g., northern Italians) against the subject’s own group (e.g., southern Italians). Subjects in the Italian-Swiss comparison condition received the same information, but in this case it was the superordinate in-group (Italians) that was discriminated against by a superordinate outgroup (Swiss). Overlapping or superordinate categorizations have consistently been shown to reduce intergroup discrimination in the social identity literature (cf. Vanbeslaere, 1991, 1996), so we hypothesized that in-groupprotective language use would be considerably stronger in the former than in the latter condition. Results confirmed this prediction. Participants described positive ingroup and negative out-group behaviors more abstractly than the remaining combinations, but this occurred mainly in the condition in which intergroup
100
ANNE MAASS
hostility was induced between southern and northern Italians. This becomes evident when considering a compound index of the degree of LIB: which was markedly greater when a comparison between southern and northern Italians was induced (M = 2.44) than when the Swiss-Italian comparison was made salient ( M = .23). An additional linguistic expectancy bias was observed such that typical behaviors were described more abstractly than atypical or unexpected behaviors regardless of the type of comparison that was made salient (southern vs. northern Italian or Italian vs. Swiss). Importantly, an increase of in-group-favoring language did by no means reduce the expectancy effect. Apparently then, the two processes are not mutually exclusive but may, at least under some circumstances, be operating independently and additively. An additional finding deserves attention. Note that southern Italians have a lower status in Italian society than northern Italians, and that this status difference is considered somewhat illegitimate, relatively stable, but potentially modifiable in the future. Under these circumstances, social identity theorists would predict a stronger in-group bias in the low-status group, which is exactly what was found on our language measure: southern Italians showed a markedly stronger LIB (compound index: M = 1.70) than northern Italians (M = .97). Taken together, these studies suggest that the same factors that have been found to activate in-group bias in social identity research also seem to affect the magnitude of the LIB. In particular, language bias increases under the influence of in-group-threatening information and in illegitimately disadvantaged groups but declines when superordinate categorizations become salient, suggesting that the LIB is indeed sensitive to motivational needs.
E. SELF VERSUS OTHER: LANGUAGE BIAS AND SELF-PROTECTIVE MOTIVATION Considering that language tends to be biased when in-group-protective needs are activated, the question arises whether a similar pattern may also be observed in self-other comparisons. There is at least one study in which self-enhancement-motivated differences in language abstraction have been tested in an interindividual context in which group membership was presumably irrelevant (Taris, in press). Subjects in this study were asked to generate positive (either moral or intelligent) and negative (either immoral or uninThe compound LIB-Index is generally calculated by adding the mean abstraction of positive in-group and negative out-group episodes and substracting the abstraction of negative in-group and positive out-group episodes. The LIB-index has a range from -6 (maximum out-group favoritism) to +6 (maximum in-group favoritism).
LINGUISTIC INTERGROUP BIAS
101
telligent) behaviors of self and others and their responses were subsequently coded for language abstraction. Findings paralleled those obtained in ingroup-out-group relations: positive behaviors of self were described more abstractly than similar behaviors of others, whereas the reverse was found for negative behaviors. The authors suggest that language abstraction seems to play a systematic role in maintaining and enhancing self-esteem regardless of whether the comparison involves in-group versus out-group or self versus others. Apparently then both cognitive (LEB) and motivational (LIB) mechanisms generalize from the intergroup to the interpersonal domain, demonstrating once more that the LIB model has implications well beyond the domain for which it was originally formulated.
F. EXPECTANCY AND IN-GROUP ENHANCEMENT: INDEPENDENT CONTRIBUTIONS?
I started this section with the question whether the LIB is driven by differential expectancies or in-group-protective motives. Considering the literature reported above, the answer to the question is that both mechanisms may contribute. Differential expectancies seem to be sufficient to produce differential language use in intergroup settings and probably in a wide range of additional settings. The link between cognition and language abstraction may be such that any expectancy (not only stereotypes) determine language abstraction, which in turn “protects” existing beliefs against disconfirmation. At the same time, the magnitude of the bias seems to vary predictably with motivational states. Intergroup competition, for example, was found to enhance this language bias in much the same way as it increases other forms of in-group favoritism and out-group discrimination. Although the possible interaction between cognitive and motivational factors has not been fully investigated at this point, it seems safe to conclude that the LIB as originally observed (see Maass et al., 1989) is the joint function of differential expectancies and in-group protective motives. This suggests an interesting link between stereotypes, in-group favoritism, and language. Although stereotypes and in-group favoritism are not completely independent (stereotypes tend to favor the in-group), they may have partially independent effects on language in intergroup contexts and may predominate in distinct situations. In the absence of in-group-threatening factors, stereotypic beliefs tend to determine language use, which in turn will contribute to the confirmation of such beliefs, thereby feeding into a self-fulfilling cycle. When in-group-protective motivations become very strong, they tend to override stereotypes and to exert a direct impact
102
ANNE MAASS
on language use, which then fulfils an in-group-favoring (rather than expectancy-confirming) function. It is not always possible to distinguish the two mechanisms a priori, and in many situations seemingly identical outcomes may result from distinct mechanisms. Despite this difficulty, it may be important to follow Wigboldus’s (1998) suggestion to distinguish the two phenomena conceptually and linguistically by referring to the LIB when desirable in-group and undesirable out-groups are described at higher levels of abstraction and to the LEB when expected behaviors are described more abstractly than unexpected behaviors.
VII. The Consequences of Abstract versus Concrete Language Use: Feeding into a Self-perpetuating Cycle I will now turn to the second part of the model in which it was argued that abstract versus concrete language use has predictable effects on the thought processes of the listener. When receiving abstract descriptions, the listener is expected to form the impression that the act reflects an enduring and stable behavior tendency of the protagonist and that the actor will behave in a similar manner in the future. Compelling evidence for the impact of language abstraction on the listener comes from studies utilizing the question-answer paradigm5 (Semin & de Poot, 1997; Semin, Rubini, & Fiedler, 1995), but studies using the LIB/LEB paradigm-which are of interest here-are surprisingly rare. The first study investigating the cognitive consequences of abstract versus concrete language use was conducted by Maass et al. (1989, Exp. 3) who presented participants with minimal sentences that had been used in a previous study to describe the experimental cartoon scenes. Participants were asked to rate how much information each phrase provided about the protagonist and how likely they thought it was that the protagonist would In the question-answer paradigm, research participants respond to questions that are formulated at different levels of abstraction (e.g., Why do you read newspaper X ? vs. Why d o you like newspaper X?). Research on this paradigm has essentially shown that question formulation biases the listener’s responses in two ways: First, question abstraction tends to drive answer abstraction, such that concrete questions elicit concrete answers and abstract questions elicit abstract responses. Second, action-verb questions (why do you read newspaper X ? ) elicit answers in which the previous subject occurs in the subject position of the answer (because I . . .). In contrast, state verb questions (why d o you like newspaper X ? ) elicit responses in which the previous object occurs in the subject position (because newspaper X i s , , .). Hence, the question-answer paradigm provides a powerful demonstration of how the speaker may (often inadvertently) manipulate the responses of the interaction partner.
103
LINGUISTIC INTERGROUP BIAS
display the same behavior or characteristic in the future. Importantly, participants had no information about the category membership of the protagonist and rated the relative informativeness and likelihood of repetition solely based on the minimal sentence. Results (see Figure 8) indicated the predicted linear increase on both ratings as a function of abstraction. The greater the abstraction at which the information was transmitted, the more information the phrase revealed about the protagonist, and the more likely it seemed that the actor would display a similar behavior in the future. Although this experiment demonstrates the predicted consequences of abstract language use, the experimental setting is far removed from a conversation-like situation. Wigboldus (1998, Exp. 3, 4, and 5 ) has recently subjected the same prediction to a much more realistic test in an interesting series of studies. In the first experiment, a group of participants received a story involving an in-group (Dutch) or an out-group member (Flemish) showing typical or atypical behaviors. They were asked to write a summary of the story for another group of participants who indeed read the stories and made inferences about the protagonist. Not only did the original subjects describe typical behaviors more abstractly than atypical ones, but the recipients also made stronger dispositional inferences about actors performing typical behaviors than about those performing atypical behaviors. Importantly, mediational analyses showed that these inferences were mediated by the writer’s language abstraction (for further support see Wigboldus, 1998, Exp. 5). Although this strongly supports the predictions of the LIB
-
= .-
lnformativeness Likelihood of repetition
3.9 3.7 3.5 3.3 3. I 2.9 2.7 &.J
.
DAV
IAV
sv
ADJ
Fig. 8. Implicit informativeness about protagonist and likelihood of repetition as a function of abstraction. (From Maass et al., 1989, Exp. 3. Copyright 01989 by the American Psychological Association. Adapted by permission of the publisher.)
104
ANNE MAASS
model, it cannot be excluded with certainty that the greater dispositional inferences may in part have been due to the recipient’s knowledge of what constituted typical versus atypical behavior for the protagonist. Even more convincing is a subsequent study (Wigboldus, 1998, Exp. 4) in which recipients had no knowledge about what constituted a typical or an atypical behavior of a given protagonist. In this study, a first group of subjects wrote two true stories in which a close friend had shown expected or unexpected behaviors. Recipients who had read these stories drew stronger dispositional inferences about the protagonist when the story described expected rather than unexpected behaviors. Importantly, the abstraction level at which the transmitter had described the episode was again found to mediate the recipient’s inferences. It is also interesting to note that language abstraction mediated the recipient’s but not the transmitter’s own dispositional inferences. Apparently, the speaker’s language has a reliable impact on the inferential processes of the listener, whereas the speaker’s own inferences are not subject to such influence-although there are findings that argue against this interpretation. There is at least two studies showing that the speaker’s own beliefs may be maintained through language use. In the first experiment, Karpinski and von Hippel (1996) induced favorable or unfavorable expectancies about an imaginary target person, Scott, whom participants imagined to be either a close, lifelong friend or a person they had disliked all their life. Scott was then described as engaging in positive or negative actions. Consistent with our model, expectancy-congruent behaviors (positive behaviors of the friend and negative behaviors of the enemy) were described more abstractly than expectancy-incongruent behaviors. More importantly, the authors also measured the extent to which participants maintained their initial impression of the target person by asking them to indicate their liking for Scott immediately after the initial description and again at the end of the experiment. Mediational analyses showed that beliefs were indeed maintained through language abstraction such that those participants that showed a strong LEB maintained their initial (positive or negative) impression of Scott throughout. However, the attitude-perpetuating function was limited to moderately confirming or disconfirming behaviors. When the protagonist’s behavior deviated too much from initial expectancies, subjects engaged in the LEB, but this did not prevent them from revising their initial beliefs about the target. Results of the second study were essentially the same, suggesting that the LIB/LEB does indeed serve a belief maintenance function unless the discrepancy between expectancy and observed behavior is too extreme: ‘Contrary to the previous experiment, participants did not display a LIBlLEB when expectancy disconfirmation was extreme, possibly due to the fact that expectancies were stronger in Exp. 1 than in Exp. 2 (see Karpinski & von Hippel, 1%6, for a discussion).
LINGUISTIC INTERGROUP BIAS
105
Taken together, the studies by Maass et al. (1989), Wigboldus (1998), and Karpinski and von Hippel (1996)-although few in numbers-suggest that biased language use fulfills both an interpersonal and an intrapersonal function of belief maintenance. On the interindividual level, it induces inferences and expectancies in the listener that are consistent with the speaker’s beliefs, thereby contributing in a subtle way to the social transmission of beliefs. It is easy to imagine how beliefs become shared knowledge within a social group, if such bias is shown consistently by its members. On the intraindividual level, such language use helps to maintain the speaker’s own beliefs in the face of incongruity. In both cases, biased language use tends to feed into a cycle in which initial beliefs such as stereotypes are protected against disconfirmation, which in turn contributes to the stability of intrapersonally as well as interpersonally shared beliefs systems.
VIII. Open Problems The above review of the literature suggests that almost 10 years of research on the LIB/LEB phenomenon have produced some important insights in the conditions under which it operates, the mechanisms by which it is driven, the methods by which it can be assessed, and the cognitive and social consequences it may produce. At the same time, there are a number of important “open problems” that have either received little attention, or that have been addressed but where a solution is not yet in sight. I will briefly discuss four of these issues concerning (a) the development of the LIB/LEB in children, (b) the role of the recipient in conversational settings, (c) the relation between abstraction and (dis-)confirmability, and (d) the question of intentional control over the bias.
A. DEVELOPMENTAL ISSUES To date, very little is known about variations of LIB/LEB across the life span and, in particular, about the development during childhood and adolescence. There is some evidence that children before 8 years of age use abstract concepts to explain human behavior to a lesser extent than older children or adults (Livesley & Bromley, 1973; Maass & Franco, 1997) and that they encounter difficulties in understanding the inferential properties of concrete versus abstract interpersonal verbs (Werkman & Semin, 1997). Both of these findings suggest that small children may simply not have the prerequisites required to make the fine-grained differences that constitute the LIBILEB.
106
ANNE MAASS
To date, there are only very few LIB/LEB studies that have involved children, and these studies have produced inconsistent results. Wigboldus (1998, Exp. 1) found no differences in language bias among participants aged 11 to 19, but it is possible that the majority of 11-year-olds already possess relevant language capacities that are comparable to those of adults. Casotti (1996/97) investigated the LIB (together with nonlinguistic in-group bias measures) in soccer players aged 8 to 19, an intergroup setting believed to be equally relevant to younger and older research participants. Contrary to predictions, no reliable LIB emerged in either age group, and no differences emerged as a function of age, although this may in part be attributable to procedural flaws of the study (see Casotti, 1996/97 for a discussion). The only study that did show reliable age differences was conducted by Werkman, Wigboldus, and Semin (1999), in which participants aged 8 to 19 described cartoons of desirable or undesirable behaviors performed by either their “best friend” or “worst enemy.” Thus, the study was concerned with interpersonal rather than intergroup relations. In line with predictions, the authors found a linear increase in bias with age; although all age groups displayed a systematic bias, it became stronger with increasing age. Unfortunately, the lack of additional language-free bias measures in this study makes it difficult to infer whether older participants experienced greater in-group favoritism or whether they were better able to express it through differential language abstraction. At this point, we are left with a small, inconsistent, and methodologically limited set of studies that shed little light about the development of the LIBlLEB in children. Considering the relevance of this issue, one can only hope that future studies will address this important issue in rigorous, multimeasure studies, possibly extending the inquiry to preschool-aged groups. B. THE ROLE OF THE MESSAGE RECIPIENT Another underresearched area regards speech behavior in communication-like interactive settings. Although the model is explicitly concerned with the interpersonal transmission of stereotypes, the predominant research paradigm is far removed from a realistic communication process. Note that most studies were conducted in rather artificial situations in which the participant either selected or produced descriptions of hypothetical scenes without knowing the recipient of the “message.” The one notable exception to this rule is Wigboldus’s (1998, Exp. 5 and 6) recent work, in which participants described true events to others who did or did not belong to the same group. This paradigm not only adds external validity to the
LINGUISTIC INTERGROUP BIAS
107
area but also allows one to investigate variations in LIB/LEB as a function of the interaction partner. In the first study (Study 5 ) , subjects described sex-role-consistent or inconsistent behaviors of a same-sex or opposite-sex friend to another recipient who was of the same or opposite sex, or whose sex was unknown. The LEB effect (greater abstraction for sex-role congruent behaviors) was obtained in all conditions except in the case in which both target and recipient belonged to the in-group, a situation that was presumably perceived as an intra-rather than intergroup situation. A subsequent study (study 6) showed that a similar argument can also be made for the LIB (greater abstraction for positive in-group and negative outgroup episodes), which again was greater when subjects communicated with an out-group rather than an in-group member. Taken together, these results suggest that speakers are sensitive to the communicative context and tend to display an enhanced LEB/LIB when the presence of an outgroup recipient defines the situation unambiguously as an intergroup setting. More generally, these findings suggest a reciprocal influence between transmitter and recipient in which the recipient’s characteristics influence the transmitter’s message, which in turn influences the recipient’s inferences. These first studies suggest that the LIBLEB phenomenon cannot be fully understood without taking into account the social context in which the interaction is embedded.
C. ABSTRACTION AND (DIS-)CONFIRMABILITY The third issue that is of particular relevance to stereotype revision, but that remains an open problem, is the relationship between abstraction and (dis-) confirmability. Assuming that stereotypic attributes present inaccurate or overgeneralized descriptions of social groups, the disconfirmability of such attributes becomes the single most important precondition of stereotype revision. For instance, intergroup contact will be conducive to stereotype reduction only to the extent that such stereotypes are potentially disconfirmable (Rothbart & Park, 1986). What then is the relation between language abstraction and (dis-) confirmabiiity? Are relatively abstract stereotypes (e.g., women are dependent) more or less open to (dis-) confirmation than concrete stereotypes (Blacks dance well)? On theoretical grounds, two opposite predictions may be derived from the literature. Semin and Fiedler (1988,1992) refer to “verifiability” as the ease or difficulty with which the actions or states described by a verb or adjective can be objectively verified. Their linguistic category model makes straightforward, linear predictions: with increasing abstraction, statements become more difficult to confirm as well as to disconfirm.
108
ANNE MAASS
Because concrete verb classes are the least ambiguous, both confirmation and disconfirmation should be easiest with concrete verbs, and become increasingly difficult as abstraction increases. This is exactly what was found when presenting participants with minimal sentences and asking them to rate “to which extent the statement can be objectively verijied” and “how easy is it to imagine specific, observable behaviors that would provide confirmation” or “disconfirmation” of the statements (Semin & Fiedler, 1988, 1992). For all three ratings, strong linear trends were observed such that abstract statements were most difficult to verify, to confirm, as well as to disconfirm. As pointed out by Semin and Fiedler (1992) this contradicts the position taken by Rothbart and Park (1986) in the domain of positive versus negative behaviors. These authors have demonstrated that, in general, negative traits are easiest to acquire and most difficult to lose-thus predicting a negative (rather than positive) relation between confirmability and disconfirmability. For negative traits-which are particularly common in outgroup stereotypes-little confirming evidence is required before they will be considered confirmed, but much disconfirming evidence is needed before they will be revised. In their research, Rothbart and Park (1986) presented subjects with a list of trait adjectives varying in favorability and asked them to rate, among others, how many confirming (or disconfirming) behaviors would be required before a given trait can be considered confirmed (or disproved). The most important result emerging from this study is that unfavorable traits require fewer instances of confirmation before they can be considered confirmed, but more disconfirming instances before they can be repudiated. As Rothbart and Park (1986, p. 135) put it, “unfavourable traits are easier to acquire and harder to lose than favourable traits” (for replications and extensions see Arcuri & Cadinu, 1992; Funder & Dobroth, 1987). A recent study by Maass et al. (1998) has applied this line of research to stereotyping looking at both confirmability and language abstraction. We wanted to test whether negative out-group and positive in-group stereotypes would be linguistically more abstract than the remaining combinations and whether they would require more disconfirming (but less confirming) evidence before being repudiated. In the first phase of the study, non-Jewish Italians were asked to indicate stereotypical beliefs about three social groups: Italians (in-group), Jews, and Germans (out-groups). Their responses were subsequently rated for abstractness. In the second phase of the study, the most frequently mentioned (hence: socially shared) stereotypical statements were then judged for confirmability and disconfirmability by an independent sample who rated how many behavioral
109
LINGUISTIC INTERGROUP BIAS
instances would be required to confirm or disconfirm each of the stereotypical traits. Results showed that positive and negative in-group stereotypes had a comparable level of abstraction, whereas for both out-groups negative statements tended to be reliably more abstract than positive ones (see Figure 9). More importantly, as can be seen in Figure 10, positive out-group traits (including traits such as intelligenr, intellectual, precise, organized ) required the greatest and negative out-group traits (including stingy, violent, milituristic) required the least amount of confirming evidence before they could be considered “true.” Also, for the outgroups, more disconfirming evidence was required for rejecting negative than for rejecting positive traits. Taken together, it is the negative out-group stereotype that requires least confirming evidence before being accepted as “true” and most disconfirming evidence before being rejected as “false.” Thus, negative out-group characteristics are not only highly abstract, but they also tend to be difficult to disconfirm. Although the design of this study precludes correlational analyses, the data pattern is clearly in line with Rothbart and Park’s observation that
Positive
0 Negative 4.5
T
I-
1
Italians
Jews
Germans
Fig. 9. Mean abstraction of positive and negative stereotypes about Italians (in-group), Jews, and Germans. (From Maass, Montalcini, B Biciotti, 1998. Copyright John Wiley & Sons Limited. Reproduced with permission.)
110
ANNE MAASS Positive
0 Negative
Confirming
Disconfirming
Fig. 10. Mean number of confirming and disconfirming behaviors required in order to consider a stereotypic trait confirmed or disconfirmed. (From Maass, Montalcini, & Biciotti, 1998. Copyright John Wiley & Sons Limited. Reproduced with permission.)
traits that require much confirmation require little disconfirmation and vice versa. How can one reconcile this finding with Semin and Fiedler’s (1988, 1992) claim that abstraction is positively related to both confirmability and disconfirmability? Theoretically, it is conceivable that interpersonal verbs that are randomly selected from the vocabulary of a given language correlate positively with both confirmability and disconfirmability-just as observed by Semin and Fiedler. It is indeed quite logical to suppose that those terms that are easiest to confirm are also the ones that are easiest to disconfirm. At the same time, stereotypes may deviate in a systematic way from such random samples of the vocabulary. Certain verbs or adjectives in a given lexicon may have a greater chance to become part of an outgroup stereotype: namely those that are negative, abstract, easy to acquire, and difficult to lose. Thus, the relation between abstraction and disconfirmability may be different in outgroup stereotypes than in the lexicon at large-although this possibility is awaiting further investigation. From an applied point of view, the above findings seem interesting for a number of reasons. First of all, they offer one possible explanation for why stereotypes that are clearly nonveridical (e.g., women are poor drivers) do not easily fade over time even in settings that provide opportunities for disconfirmation. The reason may lie in the fact that they are highly abstract and intrinsically difficult to disconfirm.
LINGUISTIC INTERGROUP BIAS
111
Second, the finding that negative out-group characteristics are particularly abstract and particularly resistent to disconfirmation may have implications for our understanding of how stereotypes change or persist over time. In particular, if concrete language is more open to disconfirmation than to disconfirmation, then one may conclude that abstract stereotype content will be particularly resistant to change. Hence, stereotypes may become increasingly abstract over time, as they will lose their concrete components but maintain the abstract ones. This, in turn, should render them even less disconfirmable over time, resulting in a selfperpetuating cycle. This possibility was addressed in an exploratory study by Maass et al. (1998, Exp. 2). As a thought experiment, let’s assume that at any time in history, new stereotypes were created as intergroup relations changed. For example, when women started to have access to professional careers not too many years ago a new stereotype, “career woman,” emerged. Let us assume that stereotypes initially contain both concrete and abstract components but that the nonveridical concrete components of the stereotype disappear relatively quickly, whereas the more abstract components survive even in the face of contradicting evidence because abstract components require less confirmation to be accepted and more disconfirmation to be repudiated. If correct, then stereotypes should lose those nonveridical components that are expressed at a lower level of abstraction, but maintain those that are highly abstract. Consequently, the older the stereotype, the more abstract it will tend to be and the more resistant it will be to any further change, whereas more recent stereotypes should contain a larger proportion of concrete components. This possibility was tested by comparing stereotypical descriptions of different groups, one with a long history of being the object of stereotyping (Jews), one of recent development (career women), and two occupying an intermediate position (homosexuals and blacks). We asked participants to indicate the stereotype for each of the four groups and rated their responses for abstraction. As can be seen in Figure 11, abstraction showed the same trend as historical order: abstraction was highest in the group with the oldest stereotype, namely Jews, followed by homosexuals and Blacks, whereas the most concrete language was used to describe the most recent group, namely career women. Converging evidence also comes from a partial correlation between abstraction and historical order while controlling for positivity, r(40) = .30, p < .05, showing that stereotypes with longer histories tended to be more abstract. Although this study is only exploratory and based on admittedly “soft” data, these findings provide a first demonstration that stereotypes may indeed become more abstract over time. If confirmed by more stringent
112
ANNE MAASS Historical order
0 Abstraction 4
3.5 3 2.5
2
I .5 1 0.5
0
II Jews
Blacks
-I
Homosexuals
Career women
Fig. 11. Historical order and mean abstractness of stereotypes about Jews, blacks, homosexuals, and career women. (From Maass, Montalcini,& Biciotti, 1Y98.Copyright John Wiley & Sons Limited. Reproduced with permission.)
historical analyses, this would suggest that language may play an important mediating role in long-term development of socially shared beliefs, with stereotypes becoming increasingly abstract over time and, hence, increasingly resistant to change.
D. STRATEGIC OR UNINTENTIONAL? A final issue that has not entirely been resolved regards the extent to which speakers are aware of and able to exert intentional control over the LIB/LEB. An intriguing possibility is that people are unlikely to consciously control language abstraction and that they are generally unaware of the subtle variations in linguistic strategies. I am obviously not arguing that language is generally used in an unintentional fashion nor that linguistic choices are in principle inaccessible to the speaker. To the contrary, at any utterance the speaker selects one option from an almost infinite pool of alternatives that any given language provides. This is frequently done in a
LINGUISTIC INTERGROUP BIAS
113
deliberate manner by adopting specific rhetorical strategies, by carefully selecting (or avoiding) specific adjectives, by adding qualifiers that enhance or soften a given statement, and so on. Yet, there are some aspects of language-including language abstraction-which people may usually pay little attention to and over which they may rarely exert intentional control. In the case of the LIB, people may well be aware of the fact that they are describing a given group positively or negatively, but they may inadvertently use a greater or lesser degree of abstraction without being fully aware of this fact. A number of authors (Franco & Maass, 1996a, 1996b; Maass & Arcuri, 1996; Schmid & Fiedler; 1996; von Hippel, Vargas, & Sekaquaptewa, 1994; von Hippel, Sekaquaptewa, & Vargas, 1995,1997) are now arguing that people have little awareness of the subtle variations in linguistic abstraction, and may even be unable to learn from feedback regarding these strategies (Schmid & Fiedler, 1996). Unfortunately, the evidence in favor of this hypothesis is not completely consistent and derives for the most part from correlational research. The first indirect evidence comes from the earlier mentioned content-analysis of journalistic reports on the Gulf War (Maass et al., 1994). Comparing the (right-wing) first and the (left-wing) third public TV channel in Italy, we found that news reporters of the third channel took a more critical stand on the Gulf War but showed the same LIB as their ChristianDemocratic colleagues from Channel 1 (Maass, Corvino, & Arcuri, 1994). One interpretation is that left-wing journalists may have been able to reduce explicit out-group discrimination but they may inadvertently have displayed the LIB. The possibility that the LIB/LEB escapes intentional control was tested more explicitly in a correlational study by Franco and Maass (1996b). Here, two basketball fan clubs were compared, one of which had a long history of intergroup aggression, whereas the other was known for considering hooligan-type behaviors unacceptable. The main goal of this study was to test whether the group that did not tolerate intergroup violence would show lesser out-group derogation than the group without such norms on explicit measures, but not on subtle language bias. If the LIB operates in a largely unconscious fashion, then the “fair” group would find it difficult to inhibit this bias despite the fact that it should encounter no difficulty in limiting out-group derogation on explicit measures. The study included two explicit and two implicit measures of out-group derogation. The explicit measures were (a) reward allocation to in-group versus out-group and (b) freely generated trait attributes describing ingroup and out-group which were subsequently rated for valence. Not surprisingly, the fan club that held norms against out-group discrimination
114
ANNE MAASS
showed reliably lesser bias on both of these measures. They distributed hypothetical goods more equally, and they described the out-group in only mildly negative terms. Yet, they showed the LIB to the same extent as the fan club that lacked fairness norms. This was true not only for the standard cartoon measure but also when analyzing the self-generated traits for derivation. Positive adjectives describing the in-group and negative adjectives describing the out-group had more abstract derivations than the remaining adjectives, regardless of the fan club’s more or less restrictive normative beliefs. One explanation of these findings is that the group that tended to present itself in a nonaggressive light succeeded in this attempt on explicit measures, but not so on more subtle measures based on language abstraction. Apparently then the inhibition of out-group derogation is particularly difficult on the LIB. Although other explanations cannot be ruled out with certainty at this point, these data can be interpreted as suggesting that reward allocation, trait attributions, and probably numerous other strategies of in-group favoritism or out-group discrimination can easily be inhibited, whereas people encounter great difficulty in suppressing the LIB. Although these data are in line with the idea that the LIB is largely unintentional, they should be treated with caution for two reasons. First, the few data available so far are purely correlational. Second, there are other findings that would argue against the assumption that the LIB escapes conscious control. In particular, the failure to find any prejudice against Blacks among European students (see Beaton & Brown, 1996; Loda, 1995-96; Mikula & Graf, 1994; Palmer, 1997) could be interpreted as indicating that people may exert some control over the LIB-unless one makes the (counterintuitive) assumption that European students are truly unprejudiced. To this point, there is no experimental evidence that people do indeed find it difficult to suppress the LIB when motivated or instructed to do so, nor has language abstraction ever been tested in those paradigms that are generally considered suitable for demonstrating automaticity (e.g., concurrent memory load paradigm, Bargh & Tota, 1988). In other words, nothing definite can be said about the degree to which people are able to exert control over the LIB until more rigorous studies are conducted. However, if confirmed by stronger evidence, then a number of interesting implications emerge which will briefly be discussed here. Most importantly, if intentional control over the LIB/LEB is, indeed, limited, then it could constitute a useful prejudice measure in those situations in which people would tend to give socially desirable responses. Von Hippel and collaborators (1995, 1997) have argued very convincingly that a LIB-based prejudice measure may have considerable predictive value and may be better suited to predict prejudiced reactions than traditional prejudice scales. The procedure used by von Hippel and collaborators
LINGUISTIC INTERGROUP BIAS
115
consists of the presentation of a series of fictitious newspaper articles about blacks and whites (Exp. 1) or women and men (Exp. 3) involved in actions that vary in both stereotype congruence and valence. The participant’s task is to rate, for each scene, the appropriateness of different descriptions corresponding to Semin and Fiedler’s (1988) linguistic category model. This measure is then used as a predictor for prejudiced reactions to or judgments of outgroup members. The results showed that the LIB-based measure reliably predicted differential evaluation of the outgroup over and above explicit measures such as the modern racism scale. Thus, a LIB-based prejudice measure may be a valid alternative to traditional, explicit prejudice measures as it limits the possibility of falsification due to social desirability concerns. There are at least two reasons why subjects may find it difficult to monitor their responses according to social desirability. On the one hand, subjects are never asked to evaluate the behavior of the protagonist but simply to describe it. Social desirability concerns may be less likely to arise under these circumstances. On the other hand, even if the participant is motivated to give a nonprejudiced image, she or he may be unaware of the link between language abstraction and prejudice that renders the LIB a potentially useful tool for measuring “true” prejudice. Although there are undoubtedly more powerful indirect measures of prejudice (such as those based on reaction time to primed stimuli-see Fazio, Jackson, Dunton, & Williams, 1995), LIB-based prejudice measures have the advantage that they can be used very easily in almost any setting without requiring laboratory space or computer equipment; more importantly, these procedures can be applied to subject populations, such as the elderly or small children, that have no familiarity with computers (see Maass & Franco, 1997). If people are indeed largely unaware of language abstraction in their speech production, then such LIB-based prejudice measures may prove particularly useful in a normative context that prevents people from expressing prejudice or in populations that tend to inhibit prejudice. In line with recent research on the paradoxical effects of thought suppression (see Macrae, Bodenhausen, Milne, & Jetten, 1994), subtle language bias may be particularly strong in subjects and in situations in which the expression of explicit out-group hostility is inhibited by normative pressures. There is now some indirect evidence that this may indeed be the case. For example, hunters in Maass et al.’s (1996) study expressed lesser explicit discrimination (reward allocation) but greater language bias than environmentalists possibly due to the fact that they enjoyed lesser public support and therefore settled for more subtle forms of out-group derogation. Another recent study by Franco and Maass (in press) compared reactions to a minority
116
ANNE MAASS
that was normatively protected against discrimination (Jews) with one that did not enjoy such protection (Islamic fundamentalists) in Italian society. Prejudiced responses were assessed through various explicit measures (reward allocation, liking) as well as LIB. In line with predictions, the LIB measure correlated well with the explicit measures but only in the absence of norms against discrimination (Islamic fundamentalists). Explicit measures and LIB were perfectly uncorrelated (and in some cases even negatively correlated) for the Jewish out-group-which is exactly what should be expected if one assumes that people express their “true” feelings on the language measure but give socially desirable responses on the explicit measures. Taken together, the above studies suggest that the LIB may be a useful unobtrusive index of “true” feelings toward a minority group and that they may produce distinct patterns of results under those circumstances in which people are motivated to inhibit racist responses either due to normative pressure or to personal motivation. At the same time, it should be noted that there is not yet a LIB-derived prejudice measure that can be said to meet the psychometric standards generally applied to attitude and/or personality measures. Despite this limitation, the few data available so far are encouraging, as they suggest that the LIB could potentially offer a valuable alternative to popular self-report scales such as the modern racism scale (von Hippel et al., 1997).
IX. Conclusion There are four general points that I would like to stress in concluding this chapter. First, and most generally, this research program suggests that language does indeed play an important, albeit subtle, role in prejudice perpetuation. Obviously, I am not arguing that language abstraction is the only, or even the most important language strategy by which stereotypes are maintained intra- and interpersonally. One can envisage an infinite number of other, and possibly more powerful, strategies. For example, people may express subtle bias by using linguistic terms that render one group salient while remaining silent about the other. The generic masculine is the most carefully investigated bias through which one group (males) is made linguistically salient, while the other group (females) is omitted from discourse (see Maass & Arcuri, 1996; Moulton, Robinson, & Elias, 1978; Ng, 1990). Another, rarely investigated, way in which speakers may express bias is the use of modifiers such as adverbs. Sentences like “Italians are sociable” versus “Italians are usually quite sociable” certainly imply differ-
LINGUISTIC INTERGROUP BIAS
117
ent degrees of generality. Even the choice of nouns versus adjectives may have nontrivial effects on how a communication is interpreted and what associations are evoked, although to my knowledge, this issue has not yet been investigated systematically. Already Allport (1954) argued that a statement like “ M . is a Jew” may carry more information and elicit stronger affective associations than the statement “ M . is Jewish.” Yet another way to express bias is through the choice of past versus present tense. The phrase “L. hit his daughter” (past tense) implies an isolated act, whereas the phrase “L. hits his daughter” (present tense) implies an enduring tendency toward child abuse. These, and many other subtle (yet largely unexplored) variations in language may contribute to the maintenance of stereotypic beliefs. What our research program shows, more than anything else, is that it is indeed worthwhile to investigate such subtle linguistic strategies through which stereotypic beliefs and dislike for outgroups are maintained and communicated interpersonally. The second conclusion that can be drawn with some degree of certainty is that there are two distinct processes that drive language abstraction, one related to (stereotypic) expectancies, the other to ingroup- or self-protective motivation. Although the two mechanisms tend to be naturally confounded in many real-world settings, they have successfully been isolated in experimental situations. This has led to two important advances in understanding the LIBlLEB phenomenon. On one side, we are now able to predict which conditions are likely to enhance (or reduce) the bias, including any factor that renders prior expectancies salient or that increases the motivation to protect the status of the in-group. On the other side, the identification of the underlying mechanisms has led to a series of important extensions beyond the intergroup situation for which the LIB model was originally formulated. Consequently, the third implication of this review is that the phenomenon is much more general than initially hypothesized. Although the main interest of our research team is in intergroup relations and stereotyping, it has now become clear that similar processes that drive language abstraction in intergroup situations also operate in interpersonal settings. This is true for both differential expectancy and self-protective motivation, suggesting that the two functions of language abstraction, belief stabilization and selfesteem maintenance, apply both in the realm of interpersonal and intergroup relations. Finally, I would like to stress a particularly intriguing feature of the LIB, namely the subtlety of its contribution. Although LIB research has not been sufficiently systematic in this respect, many observations suggest that language abstraction plays a powerful but hidden role in stereotype maintenance. As most dramatically demonstrated by Semin and de Poot’s (1997)
118
ANNE MAASS
research on the question-answer paradigm, speakers and listeners may be completely unaware of language abstraction, yet the listener’s thought processes may be strongly affected by it. The fact that speakers and listeners rarely pay attention to the linguistic abstraction in everyday discourse makes it a potentially very valuable research tool for prejudice researchers. References Allport, G. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Arcuri, L., & Cadinu, M. R. (1992). Asymmetries in the attributional processes: The role of linguistic mediators. In L. Arcuri & C. Serino (Eds.) Asymmetry phenomena in interpersonal comparison: Cognitive and social issues, Napoli: Liguori, 1992. Arcuri, L., Maass, A., & Portelli, G. (1993). Linguistic intergroup bias and implicit attributions. British Journal of Social Psychology, 32, 277-285. Bargh, J. A,, & Tota, M. E. (1988). Context-dependent automatic processing in depression: Accessibility of negative constructs with regard to self but not others. Journalof Personality and Social Psychology, 54, 925-939. Beaton, A., & Brown, R. (1996). Reactivity in measures of prejudice. Unpublished manuscript, University of Kent, UK. Casotti, P. (1996/97). Linguistic Intergroup Bias: Quattro fasce d’etu a confronfo ([Linguistic Intergroup Bias: A comparison of four age groups]. Unpublished master’s thesis, University of Padua, Italy. Dal POS,G. (1994). Linguistic Intergroup bias e mass media: analisi di articoli di cronaca riguardanti delitti di mafia; verifica del potere confermante e falsifcante di aggettivi a diversi livelli di asrrazione [Linguistic intergroup bias and mass media: analysis of articles regarding mafia crimes; the verification of the confirming and disconfirming power of adjectives at different levels of abstraction]. Unpublished master’s thesis, Padua University, Italy. Fazio, R. H., Jackson, J. R., Dunton, B. C., & Williams, C. J. (1995). Variability in automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of Personality and Social Psychology, 6Y, 1013-1027. Fiedler, K., Semin, G. R., & Finkenauer, C. (1993). The battle of words between gender groups: A language-based approach to intergroup processes. Human Communication Research, 19, 409-441. Franco, F. M. & Maass, A. (1996a). The linguistic intergroup bias: Is it under intentional control? Paper presented at the Conference of the European Association of Experimental Social Psychology, Gmunden, Austria. Franco, F., & Maass, A. (1996b). Implicit vs. explicit strategies of outgroup discrimination: The role of intentional control in biased language use and reward allocation. Journal of Language and Social Psychology, 15, 335-359. Franco, F. M., & Maass, A. (in press). Intentional control over prejudice: When the choice of the measure matters. European Journal of Social Psychology. Funder, D. C., & Dobroth, K. M. (1987). Differences between traits: Properties associated with interjudge agreement. Journal of Personality and Social Psychology, 52, 409-418. Giles, H., & Johnson, P. (1987). Ethnolinguistic identity theory: A social psychological approach to language maintenance. International Journal of the Sociology of Language, 68, 69-99.
LINGUISTIC INTERGROUP BIAS
119
Graumann, C. F., & Wintermantel, M. (1989). Discriminatory speech acts. In D. Bar-Tal, C. F. Graumann, A. W. Kruglanski, & W. Stroebe (Eds.), Stereotyping and prejudice: Changing conceptions. (pp. 183-204). New York: Springer Verlag. Guerin, B. (1994). Gender bias in the abstractness of verbs and adjectives. Journal of Social Psychology, 134, 421-428. Hamilton, D. L.. Gibbons, P. A,. Stroessner, S. J., & Sherman, J. W. (1992). Language, intergroup relations and stereotypes. In G. R. Semin & K. Fiedler (Eds.), Language, interaction and social cognition (pp. 102- 128) London: Sage. Hamilton. D. L., & Sherman, S. J. (1996). Perceiving persons and groups. Psychological Review, 103, 336-355. Harris, S. E. (1993). The language of bias: The effect of social categorization on the linguistic intergroup bias. Unpublished master's thesis, Michigan State University, East Lansing, MI. Howard, J., & Rothbart. M. (1980). Social categorization and memory for ingroup and outgroup behavior. Journal of Personality and Social Psychology, 38, 301-310. Karpinski. A,, & von Hippel, W. (1996). The role of the linguisticintergroup bias in expectancymaintenance. Social Cognition, 14, 141-163. Lazzarato, G. (1989). L'uso del linguaggio nelle situazioni intergriippo: Applicazione del modello delle categorie linguistiche. (Language use in intergroup situations: Applying the linguistic category model]. Unpublished master's thesis. University of Padua, Italy. Livesley, W. J., & Bromley. D. B. (1973). Person perception in children and adolescence. London: Wiley. Loda, F. (1995-96). Contatto intergruppi e cambiamento degli stereotipii Una ricerca sul campo. [Intergroup contact and stereotype change: A field study]. Unpublished master's thesis, Padua University, Italy. Maass, A., & Arcuri, L. (1992). The role of language in the persistence of stereotypes. In G. Semin & K. Fiedler (Eds.), Language, interaction and social cognition (pp. 129-143). Newbury Park: Sage. Maass, A,, & Arcuri, L. (1996). Language and stereotyping. In N. Macrae, C. Stangor, & M. Hewstone (Eds.) The foundations of stereotypes and stereotyping, pp. 129-143. New York: Guilford Press. Maass, A., Ceccarelli, R., & Rudin. S. (1996). The linguistic intergroup bias: Evidence for ingroup-protective motivation. Journal of Personality and Social Psychology, 71,512-526. Maass, A., Corvino, P., & Arcuri, L. (1994). Linguistic intergroup bias and the mass media. Revue de Psychologie Sociale, 1, 31-43. Maass. A,, & Franco, F. (1997). The role of adjectives in reasoning, memory, and communication. Paper presented at the Conference on Social Cognition, Language, and Connectionism, Arnhem, 9-13 November 1997. Maass, A,, Giordana, C., & Fontana, F. (1990). The linguistic intergroup bias: Is it a function of social identity protection or stereotypic expectations? Unpublished manuscript, University of Padua, Italy. Maass, A., Milesi, A., Zabbini, S., & Stahlberg, D. (1995). The linguistic intergroup bias: Differential expectancies or ingroup protection'? Journal qfPersonality and Social Psychology, 68, 116-126. Maass, A,, Montalcini, F. & Biciotti. E. (1998). On the (dis-) confirmability of stereotypic attributes. European Journal of Social Psychology, 28, 383-402. Maass, A,, Salvi, D., Arcuri, L., & Semin, G. R. (1989). Language use in intergroup contexts: The linguistic intergroup bias. Journal of Personality and Social Psychology, 57,981-993. Macrae, C. N., Bodenhausen, G. V., Milne. A. B., & Jetten, J. (1994). Out of mind but back into sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67. 808-817.
120
ANNE MAASS
Mininni, G. (1982). Psicosemiotica. Bari: Adriatica Editrice. Mikula, G., & Graf, I. (1994). Fremdenfeindlichkeit und Sprachverhalten. Eine Anwendung des linguistischen Katerorienmodells auf das Phaenomen der Fremdenfeindlichkeit (Xenophobia and language behavior. An application of the linguistic category model to the phenomenon of xenophobia). Unpublished manuscript, University of Graz, Graz, Austria. Moulton, J., Robinson, G. M., & Elias, C. (1978). Sex bias in language use: Neutral pronouns that aren’t. American Psychologist, 33, 1032-1036. Ng, S. H. (1990). The androcentric coding of man and his in memory by language users. Journal of Experimental Social Psychology, 26, 455-464. Ng, S. H., & Chan, K. K. (1996). Biases in the description of various age groups: A linguistic category model analysis. Bulletin of the Hong Kong Psychological Society, 36/37, 5-20. Ng, S. H., & Tait, J. (1994). Biases of journalists’ linguistic representation of intergroup events. Unpublished manuscript, Victoria University of Wellington, N.Z. Palmer, K . (1997). Correlates of linguistic ingroup bias. Unpublished manuscript, University of Kent, UK. Rothbart, M., Evans, M., & Fulero, S. (1979). Recall for confirming events: Memory processes and the maintenance of social stereotypes. Journal of Experimental Social Psychology, 15, 343-355. Rothbart, M., & Park, B. (1986). On the confirmability and disconfirmability of trait concepts. Journal of Personality and Social Psychology, 50, 131-142. Rubini, M., & Semin, G. R. (1994). Language use in the context of congruent and incongruent in-group behaviors. British Journal of Social Psychology, 33, 355-362. Rusher, J. B. (1998). Prejudice and stereotyping in everyday communication. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 241-307). San Diego: Academic Press. Scalise, S . (1995). La formazione delle parole [The formation of words]. In L. Renzi, G. Salvi, & A. Cardinale (Eds.), Grande Grammatica Italiana di Consulrazione (Vol. 8, pp. 471-514). Bologna: Mulino. Schmid, J. (1995, June). Good shouldn’t always win-Sprachanalysen von WrestlingBerichten [Good shouldn’t always win-Language analyses of wrestling reports]. Paper presented at the Annual Meeting of the Fachgruppe Sozialpsychologie, Potsdam. Schmid, J., & Fiedler, K. (1996). The backbone of closing speeches: The impact ofprosecution versus defense language on juridical attributions. Unpublished manuscript, University of Heidelberg, Germany. Semin, G. R., & Fiedler, K. (1988). The cognitive functions of linguistic categories in describing persons: Social cognition and language. Journal of Personality and Social Psychology, 54, 558-568. Semin, G. R., & Fiedler, K. (1991). The linguistic category model, its bases, applications and range. In W. Stroebe & M. Hewstone (Eds.), European review of socialpsychology (Vol. 2, pp. 1-50). Chichester: John Wiley. Semin, G. R., & Fiedler, K. (1992). The inferential properties of interpersonal verbs. In G. Semin & K. Fiedler (Eds.), Language, interaction and social cognition (pp. SS-78). Newbury Park Sage. Semin, G. R., & Poot, C. J. de (1997). You might regret it if you don’t notice how a question is worded! Journal of Personality and Social Psychology, 73, 472-480. Semin, G. R., Rubini, M., & Fiedler, K. (199.5). The answer is in the question: The effect of verb causality on locus of explanation. Personality and Social Psychology Bulletin, 21, 834-841. Tajfel, H., & Turner, J. C. (1987). Social identity theory of intergroup relations. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago: Nelson Hall.
LINGUISTIC INTERGROUP BIAS
121
Taylor, D. M., & Moghaddam, F. M. (1987). Social identity theory. In D. M. Taylor & F. M. Moghaddam, Theories of intergroup relations: International Social Psychological Perspectives (pp. 59-84). New York: Praeger. Tans, T. W. (in press). Describing behaviors of self and others: Self-enhancement and language abstraction level. European Journal of Social Psychology. Vanbeselaere, N. (1991). The different effects of simple and crossed categorizations: A result of the categorization process or of differentia! category salience? In W. Stroebe & M. Hewstone (Eds.), European Review of Social Psychology. Chichester: Wiley. Vanbeselaere, N. (1996). The impact of differentially valued overlapping categorizations upon the differentiation between positively, negatively, and neutrally evaluated social groups. European Journal of Social Psychology, 26, 75-95. Van Dijk, T. A. (1984). Prejudice and discourse. An analysis of ethnic prejudice in cognition and conversation. Amsterdam: Benjamins. Van Dijk, T. A. (1987). Communicating racism. Ethnic prejudice in thought and talk. Newbury Park. CA: Sage. Van Dijk. T. A. (1988). New analysis. Case studies of international and national news in the press. Hilldale, NJ: Erlbaurn. von Hippel. W., Sekaquaptewa, D.. & Vargas, P. (1995). On the role of encoding processes in stereotype maintenance. In M. P. Zanna (Ed.), Advances in experimental socialpsychology (Vol. 27. pp. 177-254). San Diego, CA: Academic Press. von Hippel, W., Sekaqueptewa, D.. & Vargas. P. (1997). The linguistic intergroup bias as an implicit indicator of prejudice. Journal of Experimental Social Psychology, 33, 490-509. von Hippel, W., Vargas, P., & Sekaquaptewa, D. (1994). Prejudice as process. Unpublished manuscript, Ohio State University, Columbus, OH. Webster, D. M., Kruglanski, A. W.. & Pattison, D. A. (1997). Motivated language use in intergroup contexts: Need-for-closure effects on the linguistic intergroup bias. Journal of Personality and Social Psychology, 72, 1122-1 131. Werkman, W. M., & Sernin, G. R. (1997). Children’s competencies in explaining interpersonal states and actions Unpublished manuscript, Free University. Amsterdam, The Netherlands. Werkman. W. M., Wigboldus, D. H. J., & Semin. G . R. (1999). Children’s communication of the Linguistic Intergroup Bias and its impact upon cognitive inferences. European Journal of Social Psychology, 29, 95-104. Wigboldus, D. (1998). Stereotyping, language, and communication. Unpublished doctoral dissertation. Free University, Amsterdam, The Netherlands.
This Page Intentionally Left Blank
RELATIONSHIPS FROM THE PAST IN THE PRESENT SIGNIFICANTOTHER REPRESENTATIONS AND TRANSFERENCE IN INTERPERSONAL LIFE Serena Chen Susan M. Andersen
Close relationships have long captured the interest of both the social psychologist and layperson. Among social psychologists, this interest has flourished in the last few decades, with considerable emphasis being given to processes and phenomena in present-day close relationships (for reviews, see Berscheid, 1994; Clark & Reis, 1988). Also common to much recent empirical work on close relationships is a focus on social-cognitive structures and processes, as reflected in recent volumes devoted entirely to the growing cross-fertilization between social cognition and close relationships research (Fletcher & Fincham, 1991; Fletcher & Fitness, 1996). In this chapter, we present our approach to the study of close relationships, an approach grounded in social-cognitive theory and research, but departing somewhat from the domain of present-day close relationships by focusing specifically on past relationships with significant others and their role in shaping current social relations. We aim not only to demonstrate the powerful and multifaceted effects that prior experiences with significant others exert on present-day interpersonal life, but also to precisely articulate the social-cognitive principles that govern the manner in which this influence of the past on the present occurs.
I. The Transference Phenomenon
In the course of everyday social life, it is not uncommon for a newly encountered person to remind a perceiver of someone highly significant whom he or she knows or has known in the past. Intuition suggests that ADVANCES IN EXPERIMENTAL SOCIAL PSYCHOLOGY VOL 31
123
Copyright 0 1999 by Acddcmlc P r m All right5 of reproduction in any form rr\erved 0065 2601199 $30 00
124
SERENA CHEN AND SUSAN M. ANDERSEN
such remindings influence perceptions of the new person. Indeed, as a result of such remindings, feelings, expectations, and behavioral responses experienced in relation to the new person may come to reflect, at least in part, past experiences with the particular significant other. Consider, for example, a man whose new neighbor reminds him of his estranged father and who thus bristles with anger upon passing this neighbor on the street. Or consider the child who anticipates harsh discipline from a new authority figure because she has been disciplined harshly by a parent. The general notion that patterns of interpersonal relating, learned in prior relationships, may resurface in later social relations is shared by a wide range of psychological theories (e.g., Baldwin, 1992; Bartholomew & Horowitz, 1991; Bowlby, 1969,1973,1980; Bugental, 1992;Feeney & Noller, 1990; Freud, 1912/1958; Greenberg & Mitchell, 1983; Guidano & Liotti, 1983; Hazan & Shaver, 1987; Horney, 1939; Horowitz, 1991; Kelly, 1955; Luborsky & Crits-Christoph, 1990; Rogers, 1951; Safran & Segal, 1990; Shaver & Rubenstein, 1980; Sullivan, 1953; Wachtel, 1981). In our work, we focus on the clinical concept of transference (e.g., Freud, 191211958), conceptualized in social-cognitive terms. Transference refers to the very phenomenon whereby aspects of past relationships with significant individuals reemerge in present interpersonal life (see also Sullivan, 1940, 1953). Indeed, basic to any definition of transference is the experience of assumptions and feelings about a significant other in relation to some other person, generally toward an analyst in psychoanalysis (Luborsky & Crits-Christoph, 1990; see also Ehrenreich, 1989), but also potentially to any individual encountered in daily social interactions. Although the transference concept has existed for nearly a century, little experimental work has been done to examine whether or not transference actually occurs, leaving scant knowledge concerning its underlying mechanisms. Our central argument is that the activation and application of a stored, mental representation of a significant other to a new person constitute the fundamental processes by which transference occurs. Thus, when something about a new person reminds a perceiver of a significant other, the stored representation of the significant individual is likely to come to mind, consciously or nonconsciously, and in turn, to be applied to the new person, coloring the perceiver’s interpretations of, and responses toward, the person. We now turn to our social-cognitive model of transference, and the program of research emerging from it (Andersen & Baum, 1994; Andersen & Cole, 1990; Andersen, Glassman, Chen, & Cole, 1995; Andersen, Reznik, & Manzella, 1996; Baum & Andersen, in press; Chen, 1997; Hinkley & Andersen, 1996). We begin by articulating the basic assumptions of the model. We then describe the experimental paradigm used in our re-
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
125
search and initial evidence supporting the model. Next, we present a series of empirical efforts designed not only to demonstrate the many forms in which past experiences with significant others may reappear in present social relations, but also to provide evidence for the social-cognitive mechanisms that guide the transference process. Finally, we locate our work in the broader literature on close relationships by highlighting common theoretical and empirical ground.
11. The Social-Cognitive Model of Transference
Fundamental to the social-cognitive model of transference is the assumption that mental representations of significant others, developed in numerous encounters with these individuals, are stored in memory, are highly affectively laden, and are connected with the self in memory via selfhignificant-other or relational linkages. As indicated, the model proposes that transference can be understood in terms of the activation and application of these stored representations to new others in social perception. In this section, we elaborate on these basic contentions of the model, describing both its social-cognitive and clinical underpinnings.
A. SOCIAL-COGNITIVE ORIGINS
To understand how transference occurs by means of the activation and application of a stored, mental representation of a significant other to a new person, one needs to begin by considering the different types of social knowledge that are thought to be represented in memory. Social constructs have most often been studied as types of persons, such as a “nerd” o r “adventurous” type (e.g., Higgins & King, 1981). As with roles and stereotypes, these constructs represent classes of people, typically designated by trait-adjective terms, which can then operate as points of comparison in assessing new others (e.g., Andersen & Klatzky, 1987; Schneider & Blankmeyer, 1983). A “proper” construct, as in a “proper” name for a specific person (e.g., Higgins & King, 1981), can also operate as do other social constructs (e.g., Andersen & Glassman, 1996; S.T. Fiske, 1982; Smith & Zarate, 1990,1992). Such constructs, representing individual persons, have been referred to as exemplars, and considerable research has shown that they are often also used as points of comparison for judging new individuals (e.g., Judd &
126
SERENA CHEN AND SUSAN M. ANDERSEN
Park, 1988; Smith & Zarate, 1990, 1992; see also Linville & Fischer, 1993; Mullen, 1991; Park & Hastie, 1987). Building on this basic notion, we define transference in terms of the activation of a perceiver’s representation of a significant other-or exemplar of this important other-and its subsequent application to a new person. As a result, encounters with this new person may be experienced nearly as if he or she were the significant other-that is, in ways that are analogous to past experiences with the significant other. Responding by analogy has been shown to occur in a variety of domains, including ones not involving significant-other representations (e.g., Gentner & Markman, 1997;Gilovich, 1981;Holyoak & Thagard, 1995;Read & Cesa, 1991; Seifert, McKoon, Abelson, & Ratcliff, 1986; see also Lewicki, 1985, 1986; Logan, 1988; Ross & Bradshaw, 1994; Ross, Perkins, & Tenpenny, 1990; Schank, 1982). B. SOURCES O F TRIGGERING TRANSFERENCE Beyond simply identifying the activation and application of significantother representations to new others as the fundamental processes underlying transference, our model maintains that the same basic principles thought to govern other social constructs (e.g., traits, roles, and stereotypes) similarly guide the activation and use of significant-other representations in social perception. In particular, in accord with prevailing social-cognitive models of construct activation and use (Higgins, 1989a, 1996; see also Bargh, 1997; Sedikides & Skowronski, 1990,1991; Wyer & Srull, 1986), the model delineates both chronic and transient sources of triggering significant-other representations. Chronic sources of construct activation, or the chronic accessibility of a construct, are thought to arise from a construct’s sheer frequency of use (e.g., Higgins, 1989a, 1990, 1996; Higgins & Brendl, 1995; Higgins & King, 1981; Higgins, King, & Mavin, 1982). The chronic accessibility of a construct is of considerable significance in social perception in that it refers to the chronic readiness for the construct to be brought to bear on perceiving and interpreting others-even with minimal or no other sources of activation present. By definition, the degree to which any given construct is chronically accessible depends on factors contributing to the frequency with which the construct comes to mind. Significant-other representations are highly familiar, frequently thought about, and of relevance to the self (Andersen & Glassman, 1996; Andersen, Glassman, & Gold, 1998; see also Prentice, 1990)-factors that imply that they are likely to be chronically accessible.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
127
Indeed, the chronic accessibility of significant-other representations has been documented in research to be described below (Andersen et al., 1995), suggesting the ubiquity of the transference phenomenon in everyday social relations. Of course, beyond chronic sources of accessibility, transient cues in the environment can also contribute to the activation and application of stored constructs (e.g., Higgins, 1989a, 1996). In particular, a large literature on priming has examined the impact of a construct’s recent activation on the extent to which it is subsequently used in social perception (e.g., Bargh, Bond, Lombardi, & Tota, 1986; Higgins, Rholes, & Jones, 1977; Srull & Wyer, 1979). This research has typically shown that the activation of a construct just before a stimulus person is encountered, or the priming of the construct, is likely to result in the assimilation of the person into the construct-that is, the person is likely to be interpreted in construct-derived ways in a subsequent impression-formation task (e.g., Higgins et a]., 1977). Although contrast effects, whereby the impression of the target is contrasted away from the activated construct, are observed under certain conditions (e.g., Higgins & Stangor, 1988; Lombardi, Higgins, & Bargh, 1987; Manis, Paskewitz, & Cotler, 1986; Martin, 1986; Martin, Seta, Crelia, 1990; Petty & Wegener, 1993; Schwarz & Bless, 1991), assimilation effects are quite common. In accord with social-cognitive work on priming, our model assumes that such transient sources of construct activation may contribute to the triggering of transference. That is, the recent activation, or priming, of a significant-other representation ought to increase the subsequent likelihood that the representation will be activated and applied to a newly encountered person. Indeed, research on transference supports this, as will be detailed later (Andersen et al., 1995, Study 1). Most work on transient priming sources of construct activation has tended to ignore the potential for a to-be-interpreted stimulus itself to contribute to the likelihood of construct activation (for exceptions, see Banaji, Hardin, & Rothman, 1993; Bruner, 1957; Hardin & Rothman, 1997; Higgins & Brendl, 1995; Higgins et a]., 1977; Srull & Wyer, 1979,1980). Central to our model of transference is the notion that a target person’s resemblance to a significant other increases the likelihood that the representation of this individual will be activated and applied to the new person. That is, the “match” between cues in the target person and stored knowledge about a significant other-termed applicability (Higgins, 1989a, 1996)-is another potential transient source of triggering transference. Considerable research converges on the notion that the activation and application of a stored construct to a stimulus are particularly likely to occur when there is some perceived “match” between the stimulus and the construct (e.g., Tversky, 1977;
128
SERENA CHEN AND SUSAN M. ANDERSEN
Tversky & Gati, 1978; see also Cantor & Mischel, 1979; Hardin & Rothman, 1997; Higgins, 1989a, 1996; Nosofsky, 1986; Rosch, 1978; Schneider & Blankmeyer, 1983; Sedikides & Skowronski, 1990, 1991; Smith & Zarate, 1990). In our view, applicability-based triggering in the form of a newly encountered person’s “match” or resemblance to a significant other is especially relevant to transference in that the phenomenon may be particularly likely to occur in the context of actual face-to-face encounters with new others. Thus, we suggest that both chronic and applicability sources of activation are typical contributors to the activation and use of significant-other representations, and much research on transference supports this claim, as will be seen below (Andersen et al., 1995, Study 2; see also Andersen & Glassman, 1996; Andersen, Reznik, & Chen, 1997; Chen, Andersen, & Hinkley, in press).’ Overall, our model of transference maintains that chronic sources, either alone or in combination with transient priming and applicability sources of activation contribute to the triggering of transference (see also Bargh et al., 1986; Higgins, 1996). That is, although significant-other representations should have a chronic readiness to be used in social perception, they should also be particularly likely to be activated and applied to new others when transient sources of construct activation are also present. We discuss issues pertaining to the triggering of transference in greater depth in later sections. C. MANIFESTATIONS OF TRANSFERENCE In our model, transference is seen as a multifaceted phenomenon, bringing a wide range of past experiences in relationships with significant others into the present. Namely, we contend that its occurrence manifests itself
’
In defining transference as reflecting the activation and application of a significant-other representation to a new person, one question that might arise is whether or not perceivers are consciously aware of these processes. In particular, are perceivers aware that the new person reminds them of a significant other? We would argue that transference may occur whether perceivers are or are not consciously aware of the resemblance of a new person to a significant other. Indeed, recent research examining the subliminal activation of significantother representations is consistent with this (Glassman & Andersen, in press). On the other hand, social-cognitive work on assimilation and contrast effects would suggest that awareness is a factor that might lead a perceiver to try to avoid interpreting a new person in terms of a significant-other representation (for review, see Higgins, 1996). Such inhibition of transference is likely to occur to the extent that the perceiver’s awareness of its possible occurrence is coupled with the assessment that using the significant-other representation to interpret the new person would be inappropriate or irrelevant (for a discussion of related issues, see Andersen et al., 1995; Chen, Andersen, & Hinkley, in press).
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
129
not only in terms of perceptions and interpretations of, and responses toward, a new person to whom an activated significant-other representation has been applied, but also in terms of the experience of the self in relations with the new other. In more specific terms, we assume that the activation and application of a significant-other representation to a new person lead perceivers to “go beyond the information given” (Bruner, 1957) about the person by inferentially filling in the blanks about the person on the basis of the representation. That is, upon such activation and application, perceivers’ stored inferences about the significant other are used to make inferences about the new person at encoding (i.e., while learning about the person). Subsequently, perceivers should be more likely to remember the new person in terms of the significant other, believing that they learned things about a new person that they had actually inferred at encoding (for related work, see Johnson, Hastroudi, & Lindsay, 1993; Johnson & Raye, 1981). Stated somewhat differently, we argue that significant-other representations have the inferential power to simplify what is known about a new person by enabling inferences about him or her, just as a schema or prototype does (see also Andersen & Klatzky, 1987; Andersen, Klatzky, & Murray, 1990; Bargh et al., 1986; Cantor & Mischel, 1977,1979; Hamilton, 1979; Higgins, Rholes, & Jones, 1977; Taylor, 1981). In short, the emergence of representation-derived inferences based on an activated significant-other representation serves as a basic index of transference in our research on the phenomenon (see also Andersen & Glassman, 1996). Our model of transference also argues that the activation and application of a significant-other representation to a new person should lead to evaluations of the person that are derived from the representation. More specifically, drawing on S.T. Fiske’s model of schema-triggered affect (e.g., S. T. Fiske, 1982), we contend that “affective” or “evaluative” information linked to the representation in memory is activated upon the activation of the representation, leading the new person to be evaluated in these terms (e.g., Andersen & Baum, 1994)-analogous to what has been shown to occur upon activation of a social category or stereotype (S. T. Fiske & Pavelchak, 1986; Pavelchak, 1989). Thus, we interpret the emergence of representation-derived evaluation of a new person based on an activated significant-other representation as another basic index of transference, along with representation-derived inference and memory (Andersen et al., 1997; Andersen & Glassman, 1996). Finally, our model assumes that representations of the self and significant others are linked in memory (see also Baldwin, 1992; Bugental, 1992; Ogilvie & Ashmore, 1991). We maintain that these selfhignificant-other linkages represent the relationship between self and other, and thus embody motiva-
130
SERENA CHEN AND SUSAN M. ANDERSEN
tions, affect, and roles experienced in relation to the other, as well as interpersonal expectancies involving the other (Andersen et al., 1996; Baum & Andersen, in press; see also Andersen et al., 1997). As such, the activation and application of significant-other representations to new others in transference should lead not only to interpretive and evaluative effects, but also to motivational and affective consequences that reflect aspects of past relationships with significant others. Beyond this, stored seiflsignificantother linkages imply that when a significant-other representation is activated and applied to a new person, aspects of the self that are or have been experienced in the relationship with the relevant significant other are also activated, leading to changes in the experience of the self in relations with the new other (e.g., Andersen et al., 1997; Hinkley & Andersen, 1996; see also Baldwin & Holmes, 1987; Baldwin, Carrell, & Lopez, 1990).
D. CLINICAL ORIGINS
The clinical origins of our model lie in the work of both Freud (1912/ 1958) and Sullivan (1940,1953). Freud developed the transference concept and assumed it to have unconscious psychosexual content and motivation that emerge in therapy as a defense mechanism. The notion that a “patient” superimposes childhood fantasies, conflicts, and wishes about a parent onto an “analyst,” weaving “the figure of the physician into one of the ‘series’ already constructed in his mind” (Freud, 1912/1963, p. 107; cited in Andersen & Glassman, 1996) is the essence of the concept. Although we do not endorse Freud’s psychosexual-conflict model (which has implications for the content of transference), his assumption about its defensive roots, nor his nearly exclusive focus on parental figures, we clearly concur that responses to a significant other may be experienced with new people. Particularly relevant to our model is the interpersonal view of transference proposed by Sullivan (1940, 1953) in which he termed the process parataxic distortion. His view assumes that children form “personifications” of themselves and of significant others, as well as “dynamisms,” or the dynamics that characterize the typical interplay between oneself and the other (see also Greenberg & Mitchell, 1983; Mullahy, 1970). He saw both as being learned in the context of “real” interpersonal relations experienced in one’s family of origin in relation to particular significant individuals, and we would agree. However, Sullivan did not, as we do, postulate the existence of mental representations, instead viewing personifications and dynamisms
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
131
in terms of energy and energy transformations.* Nonetheless, we see personifications and dynamisms as readily interpretable in terms of mental representations designating the self and others, and the relational patterns linking the self with these others (see also Andersen & Glassman, 1996). Sullivan also made assumptions about motivation, arguing that certain basic human needs guide the development of the self’s relations with others (see also Andersen et al., 1997). In his model, he discusses the need for satisfaction, defining it in terms of the need for warm or “integrative” encounters with others in which mutual tenderness is expressed. His “tenderness theorem” posits that encounters with significant others that are tender and caring are essential for satisfaction-even when one is pursuing seemingly unrelated tasks, such as developing one’s own skills and talents. In doing the latter, one is fulfilling another aspect of the need for satisfaction involving the need to express one’s own emotions and talents, and to develop one’s own competencies. In his model, the aim is to do the latter without disrupting or damaging the emotional connection with the other. Beyond the need for satisfaction, a need for security is also assumed-that is, the need to feel and be safe (Greenberg & Mitchell, 1983). In Sullivan’s view, the experience of tenderness with significant others or its alternatives-experiencing the other as cold, distant, or distressed-impacts on one’s sense of security and safety. Hence, the individual is motivated to keep insecure feelings about significant others at bay in the context of needing to feel tender, connected responses from them. In short, these motivations for satisfaction and security define a person’s individualized pattern of pursuing satisfaction and security with particular significant others, and are thus represented in personifications and dynamisms. In Sullivan’s version of transference, then, the process invokes not only personifications and dynamisms, but also these motivations. Our model of transference also has motivational underpinnings, alongside its social-cognitive bases. Thus, we concur with Sullivan that basic motivations pertaining to satisfaction and security-among others such as a basic need for human connection-are fundamental to the development of, from our perspective, representations of the self, significant others, and the self
* In terms of distinctions, Sullivan proposed that actual interpersonal relations experienced with a significant other in a family is what leads to patterns of responding learned with significant others. whereas Freud assumed ”reality” was less relevant than unconscious fantasy about interactions with others (Greenberg & Mitchell, 1983). driven by psychosexual urges and not by affiliative or mastery needs. The latter needs are of course fundamental to Sullivan’s ideas about the need for satisfaction, as described in this section. For Sullivan, then, the content of self and other knowledge is not necessarily psychosexual, as it is in Freud’s theory, and the id. ego, superego structures of Freud have no place. Instead, Freud’s tripartite structure of the mind is replaced with self-representations and object representations.
132
SERENA CHEN AND SUSAN M. ANDERSEN
in relation to these others (for a more detailed discussion, see Andersen et al., 1997). As a final note, despite the clinical relevance of our model, we contend that transference can occur not only in therapy, but also in daily life (see also Singer, 1988; Wachtel, 1981; Westen, 1988)-a claim also made by Freud and Sullivan, although they both focused on the therapy situation (see Luborsky & Crits-Christoph, 1990). Indeed, although some psychodynamic theorists focus on transference exclusively in the client-therapist relationship, and also emphasize the particular importance of parental representations (see Ehrenreich, 1989;Luborsky & Crits-Christoph, 1990), we consider transference ubiquitous in daily life, and regard its therapeutic implications in broad-based terms as well (see Andersen & Berk, 1998). Thus, we assume that it may occur on the basis of representations of various significant others-such as a sibling, best friend, spouse, or other romantic partner-as well as on the basis of parental representations. In sum, we assume that transference can be understood in basic social-cognitive terms, that it can be measured empirically, and that it has multifaceted effects that result in the reemergence of relationships from the past in the present. We turn now to the program of research that has examined the basic assumptions of the social-cognitive model of transference. We begin by describing the experimental paradigm used in this work and initial evidence for the model, and continue on through several bodies of empirical evidence on the transference phenomenon pertaining to its affective and motivational consequences, as well as to the underlying social-cognitive principles that govern the process. We then consider what is known from research on transference in terms of the broader literature on close relationships, highlighting ways in which our work and this broader body of research can be mutually informed. 111. The Basic Transference Paradigm
Given the complexity of our research methods, we discuss them in some depth before presenting the evidence for the social-cognitive model of transference. The experimental paradigm that we describe in this section serves as the backbone of all the research presented in later sections. A. COMBINING IDIOGRAPHIC AND NOMOTHETIC RESEARCH METHODS Essential to our view of the role of significant-other representations in transference is that these representations are personally meaningful, thus
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
133
requiring idiographic research methods (Allport, 1937; Kelly, 1955) to assess the idiosyncratic content of people's representations of their significant others. To understand the basic processes underlying transference, however, it is also necessary to use a nornorhetic experimental design in which idiographic content is used. Thus, we use a combined idiographic-nomothetic methodology that allows us to track normative processes generalizing across people, while nonetheless making use of idiographic information. (For pertinent reviews of idiographic methods, see Higgins, 1987,1990; Pelham, 1993; Pervin, 1976.) An outline of our basic experimental paradigm appears in Table I. In the pretest session, participants are asked to idiographically generate descriptive sentences about a significant other, and in some cases, about various control representations as well. In a subsequent, experimental session, usually held at least 2 weeks later, the learning and test phases of the
TABLE I THEBASKTRANSFERENCE PARADIGM" A. Pretest Session 1. Participants name a significant other and possibly other people or social categories to serve as control representations. 2. Participants generate a series of sentences (usually 14) to characterize each person/ category named, and then rank order these sentences in terms of how descriptive they are of each personkategory. 3. For each personkategory, participants select from a list of adjectives those they view as neither descriptive nor counter-descriptive of the personkategory-that is, those that are essentially irrelevant to the persodcategory. B. Experimental Session 1. Typically, at least 2 weeks after the pretest session, participants engage in the learning phase in which they are presented with descriptive sentences about one or more new target persons. 2. In the learning phase, one target person is always characterized by some of the descriptive sentences participants themselves had generated earlier to describe their significant other (as well as by some irrelevant filler sentences). The other targets, when present, correspond to the control representations. 3. For each target, participants complete a recognition-memory test in which they rate their confidence that they saw each of a series of descriptive sentences in the learning phase. Some of these sentences were actually presented about the target, and some were not presented but were derived from the corresponding representation. For the target corresponding to participants' significant other. the confidence ratings given for the representation-derived sentences that were not actually presented in the learning phase serve as a basic index of transference in the form of representation-derived inference and memory about the target based on an activated significant-other representation. Representationderived evaluation of the target also serves as a basic index of transference. Adapted from Andersen and Glassman (1996).
134
SERENA CHEN AND SUSAN M. ANDERSEN
paradigm occur. In the learning phase, participants encounter descriptive sentences about one or more new target persons. Some of these target descriptors were derived from ones they had idiographically generated in the earlier session, and others are irrelevant filler descriptors. We construe descriptors in a target person as an analog for cues perceivers would encounter on-line in an actual social encounter. One target always resembles the significant other participants had described earlier. That is, descriptive cues in this target “match” or overlap with some of the knowledge participants have stored about a significant other. Thus, applicability-based triggering of significant-other representations is operationalized in the form of significant-other resemblance in the new target person. Said differently, significant-other resemblance in the target constitutes applicability sources of activation for the relevant significant-other representation in the learning phase of our paradigm. The standard significant-other-resemblance condition is compared to one or several control conditions in which no-signijicant-otherresemblance exists in the target person(s). For example, in a standard control condition, the target is characterized in the learning phase by descriptors generated by a “yoked” participant to describe his or her significant other. Typically, each participant in the significant-other-resemblance condition is yoked with one participant in the no-resemblance condition on a one-to-one basis. This ensures that the participant in the latter condition, the yoked-participant control condition, sees exactly the same descriptors as the participant in the significant-other-resemblance condition, thereby controlling for the exact content of the descriptors about the new target. The extent to which participants’ own significant-other representations are activated and applied should be greater in the standard significant-other-resemblance condition relative to the no-significant-other-resemblance condition(s). After learning about the target person(s), participants take part in the test phase of the experimental session. Specifically, they complete a variety of dependent measures, including a standard recognition-memory test composed of representation-derived descriptive sentences (as well as some filler sentences) that either were or were not presented about each target in the learning phase. Table I1 depicts the stimuli used in a typical learning phase to describe the target(s), along with stimuli used as test items in the recognition-memory test to assess representation-derived inference and memory for both the standard significant-other-resemblance condition and the yoked-participant control condition in which no significant-other resemblance exists in the target person. For each target, participants’ recognition-memory confidence ratings for representation-derived descriptors that were not actually presented in the learning phase serve as a measure of the nomothetic process of
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
135
TABLE I1 TARGET STIMULI A N D RECOGNITION-MEMORY TESTITEMSUSEDIN THE EXPERIMENTAL SESSION Target condition Resembles the participant’s own significant other
Resembles a yoked-participant’s significant other
Target descriptors used in the learning phase
Some of the pnrticipnnr’s own idiographically generated significant-other descriptors, along with some irrelevant filler descriptors
Some of a yoked participant’s idiographically generated significant-other descriptors. along with some irrelevant filler descriptors
Items used in the recognitionmemory test
Some of the parkipant’s own idiographically generated significant-other descriptors that were and were not presented in the learning phase, and some irrelevant filler descriptors that were and were not presented in the learning phase
Some of a yoked participant’s idiographically generated significant-other descriptors that were and were not presented in the learning phase, and some irrelevant filler descriptors that were and were not presented in the learning phase
“going beyond the information given” inferentially about the target in representation-derived ways at encoding (see Cantor & Mischel, 1977). That is, higher confidence ratings in the significant-other-resemblance condition relative to the control condition(s) serve as a basic index of transference in the form of representation-derived inference and memory. The emergence of evaluative responses toward the target in the significant-other-resemblance condition that reflect the overall evaluation or affect linked with the relevant representation is also taken as basic evidence for the transference phenomenon. Such representation-derived evaluation should not occur in the no-significant-other-resemblance condition(s). Hence, as indicated earlier, evaluation ratings serve as another basic index of transference.
B. INITIAL EVIDENCE: REPRESENTATION-DERIVED INFERENCE/MEMORY AND EVALUATION The first demonstration of transference examined representation-derived inference and memory based on an activated significant-other representation (Andersen & Cole, 1990, Study 3). In the pretest session of the study, participants identified a significant other, a nonsignificant other, a stereo-
136
SERENA CHEN AND SUSAN M. ANDERSEN
type label (e.g., “redneck”), and a trait label (e.g., “liberal”). They then generated 14 sentences to describe each person or social category, and rank-ordered these sentences in terms of their descriptive importance. Next, they selected from a list of adjectives those they considered “irrelevant” to each representation. These adjectives were used as filler target descriptors in the learning phase, and as filler test items in the test phase of the experimental session. In the learning phase, each participant was presented with descriptive sentences about four target persons via microcomputer-one of whom resembled hidher own significant other. The other targets served as nosignificant-other-resemblance control conditions and resembled, respectively, the participant’s previously identified nonsignificant other, stereotype, and trait. The sentences describing each target were derived from ones participants had generated in the pretest session. More specifically, each of four fictional-character names (e.g., “Terry”) was paired with a fixed proportion of participants’ identically ranked sentence-predicates for each representation (e.g., “. . . has strong views” or “. . . makes others feel inadequate”), along with some irrelevant filler adjectives (each preceded by the verb “is”) in an individual random order. Each target resembled one of the participant’s own representations with an identical degree of applicability-that is, overlap between the descriptors previously generated to characterize each representation and those learned about the corresponding target. After the learning phase, participants completed a standard recognitionmemory test about each target used to assess representation-derived inference and memory. Resemblance in each target to the corresponding representation should increase the extent to which the representation is activated and applied to the target, as would be evidenced in higher recognitionmemory confidence ratings for representation-derived descriptors that were not actually presented about the target. As predicted, participants gave higher confidence ratings when the target resembled their own significant other relative to the no-significant-other-resemblance control targets. This evidence for transference in the form of representation-derived inference and memory based on an activated significant-other representation has been found in numerous studies, using designs varying in the number of target persons about whom participants learn, the control conditions used, and the manner in which participants “encounter” the descriptors about the target persons (e.g., Andersen & Baum, 1994; Andersen et al., 1995,1996; Baum & Andersen, in press; Chen et al., in press; Hinkley & Andersen, 1996). Indeed, recent studies have also demonstrated how robust this basic inference and memory effect is in that it persists and may even be exacerbated over time (Glassman & Andersen, in press).
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
137
As indicated, beyond such representation-derived inference and memory, evaluative responses of a new person based on an activated significantother representation also serve as a basic index of transference. In research examining such representation-derived evaluation in transference, the overall evaluative tone of participants’ significant-other representations is determined at pretest-specifically, participants are asked to name and describe positively and negatively evaluated significant others (e.g., Andersen & Baum, 1994; Andersen et al., 1996). Representation-derived evaluation in transference is then examined by manipulating significant-other resemblance and the overall evaluative tone of the significant-other representation in the experimental session. The initial research examining representation-derived evaluation in transference used a between-subjects design in which participants learned about a new target person in a “real” social context in which the person was allegedly seated next door and a future interaction was imminent (Andersen & Baum, 1994). The new person resembled either participants’ own positively or negatively toned significant other or a yoked participant’s positively or negatively toned significant other. Perfect one-to-one yoking across conditions ensured that the stimuli encountered by participants learning about a target resembling their own significant other were identical to those encountered by participants learning about a yoked participant’s significant other. The use of identical descriptors across conditions rules out differences based on the content of the stimuli, as well as the notion that something distinctive about the descriptors of any significant other, even a yoked participant’s, could somehow account for the results. Because the target resembled either the participant’s own or a yoked participant’s positively or negatively toned significant other, the extent to which evaluation of this target was derived from the overall evaluative tone of the relevant significant-other representation could be assessed. As stated in an earlier section, in line with the theory of schema-triggered affect (e.g., S. T. Fiske, 1982), we argue that when a significant-other representation is activated and applied to a new person, the evaluation linked to this representation should be attached to the new other. We refer to this evaluation prediction as “representation-derived’’ instead of “schema-triggered” simply because it takes no stand on the internal architecture or schematicity of significant-other representations, and thus is a language we prefer-even though predictions based on the theory of schema-triggered affect are the predictions we make. As in the prior study (Andersen & Cole, 1990, Study 3), the results showed greater representation-derived inference and memory for the target resembling the participant’s own rather than a yoked participant’s signifi-
138
SERENA CHEN AND SUSAN M. ANDERSEN
cant other, as portrayed in Figure 1. These data verify evidence for transference using this index. Importantly, this effect occurred whether the target resembled a positively or negatively toned significant other, thereby indicating that transference occurs on the basis of both positively and negatively toned significant-other representations. Of still greater importance was evidence indicating that representationderived evaluation does in fact occur as part of the transference process. That is, participants evaluated the target person more favorably when he/ she resembled their own positively versus negatively toned significant other. This effect did not occur when the target resembled a yoked participant's positively versus negatively toned significant other. These findings are depicted in Figure 2. In presenting these results, it is worth noting that participants had been asked to generate both positive and negative descriptors about their positively and negatively toned significant others in the pretest session so that the targets in both significant-other-resemblance conditions could be described with an equal number of positive and negative descriptors. As such, participants' evaluative responses toward the target resembling their own significant other were apparently derived from the overall evaluative tone of the relevant significant-otherrepresentation, rather than from the evaluative tone of the individual target descriptors. In sum, initial research demonstrating transference has suggested that representation-derived inference and memory, as well as representation2.2 Posilive slgniflcaot other
W Negative signlflelot other
2 1.8
1.6 1.4
1.2
1 Partlclpant's own
A yoked participant's
Fig. 1. Average representation-derived inference and memory rating as a function of significant-other resemblance and overall evaluative tone. (Adapted from Andersen & Baum, 1994.)
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
139
5 1
Participant's own
Posibvc Significantother
Negabve signififmt other
A yoked psrticlpant's
Fig. 2. Average representation-derived evaluation rating as a function of significant-other resemblance and overall evaluative tone. (Adapted from Andersen & Baum. 1994.)
derived evaluation, constitute basic indices of transference. That is, inferences and evaluations stored in memory about past or present significant others can be transferred to newly encountered others.
IV. Motivation, Expectancies, and Affect in Transference Beyond stored inferences and evaluations about a significant other, we argue that significant-other representations include knowledge reflecting motivations, expectancies, and affect experienced previously in relationships with significant others. This implies that such forms of knowledge are also likely to be activated and applied to newly encountered others in transference. In this section, we describe a study designed to test these assumptions (Andersen et al., 1996). Participants in this study learned about a new target person who resembled either their own or a yoked participant's positively or negatively toned significant other-with perfect yoking in a design nearly identical to the study described just above (Andersen & Baum, 1994). As in all prior studies, in the significant-other-resemblance conditions, the target descriptors were derived from those the participant had previously generated to describe either a positively or negatively toned significant other. The new target person about whom participants learned was described with an equal num-
140
SERENA CHEN AND SUSAN M. ANDERSEN
ber of positive and negative descriptors, whether derived from a positively or negatively toned significant other. After learning about the target, participants completed several dependent measures including ones assessing the basic inference/memory and basic evaluation effects, as well as a variety of other representation-derived effects involving motivation, expectancies, and affect.
A. REPRESENTATION-DERIVED INFERENCE/MEMORY AND EVALUATION As predicted, our basic inference and memory effect emerged in this study in the form of higher memory confidence ratings about representation-derived descriptors that were not presented about the new target person in the significant-other-resemblance condition relative to the noresemblance control condition. As in earlier work (Andersen & Baum, 1994), the effect held both for positively and negatively toned significant others. The basic evaluation effect was also found in this study. Specifically, participants evaluated the target more favorably when the target resembled their own positively toned significant other rather than own negatively toned significant other. This effect was not found when the target resembled a yoked participant’s significant other-that is, in the no-significant-otherresemblance conditions.
B. REPRESENTATION-DERIVD MOTIVATION Although the theory of schema-triggered affect does not specify the triggering of motivation reflecting the evaluative tone of a category (S. T. Fiske & Pavelchak, 1986),in our view, the basic notion that evaluative and affective aspects associated with a significant-other representation can be activated and applied to new others is readily extended into the motivational domain. Specifically, we contend that motivation concerning a significant other is likely to be stored in the linkages between the self and otherreflecting what the self seeks to accomplish with the other. Thus, upon the activation of a significant-other representation, representation-derived motivation should occur. That is, motivational material stored in selfhignificant-other linkages should be activated and applied to a new person when a significant-other representation is activated, leading motivations pursued in the relationship with the significant other to become operative in relations with the new other. These motivational assumptions are
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
141
grounded in part in a growing body of work indicating that motivational constructs, like any other construct, are represented in memory, and thus can be similarly activated and used (e.g., Bargh, 1990, 1997; Bargh & Gollwitzer, 1994). To examine representation-derived motivation in transference, in the pretest session of this same study (Andersen et al., 1996), participants had been asked to name a positively toned significant other whom they felt close to and wanted to be still closer to, and a negatively toned significant other whom they did not feel close to and from whom they wanted to be even more distant. In the experimental session, participants' self-reported motivation for interpersonal closeness with the new target person was assessed after they learned about the target. Thus, representation-derived motivation in transference was measured in terms of the desire for interpersonal closeness in this study. As predicted, the results showed that motivations pursued in relationships with significant others are activated and applied along with significant-other representations. As seen in Figure 3, participants were more motivated to emotionally approach (and not distance themselves from) the target who resembled their own positively toned significant other relative to the target resembling their negatively toned significant other. This pattern did not hold when the target resembled a yoked participant's significant other-that is, in the no-significant-other-resemblance conditions.
4.6 4.4 4.2 4 3.8 3.6 3.4 3.2 3 2.8 2.6 2.4 Partlclpmt's own
A yoked partleipant's
Fig. 3. Average representation-derivedmotivation rating as a function of significant-other resemblance and overall evaluative tone. (From Andersen, Reznik, & Manzella, 1996. Copyright 1996 by the American Psychological Association. Adapted by permission of the publisher.)
142
SERENA CHEN AND SUSAN M. ANDERSEN
These data document representation-derived motivation in transference, and thus extend research on the theory of schema-triggered affect (S. T. Fiske, 1982) into the motivational domain. Further, the particular motivation examined-of emotionally approaching or avoiding another person-is arguably one of the most fundamental in that the need to bond or attach to others appears to be basic to survival (Andersen et al., 1997; see also Adler, 1927/1957;Bakan, 1966; Baumeister & Leary, 1995; Blatt, 1990; Bowlby, 1969; Deci & Ryan, 1985; Fairbairn, 1954; Gilligan, 1982; Greenberg & Mitchell, 1983; Guisinger & Blatt, 1994; Helgeson, 1994; Horney, 1939, 1945; McAdams, 1985, 1989; Mullahy, 1970; Rogers, 1951; Safran, 1990a, 1990b; Sullivan, 1940,1953). In this sense, demonstrating the emergence of this motivation in transference may be of considerable significance.
C. REPRESENTATION-DERIVED EXPECTANCIES Representation-derived expectancies based on an activated significantother representation were also examined in the study just described (Andersen et al., 1996). In our view, expectancies about a significant other’s responses toward the self should be stored in memory in the linkages between self and the significant other, and thus are likely to be activated and applied to a new person in transference. To examine this prediction, participants’ expectancies about how positively or negatively the new target person would evaluate them were assessed in the experimental session of this study. If a significant other is or has been harsh and rejecting, this should be manifested as an expectancy held by perceivers for how a new person onto whom the corresponding significant-other representation has been activated and applied will respond to them-namely, with rejection. Conversely, if people perceive themselves as being loved or having been loved by a significant other, experiencing transference based on the representation of this individual should lead them to expect more love (or liking) from the new person. As predicted, when the target person resembled participants’ own positively toned significant other, participants expected the target to like and accept them more than when the target resembled their own negatively toned significant other. No such pattern of positive expectancies was found in the no-significant-other-resemblance conditions. These data appear in Figure 4. Thus, stored expectancies about acceptance and rejection are activated when a significant-other representation is activated and applied to a
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
143
Fig. 4. Average representation-derived expectancy rating as a function of significant-other resemblance and overall evaluative tone. (From Andersen, Reznik, & Manzella, 1996.Copyright 1996 by the American Psychological Association. Adapted by permission of the publisher.)
new p e r ~ o n .Such ~ representation-derived expectancies implicate self/ significant-other linkages stored in memory in that they reflect perceivers’ perceptions of a relevant significant other’s feelings toward them. On a broader level, these data are of significance in light of a growing body of work on the consequences of chronic expectancies about acceptance versus rejection in interpersonal relations (see also Downey & Feldman, 1996; Feldman & Downey, 1994),as well as theory and research on the role of outcome contingencies in interpersonal relations (e.g., Bandura, 1986; Higgins, 1989b).
D. REPRESENTATION-DERIVED FACIAL AFFECT Finally, affective responses were also examined in the above study (Andersen et al., 1996). Given that emotional responses may last only a few seconds (Ekman, 1992), the focus was on affect shown in momentary facial expressions. Specifically, participants’ facial expressions were covertly vidA reader might wonder about the distinctiveness of the measures of representationderived evaluation, motivation, and expectancies used in this study. A factor analysis was conducted on the entire pool of self-report items meant to comprise these three measures. The results of this analysis generally verified the distinct psychological meaning of these measures (for a further discussion, see Andersen et al., 1996, pp. 1112-1113).
144
SERENA CHEN AND SUSAN M. ANDERSEN
eotaped as they learned each descriptor about the new target person. Monitoring the face in this way provided a nonverbal measure of the affect experienced immediately upon encoding each target descriptor. Participants’ facial affect was expected to reflect the overall evaluative tone of the significant-other representation. Two trained judges, blind to condition, rated participants’ facial expressions for pleasantness (i.e., positive versus negative feeling) as they read the descriptors, one at a time. As anticipated, participants responded with more positive facial affect when learning about the target resembling their own positively toned rather than their own negatively toned significant other-a pattern not seen when the target resembled a yoked participant’s significant other. These findings are shown in Figure 5. Beyond examining participants’ facial affect as a function of the overall evaluative tone of the relevant significant-other representation, analyses were also conducted to assess whether participants’ facial affect was responsive to the valence of the individual target descriptors, which was manipulated independently of the overall evaluative tone of the representation, as noted earlier. The results suggested differences in the extent to which the predicted effect of the overall evaluative tone of the representation emerged as a function of descriptor valence. Specifically, although the predicted effect was found across target-descriptor valence, it emerged more unambiguously for negative target descriptors than for positive ones.
4.2
4.1
4
3.9 Partlcipant‘sown
A yoked partlclpmt’s
Fig. 5. Judges’ average rating of positivity in facial affect at encoding as a function of significant-otherresemblance and overall evaluative tone. (From Andersen, Reznik, & Manzella, 1996.Copyright 1996by the American PsychologicalAssociation. Adapted by permission of the publisher.)
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
145
As shown in Figure 6, for negative descriptors, participants showed the most positive affect when the target resembled their own positively toned significant other relative to negatively toned significant other. Stated differently, participants responded to negative target descriptors associated with a positively evaluated significant other with especially positive facial affect. One interpretation of this finding is that the inconsistency of the negative target descriptors in relation to the overall positive tone of the significant-other representation led perceivers to transform the negative information into a positive response. This possibility is intriguing and fits recent evidence in the close relationships literature suggesting that when faced with a romantic partner’s fault, perceivers tend to formulate “theories” that dampen or dispel the negativity implied by the fault (Murray & Holmes, 1993,1994,1996). In both cases, perceivers appear to be motivated to maintain positive conceptions of their significant others. Interestingly, because self-relevance is thought to be essential to the experience of emotion (Lazarus & Averill, 1972; Greenwald, 1982; Greenwald & Pratkanis, 1984), this type of finding may suggest the special self-relevance of positively toned significant others. We discuss these provocative notions in more detail in a later section. Overall, using a virtually instantaneous measure of affect, the results constitute evidence for representation-derived facial affect in transference. Learning about a new person who resembles a significant other influences facial expressions so as to reflect the overall tone of the relevant significant4.4 , Posltive dncriplar valence
Negative descriptor valence
4.3
4.2
4.1
4
3.9 Positive
Negative
Puncipsnt’s own ri@lcmt otber
Positive
Negative
A yoked pYndp8nt’s signitlclnt other
Fig. 6. Judges’ average rating of positivity in facial affect at encoding as a function of significant-other resemblance, overall evaluative tone, and target-descriptor valence. (From Andersen, Reznik. & Manzella. 1996. Copyright 1996 by the American Psychological Association. Adapted by permission of the publisher.)
146
SERENA CHEN AND SUSAN M. ANDERSEN
other representation. At the same time, these data also suggest the special role of negative descriptors in the affect experienced in transference involving positively toned significant others. On the other hand, we note that analogous evidence in the form of self-reported transient mood, also assessed in this study, was not found, suggesting that the observed facial expressions reflected fleeting affective experiences and not phenomenal mood states (see section V for a further discussion). Nonetheless, as indicated, the data clearly show that representation-derived facial affect emerges in transference, extending prior work on transference into the domain of emotion. Taken as a whole, this study extended initial evidence for the socialcognitive model of transference by replicating the basic inference/memory and evaluation effects in transference, and by demonstrating that motivation, expectancies, and facial affect derived from an activated significantother representation emerge as part of transference (Andersen et al., 1996). These data clearly suggest the breadth of the phenomenon, revealing the multifaceted ways in which past experiences in relationships with significant others come to play a role in present interpersonal relations.
V. Transient Mood States and Interpersonal Roles in Transference Given that significant-other representations are likely to be highly affectively laden, their activation and application to a new person might, at least under some circumstances, influence self-reported transient mood states. Specifically, if perceivers feel unhappy around a significant other, they might be more likely to experience depressive mood states when the representation of this significant other is activated as compared to when a representation of a significant other around whom they feel happy is activated. Essentially, transference may involve representation-derived transient mood states. In one study, self-reported transient mood states did in fact tend to support this possibility (Andersen & Baum, 1994). However, the effect size in this study was small, and the effect failed to replicate, as noted in the study above (Andersen et al., 1996). In this section, we take another look at representation-derived transient mood by considering the impact of a contextual factor on the emergence of this mood effect in transference. Specifically,we describe a study focused on the interpersonal role that perceivers typically experience in relation to their significant others (Baum & Andersen, in press). From our view, the role one experiences in a relationship with a significant other should be embodied in the linkages between the self and this other in memory. Thus,
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
147
when a new person resembles a perceiver’s significant other, role information linking the significant other with the self should be activated, leading the perceiver to respond to the new person in part as a function of this role (cf. A. P. Fiske, 1992; A. P. Fiske & Haslam, 1996). One implication of this prediction is that the role congruence or “fit” of the new person’s role with that typically experienced in the relationship with the significant other may have consequences for transient mood states. More specifically, considerable research suggests that role relationships entail goals and expectancies pertaining to how the self and others typically relate, as well as to typical successes and disappointments experienced in these relationships (Andersen et al., 1996; Baum & Andersen, in press: see also Baldwin & Sinclair, 1996; Bugental, Lyon, Krantz, & Cortez, 1997; Downey & Feldman, 1996; Downey, Khouri, & Feldman, 1997; Luborsky & Crits-Christoph, 1990; Oatley & Bolton, 1985). Moreover, the ways in which goals are or are not achieved have shown to predict affect (e.g., Martin, Tesser, & McIntosh, 1993), and role violations in relationships are often experienced negatively (e.g., Williamson & Clark, 1992; Williamson, Clark, Pegalis, & Behan, 1996). Together, these lines of research suggest that the mood states experienced with a significant other are likely to be associated with one’s role relationship with the other-namely, with the satisfaction of goals in this relationship. Thus, when a new person resembles a significant other, the goals one pursues in relation t o the significant other should become operative in relations with the new other. If these goals are then contradicted or frustrated with the new person, negative affect is likely to result. Stated somewhat differently, a congruent role in transference is likely to signal that typical interaction patterns can be played out, whereas an incongruent role is likely to signal their disruption. Hence, when a significant-other relationship is positive, experiencing a congruent role in transference should be associated with relatively positive mood, whereas experiencing an incongruent role should frustrate goals and violate expectations, leading to negative mood. In the study designed to examine these assumptions, participants learned about a new target person resembling their own or a yoked participant’s positively toned significant other (Baum & Andersen, in press). Although this study only examined positively toned significant others, the target was described, as usual, with an equal number of positive and negative descriptors generated earlier by the participant or yoked participant. All participants anticipated an interaction with the target in which the target was in the role of either an “expert” or a “beginner”-and they were in the complementary role. In the pretest session, participants had been asked to identify a significant other who was an authority figure to them-that
148
SERENA CHEN AND SUSAN M. ANDERSEN
is, whose relational role was that of a more knowledgeable, powerful other. As such, the target’s interpersonal role was manipulated in the experiment to be either similar to or opposite from that of participants’ significant other.
A. REPRESENTATION-DERIVED INFERENCE/MEMORY AND EVALUATION
As usual, transference was assessed in terms of representation-derived inference and memory, as well as representation-derived evaluation, as a function of significant-other resemblance in the new target person. Both the basic inference and memory and basic evaluation effects emerged in the form of greater representation-derived memory confidence ratings and more positive evaluation in the significant-other-resemblance condition than in the yoked-participant control condition. Given that all significantother representations were positively toned, these simple findings verify the occurrence of transference in the significant-other-resemblance condition.
B. REPRESENTATION-DERIVED TRANSIENT MOOD STATES
As predicted, participants’ reports of their current mood states after learning about the target reflected the overall tone of the relevant representation (i.e., they felt better) when the target resembled their own positively toned significant other and was in a congruent rather than incongruent role relative to the significant other’s role. Indeed, when this target was in an incongruent role, participants felt particularly badly, reporting increased depressive mood. Such affective differences did not emerge in the nosignificant-other-resemblance conditions, suggesting that the effect is unique to the transference context. These data appear in Figure 7. Overall, role fit appears to facilitate representation-derived transient mood states in transference, and lack of role fit appears to interfere with it. This study suggests that role incongruence in transference involving positively toned significant others may signal that outcomes usually achieved with a significant other will not be obtained. Thus, when a significant other is viewed positively,role incongruence in transference may lead to dysphoric mood (see also Higgins, 1989b; Kruglanski & Jaffe, 1988). Whether or not goal disruption is the precise mechanism underlying this effect, the data clearly extend work on transference into the realm of role relationships, both chronic and situational, as well as into the immediate social context of the transference encounter.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
149
0.4 ?
Role C O ~ ~ N C O I Role incongruent
0.3 0.2 0.1
0
4.1 4.2
I
Pnrtldpsnt's own
A yoked participant's
I
Fig. 7. Average transient depressive mood rating, corrected for pretest mood, as a function of significant-other resemblance and role congruence. (Adapted from Baum & Andersen, in press. 0 1999 Guilford Press. Reprinted with permission.)
Although research examining negatively toned significant-other representations and role congruence or incongruence is also needed, the results of this study clearly indicate that information about normative role relationships with significant others is stored in memory as part of selfkignificantother linkages, and thus can be activated in the context of transference. Hence, in addition to idiographic content in such linkages, there is normative role content as well (cf. A. P. Fiske, 1992; A. P. Fiske & Haslam, 1996). Overall, this evidence deepens the complexity of the transference phenomenon by suggesting that interpersonal roles experienced previously in relationships with significant others provide structure for and constrain affect in relations with new others in transference.
VI. The Self in the Transference Context Central to our view of transference is the notion that self and significantother representations are linked in memory. Much of the evidence described above supports the existence of stored, selfhignificant-other linkages in that these linkages are thought to be implicated in representation-derived motivation, expectancies, and affect in transference (Andersen et al., 1996),
150
SERENA CHEN AND SUSAN M. ANDERSEN
as well as in the impact of role information in transference (Baum & Andersen, in press). Beyond these transference effects, the existence of selfhignificant-other linkages suggests that the activation of a significant-other representation should activate aspects of the self that are linked to the relevant significant other. That is, these linkages should allow the activation of knowledge stored about a significant other to spread to the particular subset of selfknowledge that is strongly linked to this important other. As a result, the “working self-concept” (e.g., Markus & Wurf, 1987; Linville & Carlston, 1994), or the subset of self-knowledge that is active at any given moment, should come to reflect the self experienced when thinking about or being around the significant other-in other words, the self-with-sign@cant-other (see also Ogilvie & Ashmore, 1991). In terms of our basic research paradigm, then, significant-other resemblance in a new target person should lead to shifts in the working selfconcept toward the self when with the relevant significant other. Furthermore, given evidence for representation-derived affective responses in transference (e.g., Andersen et a]., 1996), described in earlier sections, selfevaluative responses reflecting the overall tone of the representation of the relevant significant other and the associated relationship with this other may also occur (see also Baldwin, 1992; Baldwin & Holmes, 1987; Baldwin et al., 1990). In this section, we describe a study designed to test these assumptions (Hinkley & Andersen, 1996). As implied above, this study tested the hypothesis that when a significant-other representation is activated and applied to a new person, stored self/significant-other linkages are also activated, bringing to mind aspects of the self experienced in the relationship with the relevant significant other, and thus resulting in predictable shifts in the content and evaluation of the working self-concept in the context of transference. In the pretest session of this study, each participant provided a general self-description as a measure of his or her general working self-concept by completing a list of sentences beginning with the prompt “Generally, I . . .” with self-descriptive predicates (e.g., “worry about my weight” or “am very direct”). Participants then named and listed both positive and negative descriptors for one positively and one negatively toned significant other. Next, participants were asked to generate self-descriptors reflecting the way they are when with each significant other by completing a list of sentences beginning with the prompt “When I’m with NAME, I . . .”) with descriptive predicates (e.g., “feel extremely feminine” or “am totally relaxed”). Finally, participants classified each of their working-self-concept descriptors as positive or negative-as a measure of self-evaluation at
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
151
the level of self-reported descriptor valence. All of this was prior to the experimental session, and allowed the overlap between each participant’s general working self-concept and self-with-significant-other to be calculated at pretest for later use as a covariate (see below). In the experimental session, participants learned descriptors about a new target person allegedly seated next door (as in Andersen & Baum, 1994). The target resembled either their own or a yoked participant’s positively or negatively toned significant other (with perfect yoking), and was described with an equal number of positive and negative descriptors. Afterward, participants’ general working self-concepts were once again measured. That is, participants were asked to describe themselves as they “are now” by generating self-descriptors. As with the self-descriptors generated in the pretest session, participants classified each of these self-descriptors as positive or negative. Once again, the overlap between the general working self-concept and the self-with-significant-other was assessed. Finally, participants completed a standard recognition-memory test which measured, as usual, representation-derived inference and memory based on an activated significant-other representation.
A. REPRESENTATION-DERIVED INFERENCE/MEMORY
As predicted, the basic inference and memory effect emerged such that participants showed greater representation-derived inference and memory about the target resembling one of their own significant others versus a yoked participant’s significant other. This effect held regardless of the overall evaluative tone of the significant-other representation from which the target descriptors were derived. Thus, the data indicate that significantother resemblance in the new target person invoked the activation and application of the significant-other representation to the target in this study, and thus implies that any effects emerging in this condition did so in the context of transference. B. SHIFTS IN THE CONTENT O F T H E WORKING SELF-CONCEPT IN TRANSFERENCE To examine the hypothesis that the activation of a significant-other representation leads to a shift in the working self-concept toward the self experienced in the relationship with the relevant significant other, a measure of overlap was used (see Prentice, 1990). That is, as indicated, the degree of overlap between participants’ general working self-concept and their self-
152
SERENA CHEN AND SUSAN M. ANDERSEN
with-significant-other was calculated on an item-by-item basis both at the pretest session and after exposure to the experimental manipulation. More specifically, at each measurement time, independent judges calculated the number of idiographic descriptors in participants' working self-concept lists that overlapped with their self-descriptors reflecting what they are like when with the significant other. When a new target person activates a significant-other representation by virtue of his or her resemblance to a significant other, the degree of this overlap should increase, controlling for pretest overlap. Such a shift in the working self-concept should not occur in the absence of significant-other resemblance. As predicted, participants' working self-concept did in fact come to overlap more with the self-with-significant-other when the target resembled the participant's own significant other rather than a yoked participant's significant other. These results, depicted in Figure 8, suggest that in the context of transference, perceivers appear to become the self they are when with the relevant significant other. Despite appearances in this figure, no interaction emerged, indicating that this overlap effect held whether the target resembled a positively or a negatively toned significant other. Importantly, no such pattern of findings was found when overlap was calculated in relation to an irrelevant self-with-significant-other (i.e., for the self-withsignificant-other with an overall tone opposite to that of the relevant significant other). Thus, the observed shifts in the content of the working self0.9
,
0.8 0.7 0.6
0.5 0.4
0.3 0.2 0.1 0 Participant's own
A yoked participant's
Fig. 8. Overlap between the working self-concept and the relevant self-with-significantother, corrected for pretest overlap, as a function of significant-other resemblance and overall evaluative tone. (From Hinkley & Andersen, 1996.Copyright 1996 by the American Psychological Association. Adapted with permission of the publisher.)
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
153
concept were specific to the relevant self-with-the-significant other and not just any significant other. C. SHIFTS IN SELF-EVALUATION O F THE WORKING SELF-CONCEPT IN TRANSFERENCE To measure shifts in the evaluative tone of the working self-concept in transference, we examined participants' positive and negative classifications of their self-descriptors in the experimental session, controlling for the valence of these self-descriptors as assessed in the pretest session. Our main analyses of self-evaluative changes were conducted using the overlapping items-the working-self-concept descriptors that came to reflect the relevant self-with-significant-other. When the target person resembled the participant's own significant other-that is, in the transference context-we expected these newly overlapping self-descriptors to reflect the evaluative tone of the relevant significant-other representation and associated selfwith-significant-other. Support was in fact found for this prediction, as depicted in Figure 9. Participants perceived their newly overlapping working self-concept descriptors as more positive when the target resembled their own positively toned significant other rather than their own negatively toned significant
Posltlve slgalflcint other
Participant's own
Negative r@mcmt other
A yoked participant's
Fig. 9. Valence of working self-descriptors overlapping with the relevant self-with-signihcant-other, corrected for pretest valence. as a function of significant-other resemblance and overall evaluative tone. (From Hinkley & Andersen, 1996. Copyright 1996 by the American Psychological Association Adapted by permission of the publisher )
154
SERENA CHEN AND SUSAN M. ANDERSEN
other, a difference not found when the target resembled a yoked participant's significant other. Thus, shifts in the working self-concept that occur in transference involve changes in self-evaluation that are derived from the overall tone of the significant-other representation and associated self-withsignificant-other. Importantly, the effect occurred for both positively and negatively toned significant others. An analysis was also conducted on the nonoverlupping items (i.e., the working self-concept items that did not overlap with the self-with-significant-other in the experiment), and this analysis revealed a very different pattern. As seen in Figure 10, the valence of the items not overlapping with the self-with-significant-other in the experiment became most positive when the target resembled the participant's own negatively toned significant other rather than positively toned significant other. A similar effect was not found in the no-significant-other-resemblance condition when the target resembled a yoked participant's significant other. These findings suggest that when a new person resembles a negatively toned significant other, positively toned aspects of a perceiver's overall pool of self-knowledge that do no?overlap with the self-with-significant-other come to predominate in his or her working self-concept in transference involving negatively evaluated significant others. Presumably, this occurs in response to the small influx of negative working-self-concept descriptors (i.e., the overlapping descriptors).
Participant's own
A yoked partlclpant's
Fig. 10. Valence of working self-descriptors nonoverlupping with the relevant self-withsignificant-other, corrected for pretest valence, as a function of significant-other resemblance and overall evaluative tone. (From Hinkley & Andersen, 1996.Copyright 1996by the American Psychological Association. Adapted by permission of the publisher.)
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
155
Interestingly, the nonoverlapping items comprised the majority of the working-self-concept descriptors listed, which implies that the compensatory self-evaluative response seen in the context of negative transference was the dominant self-evaluative response. Indeed, overall analyses of the working-self-concept descriptors verify this interpretation in the present data. Such compensatory responses are common in various domains of self research, particularly those focused on perceivers’ responses to threats to the self (e.g., Greenberg & Pyszczynski, 1985; Steele, 1988;Taylor & Brown, 1988; Taylor & Lobel, 1989; Tesser, 1988). In sum, in the context of transference, shifts in the working self-concept occurred in the direction of the relevant self-with-significant-other, or the self experienced in the relationship with the relevant significant other. Coupled with these content shifts were changes in self-evaluation in the newly overlapping aspects of the working self-concept reflecting the tone of the significant-other representation and the associated self-with-significantother. Intriguingly, the negative (or relatively less positive) self-evaluative changes in the working self-concept that occurred in the context of transference based on the negatively toned significant-other representation apparently led to self-evaluative bolstering in the overall working self-concept, rendering overall self-evaluation especially positive based on a negative significant other. Whether this bolstering occurs primarily on-line in transference, or also occurs in the relationship with the significant other, is unknown, but the result is clearly provocative in terms of the nature of transference responses based on significant others toward whom perceivers do not feel positively. Having presented an array of findings concerning the inferential, evaluative, motivational, affective, role-related, and self-relevant outcomes that emerge in transference, we now turn our attention to the basic socialcognitive principles governing the transference phenomenon. Specifically, in the next several sections, we discuss issues pertaining to how and when significant-other representations are likely to be activated and applied to new others and describe the growing body of research addressing these issues.
VII. Basic Social-Cognitive Principles Underlying Transference
A central assumption of our model of transference is that the activation and application of significant-other representations can be understood in terms of the same basic social-cognitive principles thought to guide the activation and use of any other social construct-an assumption we stated
1.56
SERENA CHEN AND SUSAN M. ANDERSEN
earlier in laying the groundwork of the model. In particular, we delineated both chronic and transient sources of activation for significant-other representations. Chronic sources, or the chronic accessibility of a construct, arises from its frequent past activation and implies a high likelihood of activation even in the absence of transient sources (e.g., Higgins & King, 1981). Transient sources may also play a role in construct activation in the form of priming, which refers to recent activation, or applicability, which refers to activation arising from the “match” or overlap between cues in a new person and knowledge stored about a construct (e.g., Higgins, 1996). Overall, then, construct activation and application can occur based on chronic sources, transient sources, or some combination (e.g., Bargh et al., 1986). Increases in the contribution from any of these sources are thought to produce corresponding increases in construct activation (e.g., Chen et al., in press; Higgins & Brendl, 199.5). Applied to the activation and application of significant-other representations, these basic principles suggest that chronic sources of triggering significant-other representations should combine with transient sources of activation to trigger transference. We conducted two studies to examine the chronic and transient activation of significant-other representations (Andersen et al., 199.5). In the first, we examined whether or not chronic and transient priming sources combine to trigger transference, and in the second, whether or not chronic and transient applicability sources combine to do so. A. CHRONIC AND PRIMING SOURCES OF TRIGGERING TRANSFERENCE
To examine whether or not chronic and priming sources combine to elicit transference, the presence versus absence of priming was manipulated in the first study (Andersen et al., 1995, Study 1). Specifically, participants took part in a pretest session in which they named and described a significant-other representation and as a control condition, a nonsignificant-other representation. They did this either immediately before learning about a series of new target persons in the experimental session of our research paradigm (priming condition), or did it several weeks beforehand (nopriming condition). In the experimental session, participants in both priming conditions learned descriptors about four new target persons in an individualized random order. These targets resembled, respectively, their own significant other, their own nonsignificantother, a yoked participant’s significant other, and a yoked participant’s nonsignificant other. Afterward, participants com-
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
157
pleted a standard recognition-memory test about each target, which was used, as usual, to assess transference in the form of representation-derived inference and memory. As depicted in Figure 11, participants showed greater representationderived inference and memory about the target resembling their own significant other than about the target resembling their own nonsignificant other. Greater representation-derived inference and memory were also found for the target resembling participants' own significant other relative to each of the yoked-participant targets. No difference was found between the yoked-participant's significant-other and nonsignificant-other targets, indicating that the reliable difference between participants' own significantother and nonsignificant-other targets cannot be attributed to objective differences in the descriptors listed for significant others versus nonsignificant others in general. Importantly, these findings held across priming conditions, suggesting the chronic accessibility of participants' own significant-other representation relative to their own nonsignificant-other representation-that is, their chronic readiness to be activated and applied to new others, even in the absence of any activation arising from priming. Beyond this, the results also showed that in the priming condition, thinking about and listing descriptors to characterize a significant other and
Sigal5cnot other
:NonrignW3cnot other
2.4
23 2.2 2.1 2 1.9 1.8 1.7
I .6 1.5
1.4 1.3
1.2 1.1 1
Participant's own
A yoked participant's Priming
Participant's own A yoked participant's No priming
Fig. 11. Average representation-derived inference and memory rating as a function of resemblance to participants' own or a yoked participant's representation, the significance or nonsignificance of the representation, and priming condition. (From Andersen, Glassman, Chen, & Cole, 1995. Copyright 1995 by the American Psychological Association. Adapted with permission of the publisher.)
158
SERENA CHEN AND SUSAN M. ANDERSEN
nonsignificant other in the pretest session just prior to the experimental session primed the representations of these individuals. That is, representation-derived inference and memory were greater in the priming condition than in the no-priming condition for the targets resembling the participant’s own significant other and nonsignificant other. This priming effect can be seen in Figure 11 in the increase from the no-priming to priming conditions for both of these representations. Also apparent is the fact that this effect occurred independently of the chronic-accessibility finding-the latter evidenced in the form of greater representation-derived inference and memory for the target resembling participant’s own significant other relative to nonsignificant other across priming conditions. Overall, then, the data indicate that both chronic sources and transient priming sources of activation contribute to the triggering of transference. B. CHRONIC AND APPLICABILITY SOURCES OF TRIGGERING TRANSFERENCE If significant-other representations are chronically accessible, they should have a high likelihood of activation in the absence of not only priming, but also applicability. And yet applicability, embodied in cues in a new person that “match” or overlap with knowledge stored about a significant other, should increase the likelihood that the relevant significant-other representation will be activated and applied to the person (e.g., Higgins, 1996; see also Tversky, 1977). That is, as with priming sources, applicability sources should combine with chronic sources to trigger transference (e.g., Bargh et al., 1986). We examined this prediction in a second study by manipulating the presence versus absence of applicability (Andersen et al., 1995, Study 2). As in all of the research described thus far, applicability-based triggering was operationalized in the form of “match” between descriptors associated with a perceiver’s significant-other representation and the cues presented about a new target person-that is, significant-other resemblance in the target. What distinguishes this study from all other research described above-in which significant-other resemblance has always been manipulated-is the manner in which the inference and memory measure was constructed, as is described below. In the pretest session of this study, particpants were asked to name and generate descriptors €or two different significant others. To provide stimuli for a new “no-representation” control condition in which the descriptors were self-generated, we asked participants to generate one sentence for each of a diverse list of well-known individuals. As a set, these descriptive
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
159
sentences would not constitute a single representation, thereby allowing an assessment of self-generation effects (Greenwald & Banaji, 1989) in the experimental session in the absence of representational coherence. Participants learned about four target persons in the learning phase of the experimental session, held two weeks after the pretest session for all participants. In the applicability condition-which is our standard significant-other-resemblance condition-participants learned about a target who resembled one of their own significant others, and later completed a standard recognition-memory test using representation-derived descriptors for this same significant other as test items. In the no-applicability (i.e., nosignificant-other-resemblance) condition, the target resembled one of a yoked participant’s significant others, but the test items for this target were descriptors derived from another of the participant’s own significant others (the one not used in the applicability condition). Thus, the extent of representation-derived inference and memory about a new target person based on one of participants’ own significant-other representations could be assessed in the presence of applicability-based triggering (i.e., significantother resemblance) versus the absence of such triggering (i.e., no-significantother resemblance). Two control conditions were also included in this study. In one, another no-significant-other-resemblance condition, descriptors about another significant other of a yoked participant were used as target descriptors, and in this condition, the recognition-memory test used descriptors that were also provided by this yoked participant. In other words, this condition was our standard yoked-participant control condition. In the other condition, the descriptors participants listed as their no-representation sentences were used as both target descriptors and as items in the recognition-memory test. As predicted, greater representation-derived inference and memory based on a significant-other representation emerged in the applicability condition relative to no-applicability condition. These data are depicted in Figure 12. Hence, applicability-based triggering appears to play a role in the activation and application of significant-other representations to new others. Importantly, however, evidence for the chronic accessibility of significant-other representations was once again found. That is, in the noapplicability condition, greater representation-derived inference and memory based on a significant-other representation were observed relative to the representation-derived inference and memory seen in each of the control conditions. This finding speaks to the chronic accessibility of significantother representations in that, even in the absence of applicability, significant-other representations appear to be activated and applied to some degree to new others. At the same time, the greater representation-derived inference and memory seen in the applicability versus no-applicability con-
160
SERENA CHEN AND SUSAN M. ANDERSEN
Fig. 12. Average representation-derived inferencehemory rating as function of target condition. (From Andersen, Glassman, Chen, & Cole, 1995. Copyright 1995 by the American Psychological Association. Adapted by permission of the publisher.)
dition speaks to the importance of applicability-based triggering in transference. Taken together, these two studies suggest that significant-other representations are chronically accessible. They have a chronic readiness to be activated and applied to new others even in the absence of activation arising from either transient priming or applicability sources. And yet both of these transient sources of activation may contribute to the activation and application of significant-other representations to new others. This is evidenced in the fact that, when present, they combined with the chronic accessibility of significant-other representations to enhance the triggering of transference. Overall, this research demonstrates that basic social-cognitive principles of construct activation and use can be readily applied to the domain of significant-other representations, providing considerable precision regarding the processes underlying transference. Beyond this, the evidence found in this research for the chronic accessibility of significant-other representations suggests the ubiquity of the transference phenomenon.
VIII. Applicability-Based Triggering of Transference The research described just above clearly indicates that applicability, defined in terms of a new target person’s “match” or resemblance to a
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
161
significant other, increases the likelihood that the relevant significant-other representation will be activated and brought to bear on interpreting him or her (Andersen et al., 1995, Study 2). Evidence for applicability-based triggering of transference also lies in the broader body of research demonstrating the emergence of inferential, affective, motivational, role-related, and self-relevant effects in the presence but not absence of significantother resemblance in a new target person. Our theorizing and research on applicability is part of a growing social-cognitive literature on this source of construct activation and use (for reviews, see Chen et al., in press; Hardin & Rothman, 1997; Higgins, 1996). In the next few sections, we consider applicability-based triggering in transference more closely. In the first section, we scrutinize existing assumptions about the underlying basis of applicability-that is, the units of knowledge that constitute “match” between a stored construct and a to-beinterpreted stimulus. Specifically, we begin by discussing longstanding feature-based approaches, which tend to view mental representations as composed of lists of features or attributes, and thus generally assume that applicability is feature-based-that is, that the activation and use of a representation to interpret a stimulus occur in part as a function of the degree of “match” between feature-based cues in the stimuIus and featurebased knowledge stored about the representation. We conclude this section by discussing recent theory-based approaches that offer an alternative to feature-based views by suggesting that the content, structure, and use of mental representations may involve theorybased forms of knowledge. We examine such approaches in considerable depth so as to lay the groundwork for the following section in which we present a theory-based approach to significant-other representations and transference, and describe a line of research designed to examine it. Ultimately, this work raises the possibility that applicability-based triggering in transference may be in part theory-based (Chen, 1997).4
A. FEATURE-BASED APPROACHES T O APPLICABILITY In all of the research described thus far, applicability has been conceptualized as a source of activation for significant-other representations arising from the overlap between feature-based cues in a new target person and Although the present focus is on feature-based and theory-based applicability, it is worth noting that there are undoubtedly various other possible bases of the role of applicability in triggering transference. For example, physical characteristics in a new person that resemble those associated with a significant other may often add to the contribution of applicabilitybased triggering to transference. This possibility warrants future attention.
162
SERENA CHEN AND SUSAN M. ANDERSEN
feature-based knowledge stored about a significant other (e.g., “Terry is sincere” or “Pat likes to go dancing”). As stated in an earlier section, we construe cues in a new target person as an analog for cues perceivers would encounter on-line in an actual social encounter. In more concrete terms, participants’ conceptions of their significant others have been assessed simply in terms of lists of descriptive features they have generated about these individuals. In turn, transference has been shown to occur in part as a result of the feature-based resemblance of a new target person to one of participants’ significant others. Indeed, as indicated, research has shown a variety of representation-derived effects reflecting the activation and application of a significant-other representation to a new person when such feature-based applicability is present versus when it is absent (e.g., Andersen et al., 1995, Study 2). Overall, then, featurebased applicability appears to play some role in triggering transference. Our feature-based approach to applicability is rather common. Featurebased assumptions are made, either implicitly or explicitly, by various socialcognitive models on construct activation and use (e.g., Higgins, 1996; Sedikides & Skowronski, 1990,1991; Wyer & Srull, 1986), as well as by various cognitive models of categorization (e.g., Rosch, 1978). The latter are of relevance because categorization processes are thought to be similar to construct activation and use in terms of basic predictions. Though their emphases vary considerably, these feature-based models converge in the basic notion that a new person (or object) is likely to be compared with stored knowledge in part as a function of matching and mismatching features between the person and the stored knowledge (e.g., Tversky, 1977; Tversky & Gati, 1978; see also Cantor & Mischel, 1979; Nosofsky, 1986; Rosch, 1978; Smith & Medin, 1981). An alternative basis for applicability, however, lies in theory-based approaches to categories and categorization. Such approaches have appeared across a wide range of research (for reviews, see Komatsu, 1992; Murphy & Medin, 1985). We turn now to a discussion of such approaches (see also Chen, 1997), emphasizing relevant theory and research in the socialcognitive domain so as to set the stage for a consideration of the potential implications that these approaches carry for significant-other representations and transference. B. THEORY-BASED APPROACHES TO APPLICABILITY Theory-based approaches argue that “theories” are basic to the nature and role of stored knowledge in perception and cognition (e.g., Murphy & Medin, 1985). Theories generally refer to knowledge that serves expressly
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
163
as explanation. Unlike unrelated, feature-based units of knowledge, theories are thought to be embodied in explanatory relations linking together bits of knowledge about a given entity. Indeed, some have argued that viewing categories simply in terms of lists of unrelated features falls short of providing a full account of what makes a category coherent (Medin, 1989; Murphy & Medin, 1985: see also Armstrong, Gleitman, & Gleitman, 1983; Ortony, 1979; Rips, 1991; Rips & Collins, 1993; cf. Goldstone, 1994; Medin, Goldstone, & Gentner, 1993). One implication of theory-based approaches is that categorization, the process whereby a stimulus is interpreted in terms of a stored category, may occur on the basis of more than feature-based matches and mismatches between a stimulus and category. In applicability terms, applicability’s role in construct activation and use may not only be based on feature-based “matches” between a stimulus and a stored construct, but also on theorybased “matches.” Theorizing and research consistent with a theory-based approach are increasingly seen in various psychological subdisciplines. Outside of the social domain, they can be seen in cognitive work on not only categorization (e.g., Wisniewski & Medin, 1994) but also analogical thinking (e.g., Gentner & Markman, 1997; Holyoak & Thagard, 1995), as well as in work on conceptual development in children (e.g., Carey, 1985; Gelman & Markman, 1986; Keil, 1989). In the social-cognitive realm, theory-based views have emerged in, for example, research on stereotype representations and stereotyping (e.g., Kunda & Thagard, 1996; Wittenbrink, Gist, & Hilton, 1997) as well as social perception more broadly (e.g., Park, DeKay, & Kraus, 1994; Read, Vanman, & Miller, 1997). Interestingly, one of the earliest contributions to social cognition documented the inadequacy of additive models of person perception treating features as unrelated entities (Asch, 1946). Reflecting its Gestalt roots, this work showed that features conceived within the context of persons are part of unified wholes rather than sets of unrelated features (see also Heider, 1958). Theoretical relations among features play a more explicit role in recent work in the same tradition. Specifically, this research has demonstrated that one way people make sense of incongruent information about others is to impose a theory-that is, to call upon knowledge to explain the incongruence (Asch & Zukier, 1984; see also Hastie, Schroeder, & Weber, 1990; Kunda, Miller, & Claire, 1990; Read et al., 1997). In a similar vein, research has shown that people tend to integrate inconsistencies in their experiences into internally coherent stories or narratives (e.g., Baumeister & Newman, 1994; see also Murray & Holmes, 1993, 1994, 1996; Pennington & Hastie, 1992, 1993; Read & Miller, 1993; Thagard, 1989; Zukier, 1986). Finally, the notion that relations among features exist is in
164
SERENA CHEN AND SUSAN M. ANDERSEN
fact fundamental to “implicit personality theories” (e.g., Schneider, 1973;see also Leyens & S. T. Fiske, 1994), which are thought to link traits together, allowing perceivers to infer some traits from the presence of others. Stereotyping research provides more suggestive evidence for a theorybased approach to the activation and use of stored knowledge. For example, research on the impact of “fit” on social categorization suggests that theorybased knowledge influences the likelihood that a target person will be categorized as a member of a social group. More specifically, this work showed that to the extent that the features (e.g., attitudes) of a target person “fit” a perceiver’s “theoretical” expectations about a social group (e.g., females), the target was more likely to be categorized as a member of the group (Oakes, Turner, & Haslam, 1991). Work on implicit stereotyping has yielded analogous data, showing that the effects of priming genderstereotypic traits (e.g., dependence) on ratings of male and female targets on these trait dimensions were greater when there was stereotype-consistent fit between the primed trait and gender of the target (Banaji et al., 1993). Both cases suggest that categorization may not simply be based on the degree of feature-based “match” between cues in a target person and stored knowledge. Still more direct evidence for the notion that applicability may operate on the basis of theory-based “match” can be found in cognitive research on analogical reasoning, which suggests that structural “parallels” between stored knowledge and a stimulus may help trigger the retrieval and use of the knowledge in interpreting the stimulus (e.g., Gentner & Markman, 1997; Gick & Holyoak, 1980, 1983; Holyoak & Thagard, 1995; see also Markman & Gentner, 1993). Similar findings have been obtained in socialcognitive research on analogy, in which retrieval of a stored explanation was facilitated when participants were asked to explain an unexpected event similar in causal structure to the one for which the explanation was originally generated (Read & Cesa, 1991). Thus, stimulus cues reflecting causal relations help to trigger and render applicable the stored explanatory knowledge. Building on this analogy work, recent research on stereotypes has explicity adopted a theory-based approach. This research argues that stereotype representations contain knowledge about “cause-effect” relations that influences the processing of stereotype-relevant information (Wittenbrink et al., 1997; Wittenbrink, Hilton, & Gist, 1998; see also Hirschfeld, 1996). One series of studies found that “matches” between cause-effect relations stored about a stereotype and those embodied in stimulus cues in the social environment contributed to the activation and use of the stereotype (Wittenbrink et al., 1997). Such evidence clearly fits the notion that applicability may at times be theory-based.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
165
The critical question for the present purposes, of course, is whether or not perceivers are likely to possess “theories” about their significant others. A wide range of evidence suggests that this is likely (for a review, see Chen, 1997). For instance, research on close relationships, described in an earlier section, has shown that when faced with negative information about a significant other, perceivers tend to formulate explanations or “theories” to dampen the negativity of the information (Murray & Holmes, 1993, 1994, 1996; see also Andersen et al., 1996; Collins, 1996). By and large, however, theories about significant others have not been precisely defined, nor has their potential role in transference been examined. One line of research has begun to address these issues, adopting a theory-based approach to significant-other representations and transference (Chen, 1997). The basic argument of this work is that perceivers are especially likely to make sense of the significant others in their lives in terms of “theories” about them, and that this should be manifested in content and structural characteristics of significant-other representations. In turn, this should have implications for the nature of applicability-based triggering in transference. We turn now to describing this research, and present initial findings emerging from three studies.
IX. A Theory-Based Approach to Significant-Other Representations and Transference The theory-based approach taken in this research was based in part on the “cognitive-social’’ model of personality and social perception (Mischel & Shoda, 1995; see also Mischel, 1968, 1973, 1990). Countering widely held assumptions about the stable and trait-like nature of personality, this approach posits IF-THEN relations as the basic units of personality-with IFs referring to situations and THENs referring to the responses elicited by them. Such an IF-THEN view of personality has implications for social perception insofar as it suggests that perceivers may at times conceive of others in terms of IF-THEN relations, recognizing that what others think, feel, and do may be contingent on the situation (Shoda & Mischel, 1993). In this regard, like other theory-based approaches, an IF-THEN approach to social perception suggests that perceivers’ conceptions of others may at times include knowledge about causal or explanatory relations-namely, ones between the IFs and THENs of others. In the research described below, theories were conceptualized in such IF-THEN terms (Chen, 1997). Specifically, they were defined as IF-THENs in which the IFs reflect perceivers’ inferences about the psychological situa-
166
SERENA CHEN AND SUSAN M. ANDERSEN
tions or states (e.g., thoughts, feelings, goals) of another person that elicit the person’s responses or THENs (e.g., “IF feeling insecure about himself, THEN Steve is hostile toward others”). In other words, such inferences reflect a perceiver’s consideration of the psychological experiences that come into play for the person in particular situations. Indeed, inferences about another person’s psychological IFs reflect perceivers’ beliefs about, arguably, the most proximal causes of the THENs to which they are linked, and as such, hold distinct explanatory power. Overall, then, this work proposes that perceivers’ conceptions of others may include not only unrelated bits of featural knowledge, but also explanatory knowledge about relations between psychological IFs and THENs. In short, they may include “psychological-state theories”-defined as IF-THEN mini-narratives composed of the psychological states (IFs) thought to elicit others’ responses (THENs). The overriding argument of this theory-based approach to significantother representations and transference is that perceivers are especially likely to possess psychological-state theories about their significant others. This argument was based in part on evidence suggesting that perceivers tend to have more psychological-state knowledge about their significant others relative to nonsignificant others (e.g., Prentice, 1990), providing a larger pool of knowledge from which to draw to construct psychologicalstate theories about significant others. Moreover, it reflects evidence suggesting that perceivers are especially motivated to make sense of significant others in terms of the private experiences of these individuals (Andersen et al., 1998). Significant others are bound up with the self. Hence, understanding the responses of these important individuals directly in terms of the inner workings of their hearts and minds is critical in terms of enabling perceivers to know what to expect and hope for in their relationships with these others, how best to negotiate these relationships, and so forth. Indeed, the construction of psychological-state theories may at times reflect motivated attempts to bolster a perceiver’s sense that a significant other loves him or her (see Murray & Holmes, 1993,1994,1996). In short, perceivers are likely to be especially motivated to make efforts to know and understand certain persons or groups (or believe that they do). Although it might be sufficient to know that “nerds” are smart and skinny, it is unlikely that such simple conceptions would suffice in knowing a significant other. Three studies were conducted to test several hypotheses derived from the present theory-based approach (Chen, 1997). The first two examined the implications of this approach for the content and structure of significantother representations, and the third examined its implications for applicability-based triggering in transference. In all three studies, participants were
SIGNIFICANT-OTHER REPRESENTATIONS A N D TRANSFERENCE
167
asked to identify a significant other, a stereotype, and a nonsignificant other-the latter two serving as control representations. A. THEORY-BASED CONTENT IN SIGNIFICANT-OTHER REPRESENTATIONS The first hypothesis was that psychological-state theories about significant others are especially available in memory, stored as part of significantother representations. Evidence was found for this hypothesis in a first study (Chen, 1997). In this study, participants were first asked to describe each of the three representations by generating IF-THEN statements in which the IFs simply referred to observable situations, and the THENs referred to responses in these situations (e.g., “IF Amy is at work, THEN she is very serious”). Thus, these statements essentially reflected empirical observations previously made about each person or stereotype. Subsequently, participants were asked to exphin those IF-THEN observations they had just listed for which they felt they had an explanation. They provided each explanation by completing a “because . . .” prompt. Two trained judges coded each of these explanations in terms of whether it reflected a psychological-state explanation (e.g., “because she wants to be the best” or “because she feels anxious”) or one that did not refer to a psychological state of the person or stereotype (e.g., “because her boss is strict” or “because it’s a tough job”). The major finding of this study was that participants spontaneously generated a greater number and proportion of psychological-state explanations for the IF-THEN observations they had listed about their significant other relative to the control representations. These explanations essentially reflect psychological-state theories. Thus, these data provide support for the argument that significant-other representations are particularly likely to include such theory-based content. B. THEORY-BASED STRUCTURE IN SIGNIFICANT-OTHER REPRESENTATIONS If perceivers tend t o conceive of significant others in terms of psychological-state theories, the stored linkages between the psychological IFs and THENs about these individuals should be especially strong, implying that activation should spread quickly across them. In other words, the component parts of psychological-state theories (i.e., the IFs and THENs) about significant others should be retrieved in tandem very readily. Said differ-
168
SERENA CHEN AND SUSAN M. ANDERSEN
ently, such theories ought to be especially accessible. To test this hypothesis, participants in a second study generated psychological-state theories on a computer which recorded their generation-latencies (Chen, 1997). They did so by first typing in a response (i.e., THEN) true of the person or stereotype being described, followed by the psychological state (i.e., psychological IF) believed to elicit the response. The results showed that participants not only generated more psychological-state theories about their significant other relative to the control representations, conceptually replicating the major finding of the first study in this line of research, but also exhibited significantly greater ease in corning up with psychological IFs to explain THENs (i.e., shorter mean latency within IF-THEN units) about their significant other. The latter finding supports the accessibility hypothesis, and suggests that there is structure to the psychological-statetheories that perceivers have stored about significant others in the form of tightly-linked IF-THEN units composed of psychological states and the THENs they explain. Indeed, this effect held even when controlling for the mean latency between IF-THEN units-verifying the distinctiveness of the latency within IF-THEN units, and bolstering the interpretation that this latency reflects IF-THEN units stored in memory that, for significant-other representations, involve especially strong IFTHEN linkages. This evidence suggesting theory-based structure in significant-other representations speaks to prior research examining the structural complexity of significant-other representations (Andersen & Cole, 1990, Studies 1 & 2). In this earlier work, significant-other representations were shown to be more featurally distinct and rich than various control representationscharacteristics associated with structural complexity. What the current evidence suggests is that the structural complexity of significant-other representations may be attributable in part to the existence of IF-THEN units of knowledge reflecting psychological-state theories about significant others. Overall, these content and structure findings support the notion that perceivers are especially likely to make sense of their significant other in terms of psychological-state theories about them. That is, theory-based knowledge is especially likely to be stored in memory as part of representations of significant others. Theory-based content and structure in significantother representations should have implications for the nature of applicability-based triggering in transference. C. A THEORY-BASED APPROACH TO APPLICABILITY-BASED TRIGGERING IN TRANSFERENCE If perceivers tend to represent their significant others in terms of psychological-state theories, then cues in a new person that “match” or overlap
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
169
with a perceiver’s theories about a significant other should be especially likely to trigger the activation and application of the relevant significantother representation to the new person. In other words, psychologicalstate theories about significant others should constitute a primary basis of applicability’s role in triggering transference. In broader terms, the cues in a stimulus person that are most likely to render a representation applicable should be those that correspond to the level at which perceivers tend to conceive of the person or group designated by the representation (Hardin & Rothman, 1997). The final study in this line of research examined the hypothesis that applicability-based triggering cues reflecting psychological-state theories about significant others play a distinctly powerful role in eliciting transference (Chen, 1997). To do so, this study used a variant of the basic transference paradigm, and compared the role of theory-based applicability to that of feature-based applicability in triggering transference. More specifically, participants learned about several target persons during the learning phase of this study, one of whom was described by either feature-based or theorybased descriptors that participants had generated in a pretest session to describe a significant other. The other targets resembled various control representations. Thus, as in all of the research described thus far, applicability-based triggering of participants’ representation of a significant other took the form of a target person’s resemblance to this important other. The one exception in this study was that applicability was based on either features or theories associated with the significant other. Thus, in the feature-based applicability condition, participants encountered feature-based stimuli in the learning phase (e.g., “Terry is an intelligent person” or “Pat plays basketball”). In contrast, participants in the theory-based applicability condition were presented with theory-based stimuli reflecting psychologicalstate theories. These theories took the form of responses (i.e., THENs) linked to the psychological states (i.e., psychological IFs) thought to elicit them (e.g., “Pat is shy if she feels insecure” o r “Terry is nice to others when he wants to get ahead”). In both the feature-based and theory-based applicability conditions, participants were later asked to complete a standard recognition-memory test that was used, as usual, to assess the emergence of transference in the form of representation-derived inference and memory based on a significantother representation. The feature-based or theory-based nature of the representation-consistent descriptors used in this memory test corresponded to the nature of the representation-consistent descriptors used as applicabilitybased triggering cues in the learning phase. Regardless of applicability condition-that is, regardless of whether the target descriptors were theory-based or feature-based-greater representa-
170
SERENA CHEN AND SUSAN M. ANDERSEN
tion-derived inference and memory were found for the target resembling participants’ own significant other compared to the control targets, providing evidence for transference in both applicability conditions. In the featurebased condition, this effect replicates prior research using feature-based triggering cues in a target person to elicit transference (e.g., Andersen & Cole, 1990,Study 3; Andersen et al., 1995; Glassman & Andersen, in press). The main hypothesis of this study, however, was that applicability-based triggering involving theory-based significant-other resemblance in the new target person would result in a stronger transference effect than applicability-based triggering involving feature-based significant-other-resemblance. Although some deviations from this strict prediction emerged, the finding of greater representation-derived inference and memory for the target resembling the participant’s own significant other relative to control targets was in fact reliably stronger in the theory-based applicability condition relative to the feature-based condition. Thus, the data suggest that when cues in a new person “match” a perceiver’s psychological-state theories about a significant other, the transference effect is more pronounced, as indexed by representation-derived inference and memory. Taken as a whole, this set of findings extends the social-cognitive model of transference by suggesting that theory-based forms of knowledge play a role in the content, structure, and triggering of significant-other representations. On a broader level, this research parallels the increasing attention being given to applicability in the social-cognitive literature on construct activation and use (for a review, see Higgins, 1996; see also Chen, 1997; Chen et al., in press; Hardin & Rothman, 1997). In our view, applicabilitybased triggering may be especially critical to examine in transference because cues that we pick up in face-to-face encounters with new others are, arguably, those that are most likely to trigger stored, significant-other knowledge. In turn, a growing body of research on transference suggests that the activation and application of this knowledge to new others is then likely to color interpretations of and relations with them, laying the groundwork for the reemergence of relationships from the past in the present. Stated somewhat differently, transference may emerge in its fullest form, and with its greatest consequences, in the context of actual face-toface encounters with another person, who embodies cues that provide grist for the activation and use of particular significant-other representations to interpret and respond to him or her.
X. Finding Common Ground Transference and Close Relationships We began by locating our theorizing and research on significant-other representations and transference in the close relationships literature, yet
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
171
at the same time distinguishing our work as focusing on how past relationships with significant others influence current interpersonal life rather than on the dynamics of present-day close relationships. Nonetheless, a substantial degree of common ground can be found between our work and the broader literature on close relationships. In this section, we highlight some ways in which our research parallels this literature, can contribute to it, and be enriched by it. A. ATTACHMENT THEORY IN CLOSE RELATIONSHIPS A central contention of our work on significant-other representations and transference is that present interpersonal experience can be informed by looking to relationships from the past. In this basic regard, our work is closely allied with the burgeoning body of research on attachment theory, particularly in the realm of social psychology (for reviews, see Feeney & Noller, 1996; Hazan & Shaver, 1994; Simpson & Rholes, 1998). Fundamental to attachment theory is the notion that adult relationships can be understood in terms of early experiences with major attachment figures (Bowlby, 1969,1973,1980), an idea that is obviously consistent with our model of transference. Also central to attachment theorizing, and in accord with our thinking, is the assumption that working models of the self and important others are stored in memory, shaped in part by prior attachment experiences (Bowlby, 1969,1973,1980,see also Bartholomew & Horowitz, 1991; Collins & Read, 1994; Griffin & Bartholomew, 1994; Mikulincer, 1995). Although research grounded in attachment theory has often conceptualized adult attachment orientations or styles essentially as individual differences that can be captured in participants’ responses to self-report items (e.g., Bartholomew & Horowitz, 1991; Hazan & Shaver, 1987), increasing theoretical and empirical precision regarding the content, structure, and processes underlying working models of attachment can be found in the attachment literature (e.g., Baldwin, Fehr, Keedian, Seidel, & Thomsom, 1993; Shaver, Collins, & Clark, 1996). For example, in terms of content, greater specification has emerged regarding attachment-relevant goals and needs (e.g., Collins & Read, 1994), emotions and styles of emotion regulation (e.g., Brennan & Shaver, 1995; Kobak & Sceery, 1988; Mikulincer & Orbach, 1995), as well as interpersonal expectancies (e.g., Baldwin et al., 1993; Downey & Feldman, 1996). Behavioral aspects of working models of attachment have also been more clearly specified in the form of, for example, responses in daily social interactions (e.g., Pietromonaco & Barrett, 1997;see also Downey, Freitas, Michaelis, & Khouri, 1998), as well as support seeking and giving among relationship
172
SERENA CHEN AND SUSAN M. ANDERSEN
partners in anxiety-provoking situations (e.g., Simpson, Rholes, & Nelligan, 1992). Research examining structural components of working models of attachment, as well as the processes that govern when these models are likely to exert an influence on interpretation and responding, is also rapidly expanding, drawing heavily on social-cognitive theory and research (e.g., Baldwin et al., 1993; Baldwin, Keelan, Fehr, Enns, & Koh-Rangarajoo, 1996). Our work on significant-other representations and transference dovetails nicely with much of this growing body of research on attachment (for a further discussion, see Andersen et al., 1997). As described in this chapter, research on transference has in fact demonstrated that motivations, expectancies, and affect experienced previously in relationships with particular significant others (e.g., Andersen et al., 1996), as well as normative role information (Baum & Andersen, in press), are stored in memory as part of “working models” or representations of these important others, as well as the linkages that bind self and significant-other knowledge. When a significant-other representation is activated, these various forms of stored information have been shown to become operative, coloring current social relations with aspects of past relationships with significant others. Inferences and evaluative responses derived from significant-other representations have similarly been shown to be activated and applied to new others in transference (e.g., Andersen & Baum, 1994; Andersen & Cole, 1990, Study 3; Andersen et al., 1995). Moreover, research has demonstrated the chronic readiness with which significant-other representations are activated and used in social perception (Andersen et al., 1995). In the process, this work has provided considerable precision regarding the basic principles that determine how and when transference is triggered (Andersen et al., 1995; see also Chen, 1997; Chen et al., in press). Finally, the activation and application of a significant-other representation to a new person have been shown to accordingly activate relevant selfkgnificant-other linkages stored in memory. This is strongly suggested by research on motivation, expectancies, affect, and interpersonal roles (Andersen et al., 1996; Baum & Andersen, in press), and is demonstrated directly in evidence indicating that the activation of a significant-other representation leads to predictable changes in how the self is experienced (Hinkley & Andersen, 1996). At the same time, unlike attachment research in which participants’ general relationship orientations are typically assessed, our research examines the particular constellation of inferences, motivation, affect, and expectations, as well as roles and selves, associated with specific significant others in people’s lives, and does so using idiographic research methods. As such, our work captures the idiosyncrasies in specific significant relationships in a person’s life rather than focusing on generic relational styles.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
173
This significant-other-specific approach assumes that individuals may have relatively unique definitions of their various significant others, as well as unique patterns of relating with them. And yet, on another level, there is no reason that such idiosyncrasy cannot coexist with more generic differences in relational learning and patterning (Baldwin et al., 1996; cf. A. P. Fiske, 1992; A. P. Fiske & Haslam, 1996). Indeed, recent research on interpersonal roles and transient mood states in transference, described earlier, has demonstrated that idiosyncrasy in selfhignificant-other relations and relational patterns reflecting normative roles may coexist (Baum & Andersen, in press). Attachment research focused on the self (e.g., Baldwin, 1992; Brennan & Morris, 1995; Griffin & Bartholomew, 1994; Mikulincer, 1995) is also compatible with our theorizing and research (Hinkley & Andersen, 1996; see also Andersen et al., 1997). Common to this attachment work and our view of the self in transference is the fundamental notion that the self is defined and evaluated in part in relation to other persons. However, once more, we diverge from much work in this broader literature on attachment by conceptualizing the self in relation to specific significant others, and by examining the self idiographically. In this regard, our approach fits especially well with the increasingly widely accepted notion of the “working self-concept” (for a review, see Linville & Carlston, 1994), or the idea that the self is essentially constructed anew as a function of contextual cues, even though nomothetic propensities may also exist. That is, from our perspective, the self is determined in part by contextual cues that trigger transference-for example, the “context” of a new person who bears some resemblance to a significant other. This is exemplified in the research demonstrating that the activation and application of a significant-other representation to a new person in such a “context” make operative those aspects of the self most relevant to this “context”that is, the relevant self-with-significant-other (Hinkley & Andersen, 1996; see also Andersen et al., 1997). In this same vein, we have recently proposed that the notion that the self is related to or “entangled” with significant others, which results in affective, motivational, and self-definitional consequences when a significant-other representation is activated, provides a model for defining personality in interpersonal terms while highlighting basic processes (Andersen & Chen, 1999). That is, we suggest that personality is defined in part in relationships with significant others (see also Andersen et al., 1997). Thus, the social-cognitive process of transference can be viewed as tracking variability in the self and personality across different interpersonal situations.
174
SERENA CHEN A N D SUSAN M. ANDERSEN
This notion that the self is in part transference-based meshes well with the recently proposed IF-THEN model of personality, also defined in socialcognitive terms (Mischel & Shoda, 1995). In an earlier section, we discussed the implications of this model for social perception-namely, we described research based on the assumption that perceivers may at times conceptualize others in IF-THEN terms. The current focus is on this model’s conceptualization of personality. According to this IF-THEN model, an individual’s personality can be defined in terms of a set of idiosyncratic IF-THEN patterns, or the individual’s personality “signature” across situations. Our model is compatible with this framework. The IFs in our model refer to the situations-especially the particular persons encountered-that trigger particular significant-other representations and associated selfwith-significant-other knowledge. The THENs in our model refer to the perceptual, affective, and behavioral responses elicited in transference (Andersen & Chen, 1999). Thus, as with the more global IF-THEN model of personality (Mischel & Shoda, 1995), our IF-THEN view suggests that transference embodies basic personality processes that capture both variability and continuity. Variability in personality arises as a function of contextual cues (i.e., new people who are encountered), and continuity in personality lies in the chronic accessibility of significant-other representations and the aspects of the self that are strongly associated with these representations that are thus likely to be chronically operative. B. SELF AND OTHER REPRESENTATIONS IN CLOSE RELATIONSHIPS Common ground can also be found in our work on the self in transference and the broader literature on close relationships extending beyond the attachment domain. Indeed, several theoretical frameworks contend, as we do, that self-with-significant-other information is represented in memory (Andersen et al., 1997; Baldwin, 1992; Bugental, 1992; Ogilvie & Ashmore, 1991). Theorizing on “relational schemas” is of particular relevance (for a review, see Baldwin, 1992). Relational schemas are thought to be composed of three elements-the self schema, associated significant-other schemas, and interpersonal scripts reflecting knowledge about expected patterns of interaction between the self and others that have been learned through past interactions with these others. These three elements are thought to be structurally associated in memory such that schematicity in each element and in the entire structure (“conjoint schematicity”) therefore exists. If conjoint schematicity accurately characterizes relational schemas, conjoint
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
175
priming should occur such that activation in one element spreads to the others (e.g., Baldwin et al., 1990, 1993; Baldwin & Holmes, 1987). Notions of conjoint schematicity and conjoint priming depend on the existence of, and spread of activation across, stored selfkignificant-other linkages. In this regard, a relational-schema approach is obviously consistent with our view that selfhignificant-other linkages are stored in memory, and are activated in transference when relevant significant-other representations are activated. Indeed, as indicated, research on transference has demonstrated this both directly (Hinkley & Andersen, 1996) and indirectly (Andersen et al., 1996; Baum & Andersen, in press). Empirically speaking, parallels between relational-schema work and research on transference are perhaps most readily seen in investigations concerned with the impact of selfkgnificant-other linkages on selfevaluation. As described earlier, research on transference has demonstrated not only content shifts in the working self-concept in the direction of the relevant self-with-significant-other in transference, but also self-evaluative changes in line with the relevant self-with-significant-other (Hinkley & Andersen, 1996). Research conducted within a relational-schema approach has shown similar self-evaluative effects (for a review, see Baldwin, 1992). For instance, participants for whom a representation of a parental figure was primed reported less enjoyment of sexually charged written passages than did participants for whom a friend representation was primed, a finding suggesting that the activation of the representation of the familial significant other versus of a friend differentially influenced participants’ personal selfevaluative standards (Baldwin & Holmes, 1987). In another study, psychology graduate students who were subliminally exposed to the approving face of their program director rated their research ideas more favorably than did those who were exposed to the disapproving face of their director (Baldwin et al., 1990). In this same line of work, it was also found that religious, Roman Catholic female participants who were subliminally exposed to the disapproving face of the Pope evaluated themselves less positively than did their counterparts who were exposed to the disapproving face of a nonsignificant other. Also relevant to conceptualizations of the self in transference is recent theory and research in which closeness in relationships is defined in terms of the overlap between representations of self and others (Aron, Aron, Tudor, & Nelson, 1991; Aron, Aron, Smollan, 1992; for a review, see Aron & Aron, 1996). That is, the degree of closeness one experiences with another person is thought to be reflected in the extent to which the other is “included in the self,” as assessed by self-other overlap (e.g., Aron et al., 1991). This “inclusion-of-other-in-the-self ” approach suggests that the self representation can be conceived of as overlapping with multiple representations of
176
SERENA CHEN AND SUSAN M. ANDERSEN
others to varying degrees as a function of closeness, a conceptualization with which we would concur (Andersen et al., 1997; Hinkley & Andersen, 1996). Finally, any consideration of the self in relation to others is likely to call to mind the rapidly expanding body of research concerned with crosscultural variations in construals of the self (e.g., Cousins, 1989; A. P. Fiske, Kitayama, Markus, & Nisbett, 1998; Markus & Kitayama, 1991; Rhee, Uleman, Lee, & Roman, 1995; see also Brewer, 1991; Smith & Henry, 1996). Much of this work has distinguished cultures in which the self is construed largely in relation to others from those in which the self is often construed as independent from these others (e.g., Markus & Kitayama, 1991). From our perspective, variations across cultures in the extent to which the self is defined in relation to others may well have implications for the breadth of the transference experience. That is, although we would argue that significant others, by definition, are closely linked to the self, and thus that changes in self-definition and self-evaluation are very likely to occur in the context of transference irrespective of culture, we speculate that these changes may hold to a greater degree among particular individuals or within particular cultures. Thus, for example, to the extent that the self is tightly bound to others, transference effects implicating the self may be especially pronounced. In broad terms, cross-cultural research clearly suggests the need to recognize that the content, structure, and use of representations of both the self and significant others may vary importantly across cultures, reflecting systematic differences in the ways people of different cultures construe and experience the self in relation to the social world. Accordingly, such variations may correspond to differences in the ways in which relationships from the past are experienced in present relations.
C. THEORIES ABOUT SIGNIFICANT OTHERS As described earlier, the “theories” or explanations perceivers have stored about their significant others have been the focus of a recent line of inquiry (Chen, 1997). This work parallels the increasing recognition that perceivers’ conceptions of objects, people, and events are likely to extend beyond simple lists of features or attributes so as to include explanatory or theory-based forms of knowledge (e.g., Medin, 1989; Murphy & Medin, 1985). Research adopting a theory-based approach to significant-other representations and transference has in fact yielded some evidence suggesting that perceivers are especially likely to formulate psychological-state theories to make sense of significant others, and that these theories play a role in the applicability-based triggering of transference (Chen, 1997). More
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
177
specifically, this work suggests that the content and structure of significantother representations may be in part theory-based, which in turn carries implications for how and when significant-other representations are likely to be activated and applied to newly encountered others. Although this research on theories about significant others initially grew out of a body of work outside the domain of close relationships, its focus on the explanations perceivers construct about their significant others is shared by an expanding body of evidence on attributions and attributional processes in close relationships (e.g., Bradbury & Fincham, 1990; Collins, 1996; Fletcher & Fincham, 1991; Murray & Holmes, 1993,1994,1996). For instance, a highly provocative line of research proposes that theories about significant others may serve to construct or maintain perceivers’ “positive illusions” about their significant others, helping to bolster confidence in their relationships with these important others (e.g., Murray & Holmes, 1993, 1994, 1996; Murray, Holmes, & Griffin, 1996; see also Holmes & Rempel, 1989). As described in earlier sections, this research has shown that when faced with negative information about a romantic partner about whom one feels positively, perceivers tend to construct narratives or theories that diminish, if not eliminate, the threat carried by the negative information. For example, perceivers may transform a romantic partner’s fault into a virtue. Analogous data have emerged in research on transference (Andersen et al., 1996), also described previously. Specifically, research has shown that upon encountering a new target person characterized by some negative descriptors associated with a positively evaluated significant other, participants responded to these descriptors with especially positive facial affect at encoding. The inconsistency of the negative features in relation t o the overall positive tone of the significant-other representation apparently led perceivers to transform the negative into a positive response-arguably a reflection of the influence of their positively valenced “theories” about their significant others. Research examining attributions in the context of marital relationships is also relevant to the idea that perceivers are likely to possess theories about the significant others in their lives. A major finding in this work is that attributional patterns play a predictive role in marital satisfaction (for a review, see Bradbury & Fincham, 1990). In general terms, the tendency to attribute positive behaviors of a spouse to dispositional factors and negative behaviors to situational ones tends to be associated with greater marital satisfaction, whereas the opposite pattern of attributions for the positive and negative behaviors of a spouse tends to be associated with lower marital satisfaction.
178
SERENA CHEN AND SUSAN M. ANDERSEN
Dovetailing with this work is research suggesting that one possible origin of differences in attributional patterns lies in attachment orientation (Collins, 1996). More specifically, this work argues that individuals with different attachment styles are inclined to perceive, interpret, and explain relationship events in ways consistent with the distinct nature of their working models of self and other. Stated somewhat differently, it proposes that working models of the self and others contain particular explanatory tendencies, among other forms of knowledge pertaining to emotional and behavioral responding, that have been developed in part on the basis of past experiences with significant others (see also Downey & Feldman, 1996). Thus, for example, this research has shown that individuals with a preoccupied attachment orientation tend to formulate explanations for hypothetical relationship events reflecting a more negative interpretation of these events, as well as more negative views of the relevant relationship partner, than do individuals with a secure attachment orientation. In turn, such variations in explanatory tendencies have been shown to exert an impact on emotional responding (i.e., distress), as well as self-reported behavioral intentionsfor instance, likelihood to engage in conflict-eliciting behavior (Collins, 1996). The notion that working models of self and others include patterns of explanation implies that transference, which occurs upon the activation and application of a significant-other representation to a new other, may involve the reemergence of explanatory tendencies exhibited in prior relationships with significant others in relations with new others. This is certainly consistent with a theory-based approach to significant-other representations and transference. In turn, the transfer of patterns of explanation may in part predict emotional and behavioral responding in relations with these new others (Collins, 1996; see also Downey & Feldman, 1996). This implies that the emergence of representation-derived affect and behavior in transference may result in part from stored patterns of explanation that have been brought to bear on making sense of events in current relations with a new other. Indeed, the occurrence of representation-derived patterns of explanation may be common to the transference phenomenon-influencing perceptions and interpretations of events in encounters with new others, and representing a fundamental way in which significant relationships from the past are re-created in current social life.
XI. Closing Remarks In this chapter, we described our approach to the study of close relationships by presenting a social-cognitive model of transference and the research
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
179
emerging from it (Andersen et al., 1995, 1996; Andersen & Baum, 1994; Andersen & Cole, 1990; Baum & Andersen, in press: Chen, 1997; Glassman & Andersen, in press; Hinkley & Andersen, 1996). At the core of this model is the notion that transference, the phenomenon whereby significant relationships from the past resurface in the present, can be understood in social-cognitive terms. Namely, the model contends that the activation and application of stored, mental representations of significant others to newly encountered others lie at the root of transference. Throughout, we discussed the basic and multifaceted effects that define the transference phenomenon and emerge in it, as well as the social-cognitive principles that govern it. In more specific terms, we reviewed well-controlled experimental research demonstrating that representation-derived inference and memory about a new person occur on the basis of an activated significant-other representation, as well as representation-derived evaluation, motivation, expectancies, facial affect, working-self-concept changes, and, under some circumstances, transient mood changes. As well, research has begun to address complexities in the content and structure of significant-other representations that carry implications for the triggering of transference. We conclude that transference occurs in everyday social cognition, and may do so rather ubuquitously, rendering examination of significant relationships from the past fundamental to understanding present interpersonal experience.
Acknowledgments The authors would like to thank Mark Zanna and John Holmes for their helpful comments on an earlier version of this chapter.
References Adler, A. (1957). Understanding human nature. New York: Fawcett Premier. (Original work published 1927). Allport, G. (1937). Personalily: A psychology inrerpretafion. New York: Holt, Rinehart, & Winston. Andersen, S. M., & Baum, A. B. (1994). Transference in interpersonal relations: Inferences and affect hased on significant-other representations. Journalof Personality, 62,4,460-497. Andersen, S. M., & Berk, M. S. (1999). Transference in everyday experience: Implications of experimental research for relevant clinical phenomena. Review of General Psychology, 2, 81-120.
180
SERENA CHEN AND SUSAN M. ANDERSEN
Andersen, S. M., & Chen, S. (1999). The entangled self A social-cognitive interpersonal theory of self and personality embodied in transference. Unpublished manuscript, New York University. Andersen, S. M., & Cole, S. W. (1990). “Do I know you?”: The role of significant others in general social perception. Journal of Personality and Social Psychology, 59, 384-399. Andersen, S. M., & Glassman, N. S. (1996). Responding to significant others when they are not there: Effects on interpersonal inference, motivation, and affect. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition (Vol. 3, pp. 262-321). New York: Guilford. Andersen, S. M., Glassman, N. S., Chen, S., & Cole, S. W. (1995). Transference in social perception: The role of chronic accessibility in significant-other representations. Journal of Personality and Social Psychology, 69, 41-57. Andersen, S. M., Glassman, N. S., & Gold, D. (1998). Mental representations of the self, significant others, and nonsignificant others: Structure and processing of private and public aspects. Journal of Personality and Social Psychology, 75, 845-861. Andersen, S. M., & Klatzky, R. L. (1987). Traits and social stereotypes: Levels of categorization in person perception. Journal of Personality and Social Psychology, 53, 235-246. Andersen, S. M., Klatzky, R. L., & Murray, J. (1990). Traits and social stereotypes: Efficiency differences in social information processing. Journal of Personality and Social Psychology, 59, 192-201. Andersen, S. M., Reznik, I., & Chen, S. (1997). The self in relation to others: Cognitive and motivational underpinnings. In J. G . Snodgrass & R. L. Thompson (Eds.), The selfacross psychology: Self-recognition, self-awareness, and the self-concept (pp. 233-275). New York: New York Academy of Science. Andersen, S. M., Reznik, I., & Manzella, L. M. (1996). Eliciting facial affect, motivation, and expectancies in transference: Significant-other representations in social relations. Journal of Personality and Social Psychology, 71, 1108-1129. Armstrong, S. L., Gleitman. L. R., & Gleitman, H. (1983). What some concepts might not be. Cognition, 1.7, 263-308. Aron, E. N., & Aron, A. (1996). Love and expansion of the self The state of the model. Personal Relationships, 3, 45-58. Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships and including other in the self. Journal of Personality and Social Psychology, 60, 241-253. Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of Other in the Self Scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63, 596-612. Asch, S. E. (1946). Forming impressions of persons. Journal of Abnormal and Social Psychology, 41, 258-290. Asch, S. E., & Zukier, H. (1984). Thinking about persons. Journal of Personality and Social Psychology, 46, 1230-1240. Bakan, D. (1966). The duality of human existence. Chicago: Rand-McNally. Baldwin, M. W. (1992). Relational schemas and the processing of information. Psychological Bulletin, 112, 461-484. Baldwin, M. W., Carrell, S. E., & Lopez, D. F. (1990). Priming relationship schemas: My advisor and the Pope are watching me from the back of my mind. Journal of Experimental Social Psychology, 26, 435-454. Baldwin, M. W., Fehr, B., Keedian, E., Seidel, M., & Thomson, D. W. (1993). An exploration of the relational schemata underlying attachment styles: Self-report and lexical decision approaches. Personality and Social Psychology Bulletin, 19, 746-754.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
181
Baldwin, M. W., & Holmes, J. G. (1987). Salient private audiences and awareness of the self. Journal of Personality and Social Psychology, 52, 1087-1098. Baldwin, M. W., Keelan. J. P. R., Fehr. B., Enns, V., & Koh-Rangarajoo, E. (1996). Socialcognitive conceptualization of attachment working models: Availability and accessibility effects. Journal of Personality and Social Psychology, 71, 94-109. Baldwin, M. W., & Sinclair, L. (1996). Self-esteem and “ i f . . , then” contingencies of interpersonal acceptance. Journal of Personalify and Social Psychology, 71, 1130-1 141. Banaji, M. R., Hardin, C.. & Rothman, A. J. (1993). Implicit stereotyping in person judgment. Journal of Personality and Social Psychology, 65, 272-281. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Bargh, J. A. (1990). Auto-Motives: Preconscious determinants of social interaction. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and cognifion (Vol. 2, 93-130). New York: Guilford. Bargh, J. A. (1997). The automaticity of everyday life. In R. S. Wyer, Jr. (Ed.), Advances in social cognition (Vol. 10, pp. 1-61). Mahwah, NJ: Erlbaum. Bargh, J. A,. Bond, R. N., Lombardi, W. L., & Tota, M. E. (1986). The additive nature of chronic and temporary sources of construct accessibility. Journal of Personality and Social Psychology, 50, 869-878. Bargh, J. A,, & Gollwitzer. P. M. (1994). Environmental control of goal-directed action: Automatic and strategic contingencies between situations and behavior. Nebraska Symposium on Motivation, 41, 71-124. Bartholomew. K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226-244. Baum, A,, & Andersen, S. M. (in press). Interpersonal roles in transference: Transient mood effects under the condition of significant-other resemblance. Social Cognition. Baumeister. R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497-529. Baumeister, R. F., & Newman, L. S. (1994). How stories make sense of personal experiences: Motives that shape autobiographical narratives. Personality and Social Psychology Bulletin, 20, 676-690. Berscheid, E. (1994). Interpersonal relationships. Annual Review of Psychology, 45, 79-129. Blatt, S. J. (1990). Interpersonal relatedness and self-definition: Two personality configurations and their implications for psychopathology and psychotherapy. In J. L. Singer (Ed.), Repression and dissociation: Implications for personality theory, psychopathology and health (pp. 299-335). Chicago: University of Chicago Press. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books. Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. New York: Basic Books. Bowlby, J. (1980). Attachmenf and loss: Vol. 3 Loss: Sadness and depression. New York: Basic Books. Bradbury, T. N.. & Fincham, F. D. (1990). Attributions in marriage: Review and critique. Psychological Bulletin, 107, 3-33. Brennan, K. A., & Morris, K. A. (1997). Attachment styles, self-esteem, and patterns of seeking feedback from romantic partners. Personality and Social Psychology Bulletin, 23, 23-31. Brennan, K. A,, & Shaver. P. R. (1995). Dimensions of adult attachment, affect regulation, and romantic attachment. Personality and Social Psychology Bulletin, 21, 267-283. Brewer, M. B. (1991). The social self On being the same and different at the same time. Personality and Social Psychology Bulletin, 17, 475-482.
182
SERENA CHEN AND SUSAN M. ANDERSEN
Bruner, J. S. (1957). Going beyond the information given. In H. E. Gruber, K. R. Hammond. & R. Jessor (Eds.), Contemporary approaches to cognition (pp. 41-60). Cambridge, MA: Harvard University Press. Bugental, D. (1992). Affective and cognitive processes within threat-oriented family systems. In I. E. Sigel, A. McGillicuddy-de Lissi, & J . Goodnow (Eds)., Parental belief systems: The psychological consequences for children (pp. 219-248). Hillsdale, NJ: Erlbaum. Bugental, D. B., Lyon, J. E., Krantz, J., & Cortez, V. (1997). Who’s the boss?: Differential accessibility of dominance ideation in parent-child relationships. Journal of Personality and Social Psychology, 72, 1297-1309. Cantor, N., & Mischel, W. (1977). Traits as prototypes: Effects on recognition memory. Journal of Personality and Social Psychology, 35, 38-48. Cantor, N., & Mischel, W. (1979). Prototypes in person perception. In L. Berkowitz (Ed.), Advances in experimentalsocialpsychology (Vol. 12, pp. 3-52). New York: Academy Press. Carey, S. (1985). Conceptual change in childhood. Cambridge, MA: MIT Press. Chen, S. (1997). Making sense of significant others: “Theories” about significant others and their role in transference. Unpublished doctoral dissertation, New York University, New York. Chen, S., Andersen, S. M., & Hinkley, K. (in press). Triggering transference: Examining the role of applicability in the activation and use of significant-other representations in social perception. Social Cognition. Clark, M. S., & Reis, H. T. (1988). Interpersonal processes in close relationships. Annual Review of Psychology, 39, 609-672. Collins, N. L. (1996). Working models of attachment: Implications for explanation, emotion, and behavior. Journal of Personality and Social Psychology, 71, 810-832. Collins, N., & Read, S. J. (1994). Cognitive representations of attachment: The structure and function of working models. In K. Bartholomew & D. Perlman (Eds.), Advances in personal relationships (Vol. 5, pp. 53-90). London: Jessica Kingsley. Cousins, S. D. (1989). Culture and self-perception in Japan and the United States. Journal of Personality and Social Psychology, 56, 124-131. Deci, E. L., & Ryan, R. M. (1085). Intrinsic motivation and self-determination in human behavior. New York: Plenum. Downey, G . ,& Feldman, S. I. (1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality arid Social Psychology, 70, 1327-1343. Downey, G., Freitas, A. L., Michaelis, B.. & Khouri, H. (1998). The self-fulfilling prophecy in close relationships: Rejection sensitivity and rejection by romantic partners. Journal of Personality and Social Psychology, 75, 545-560. Downey, G., & Khouri, H., & Feldman. S. (1997). Early interpersonal trauma and adult adjustment: The mediational role of rejection sensitivity. In D. Cicchetti & S. Toth (Eds.), Rochester Symposium on Development Psychopathology, Vol. VIlI: The effects of trauma on the development process (pp. 85-1 14). Rochester, NY: University of Rochester Press. Ehrenreich, J.H. (1989). Transference: One concept or many? The Psychoanalytic Review, 76, 1989. Ekman, P. (1992). Are there basic emotions? Psychological Review, 9Y, 550-553. Fairbairn, W. R. D. (1954). Psychoanalytic studies of personality. London: Tavistock. Feeney, J. A., & Noller, P. (1990). Attachment style as a predictor of adult romantic relationships. Journal of Personality and Social Psychology, 58, 281-291. Feeney, J.A., & Noller, P. (1996). Adult attachment. Thousand Oaks, CA: Sage Publications, Inc. Feldman, S., & Downey, G. (1994). Rejection sensitivity as a mediator of the impact of childhood exposure to family violence on adult attachment behavior. Development and Psychopathology, 6, 231 -247.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
183
Fiske, A. P.(1992). The four elementary forms of sociality: Framework for a unified theory of social relations. Psychological Review, 99, 689-723. Fiske, A. P., & Haslam, N. (1996). Social cognition is thinking about relationships. Current Directions, 5, 143-148. Fiske. A. P., Kitayama, S., Markus, H. R.. & Nisbett, R. E. (1998). The cultural matrix of social psychology. In D. T. Gilbert, S. T. Fiske, & G . Lindzey (Eds.), The handbook of social psychology (Vol. 2, pp. 915-981). Boston, MA: McGraw-Hill. Fiske, S. T. (1982). Schema-triggered affect: Applications to social perception. In M. S. Clark & S. T. Fiske (Eds.), Affect and cognition: The 17th annual Carnegie symposium on cognition (pp. 55-78). Hillsdale, NJ: Erlbaum. Fiske, S. T., & Pavelchak, M. (1986). Category-based versus piecemeal-based affective responses: Developments in schema-triggered affect. In R. M. Sorrentino & E. T. Higgins (Eds.). Handbook of motivation and cognition (pp. 167-203). New York: Guilford. Fletcher, G. J. O., & Fincham. F. D. (1991). Attribution processes in close relationships. In G. J. 0. Fletcher & F. D. Fincham (Eds.), Cognition in close relationships (pp. 7-35). Hillsdale, NJ: Erlbaum. Fletcher. G. J. 0..& Fincham, F. D. (Eds.). (1991). Cognition in close relationships. Hillsdale, NJ: Erlbaum. Fletcher, G. J. O., & Fitness, J. (Eds.). (1996). Knowledge structures in close relationships: A social psychological approach. Mahwah, NJ: Erlbaum. Freud, S. (1958). The dynamics of transference. Standard edition (Vol. 12, pp. 99-108). London: Hogarth. (Original work published 1912) Freud, S. (1963). The dynamics of transference. Therapy and technique (pp. 105-115). New York: Macmillan. (Original work published 1912) Gelman, S. A,, & Markman, E. M. (1986). Categories and induction in young children. Cognirion, 23, 183-208. Gentner, D., & Markman, A. B. (1997). Structure mapping analogy and similarity. American Psychologist, 52, 45-56. Gick, M. L., & Holyoak, K. J. (1980). Analogical problem solving. Cognitive Psychofogy, 12, 306-355. Gick, M. L., & Holyoak, K. J. (1983). Schema induction and analogical transfer. Cognitive Psychology, 15, 1-38. Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University Press. Gilovich, T. (1981). Seeing the past in the present: The effect of associations to familiar events on judgments and decisions. Journal of Personality and Social Psychology, 40, 797-808. Glassman, N. S., & Andersen, S. M. (in press). Activating transference without consciousness: Using significant-other representations to go beyond the subliminally given information. Journal of Personality and Sociul Psychology. Glassman, N. S., & Andersen, S. M. (in press). Transference in social cognition: Persistence and exacerbation of significant-other-based inferences over time. Cognitive Therapy and Research. Goldstone, R. L. (1994). The role of similarity in categorization: Providing a groundwork. Cognition, 52, 125-157. Greenberg, J. R., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory. Cambridge, MA: Harvard University Press. Greenberg, J., & Pyszczynski, T. (1985). Compensatory self-inflation: A response to the threat to self-regard of public failure. Journal of Personality and Social Psychology, 49, 273-280. Greenwald, A. G. (1982). Ego task analysis: An integration of research on ego-development and self awareness. In A. Hastorf & A. Isen (Eds.), Cognitive social psychology. New York: Elsevier.
184
SERENA CHEN AND SUSAN M. ANDERSEN
Greenwald, A. G., & Banaji, M. R. (1989). The self as a memory system: Powerful, but ordinary. Journal of Personality and Social Psychology, 67, 430-445. Greenwald, A. G., & Pratkanis, A. R. (1984). The self. In R. S. Wyer, Jr., & T. K. Srull (Eds.), The self and social life (pp. 126-145). New York: McGraw-Hill. Griffin, D., & Bartholomew, K. (1994). Models of self and other: Fundamental dimensions underlying measures of adult attachment. Journal of Personality and Social Psychology, 67, 430-445. Guidano, V. F., & Liotti, G. (1983). Cognitive process and emotional disorders. New York: Guilford. Guisinger, S., & Blatt, S. J. (1994). Individuality and relatedness: Evolution of a fundamental dialectic. American Psychologist, 49, 104-111. Hamilton, D. L. (1979). A cognitive-attributional analysis of stereotyping. In L. Berkowitz (Eds.), Advances in experimental social psychology (Vol. 12, pp. 53-84). New York: Academic Press. Hardin, C. D., & Rothman, A. J. (1997). Rendering accessible information relevant: The applicability of everyday life. In R. S. Wyer, Jr. (Ed.), Advances in social cognition (Vol. 10, pp. 143-156). Mahwah, NJ: Erlbaum. Hastie, R., Schroeder, C., & Weber, R. (1990). Creating complex social conjunction categories from simple categories. Bulletin of the psychonomic society, 28, 242-241. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 51 1-524. Hazan, C., & Shaver, P. R. (1994). Attachment as an organizational framework for research on close relationships. Psychological Inquiry, S(I,J, 1-22. Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. Helgeson, V. S. (1994). Relation of agency and communion to well-being: Evidence and potential explanations. Psychological Bulletin, 116, 412-428. Higgins, E. T. (1987). Self-discrepancy theory: A theory relating self and affect. Psychological Review, 94, 319-340. Higgins, E. T. (1989a). Knowledge accessibility and activation: Subjectivity and suffering from unconscious sources. In J. S. Uleman & J . A. Bargh (Eds.), Unintended thought (pp. 75-123). New York: Guilford. Higgins, E. T. (1989b). Continuities and discontinuities in self-regulatory and self-evaluative processes: A developmental theory relating self and affect. Journal of Personality, 57, 407-444. Higgins, E. T. (1990). Personality, social psychology, and person-situation relations: Standards and knowledge activation as a common language. In L. A. Pervin (Ed.), Handbook of personality (pp. 301-338). New York: Guilford. Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Socialpsychology: Handbook of basic principles (pp. 133-183). New York: Guilford. Higgins, E. T., & Brendl, C. M. (1995). Accessibility and applicability: Some “activation rules” influencing judgment. Journal of Experimental Social Psychology, 31, 218-243. Higgins, E. T., & King, G . A. (1981). Accessibility of social constructs: Information processing consequences of individual and contextual variability. In N. Cantor & J. F. Kihlstrom (Eds.), Personality, cognition and social interaction (pp. 69-121). Hillsdale, NJ: Erlbaum. Higgins, E. T., King, G. A., & Mavin, G. H. (1982). Individual construct accessibility and subjective impressions and recall. Journal of Personality and Social Psychology, 43,35-47. Higgins, E. T., Rholes, W. S., & Jones, C. R. (1977). Category accessibility and impression formation. Journal of Experimental Social Psychology, 13, 141-154.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
185
Higgins, E. T., & Stangor. C. (1988). The change of standard effect in person memory. Journal of Personality and Social Psychology, 54, 181-192. Hinkley, K., & Andersen, S. M. (1996). The working self-concept in transference: Significantother activation and self-change. Journal of Personality and Social Psychology, 71, 12791295. Hirschfeld, L. A. (1996). Race in rhe making. Cambridge, MA: The MIT Press. Holmes, J. C., & Rempel, J. K. (1989). Trust in close relationships. Review of Personality and Social Psychology, 10, 187-219. Holyoak, K. J., & Thagard, P. (1995). Menial leaps: Analogy in creative thought. Cambridge. MA: The MIT Press. Horney, K. (1939). New ways in psychoanalysis. New York: Norton. Horney, K. (1945). Our inner conflicts. New York: Norton. Horowitz, M. J. (Ed.). (1991). Person schenzas and maladaptive interpersotlalpatterns. Chicago: The University of Chicago Press. Johnson, M. K., & Raye, C. L. (1981). Reality monitoring. Psychological Review, 88, 67-85. Johnson, M. K., Hastroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological Bulletin, 88, 67-85. Judd, C. M., & Park, B. (1988). Outgroup homogeneity: Judgments of variability at the individual and group levels. Journal of Personality and Social Psychology, 54, 778-788. Keil. F. C. (1989). Concepts, kinds, and cognitive rievelopment. Cambridge, MA: MIT Press. Kelly, G. A. (1955). The psychology of personal constructs. New York: Norton. Kobak, R. R., & Sceery, A. (1988). Attachment in late adolescence: Working models, affect regulation, and representations of the self and others. Child Development, 5Y, 135- 146. Komatsu, L. K. (1992). Recent views of conceptual structure. Psychological Bulletin, 112, 500-526. Kruglanski, A. W., & Jaffe, Y. (1988). Curing by knowing: The epistemic approach to cognitive therapy. In L. Y. Abramson (Ed.), Social cognition and clinical psychology: A synthesis (pp. 254-291). New York Guildford Press. Kunda, Z., Miller, D. T., & Claire, T. (1990). Combining social concepts: The role of causal reasoning. Cognitive Science, 14, 55 1-577. Kunda. Z.. & Thagard, P. (1996). Forming impressions from stereotypes, trait, and behaviors: A parallel constraint satisfaction theory. Psychological Review, 103, 284-308. Lazarus, R. S.. & Averill. J. R. (1972). Emotion and cognition with special reference to anxiety. In C. D. Spielberger (Ed.), Anxiety: Current trends in theory and research (Vol. 2, pp. 242-283). New York: Academic Press. Lewicki. P. (1985). Nonconscious biasing effects of single instances on subsequent judgments. Journal of Personality and Social Psychology, 48, 563-574. Lewicki, P. (1986). Nonconscious social information processing. San Diego, CA: Academic Press. Leyens, J.. & Fiske, S. T. (1994). Impression formation: From recitals to symphonie fantastique. In P. G. Devine. D. L.. Hamilton. & T. M. Ostrom (Eds.), Social cognition: Impact on social psychology (pp. 39-75). San Diego, CA: Academic Press. Linville, P. W., & Carlston, D. E. (1994). Social cognition of the self. In P. G. Devine, D. C . Hamilton, T. M. Ostrom (Eds.). Social cognition: Impact on social psychology (pp. 143193). New York: Academic Press. Linville, P. W., & Fischer, G. W. (1993). Exemplar and abstraction models of perceived group variability and stereotypicality. Social Cognition, 11, 95-125. Logan, G. D. (1988). Toward an instance theory of automatization. Psychological Review, 95. 492-527.
186
SERENA CHEN AND SUSAN M. ANDERSEN
Lombardi, W. J., Higgins, E. T., & Bargh, J. A. (1987). The role of consciousness in priming effects on categorization: Assimilation versus contrast as a function of awareness of the priming task. Personality and Social Psychology Bulletin, 13, 411-429. Luborsky, L., & Crits-Cristoph, P. (1990). Understanding transference: The CCRT method. New York: Basic Books. Manis, M., Paskewitz, J . , & Cotler, S. (1986). Stereotypes and social judgment. Journal of Personality and Social Psychology, 50, 461-473. Markman, A. B., & Gentner, D. (1993). Structural alignment during similarity comparisons. Cognitive Psychology, 25, 431-467. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253. Markus, H., & Wurf, E. (1987). The dynamic self-concept: A social psychological perspective. Annual Review o,f Psychology, 38, 299-337. Martin, L. L. (1986). Set-reset: Use and disuse of concepts in impression formation. Journal of Personality and Social Psychology, 51, 493-504. Martin, L. L., Seta, J. J., & Crelia, R. A. (1990). Assimilation and contrast as a function of people’s willingness and ability to expend effort in forming an impression. Journal of Personality and Socia[ Psychology, 59, 27-37. Martin, L. L., Tesser, A,, & McIntosh, W. D. (1993). Wanting but not having: The effects of unattained goals on thoughts and feelings. In D. M. Wegner & J. W. Pennebaker (Eds.), The handbook of mental control (pp. 552-572). New York: Prentice Hall. McAdams, D. P. (1985). Power, intimacy, and the life story: Personological inquiries into identity. New York: Guilford. McAdams, D. P. (1989). Intimacy: The need to be close. New York: Doubleday. Medin, D. L. (1989). Concepts and conceptual structure. American Psychologist, 44, 1469-1481. Medin, D. L., Goldstone, R. L., & Gentner, D. (1993). Respects for similarity. Psychological Review, 100, 254-278. Mikulincer, M. (1995). Attachment style and the mental representation of the self. Journal of Personality and Social Psychology, 69, 1203-1215. Mikulincer, M., & Orbach, I. (1995). Attachment styles and repressive defensiveness: The accessibility and architecture of affective memories. Journal of Personality and Social Psychology, 68, 917-925. Mischel, W. (1968). Personality and assessment. New York: Wiley. Mischel, W. (1973). Toward a cognitive-social reconceptualization of personality. Psychological Review, 80, 252-283. Mischel, W. (1990). Personality dispositions revisited and revised: A view after three decades. In L. A. Pervin (Ed), Handbook ofpersonality: Theory and research (pp. 111-134). New York: Guilford. Mischel, W.. & Shoda, Y. (1995). A cognitive-affective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246-268. Mullahy, P. (1970). Psychoanalysis and interpersonal Psychiatry. The Contributions of Harry Stack Sullivan. New York: Science House. Mullen, B. (1991). Group composition, salience, and cognitive representations: The phenomonology of being in a group. Journal of Experimental Social Psychology, 27, 297-323. Murphy, G. L., & Medin, D. L. (1985). The role of theories in conceptual coherence. Psychological Review, 92, 289-316. Murray, S. L., & Holmes, J. G. (1993). Seeing virtues in faults: Negativity and the transformation of interpersonal narratives in close relationships. Journal of Personality and SocIal P~ychology,65, 707-722.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
187
Murray, S. L., & Holmes, J. G. (1994). Storytelling in close relationships: The construction of confidence. Personality and Social Psychology Bulletin, 20, 650-663. Murray, S. L., & Holmes. J. G. (1996). The construction of relationship realities. In G. J. 0. Fletcher & J . Fitness (Eds.), Knowledge structures in close relationships: A social psychological approach (pp. 91-120). Mahwah, NJ: Erlbaum. Murray, S. L., Holmes. J. G., & Griffin, D. W. (1996). The self-fulfilling nature of positive illusions in romantic relationships: Love is blind, but prescient. Journal of Personality and Social Psychology, 71, 1155-1 180. Nosofsky, R. M. (1986). Attention, similarity, and the identification-categorization relationship. Journal of Experimental Psycliology: Geiieral, 115, 39-57. Oakes, P. J., Turner, J. C., & Haslam, S. A. (1991). Perceiving people as group members: The role of fit in the salience of social categorizations. British Journal of Social Psychology. 30, 125-144. Oatley, K.. & Bolton, W. (1985). A social-cognitive theory of depression in reaction to life events. Psychological Review. 92, 372-388. Ogilvie, D. M., & Ashmore, R. D. (1991). Self-with-other representation as a unit of analysis in self-concept research. In R. C. Curtis (Ed.), The relational sel) Theorerical convergences in psychoanalysis and social psychology (pp. 282-314). New York: Guilford Press. Ortony. A. (1979). Beyond literal similarity. Psychological Review, 86, 161-180. Park, B., & Hastie. R. (1987). Perception of variability in category development: Instanceversus abstraction-based stereotypes. Journal vf Personality and Social Psychology, 53, 621-635. Park, B., DeKay. M. L., & Kraus, S.(1994). Aggregating social behavior into person models: Perceiver-induced consistency. Journal of Personality and Social Psychology, 66, 437-459. Pavelchak, M. (1989). Piecemeal and category-based evaluation: An idiographic analysis. Journal of Personality and Social Psychology, 56, 354-363. Pietromonaco, P. R., & Barrett, L. F. (1997). Working models of attachment and daily social interactions. Joitrnal of Personality and Social Psychology, 73, 1409-1423. Pelham, B. W. (1993). The idiographic nature of human personality: Examples of the idiographic self-concept. Journal of Personality and Social Psychology, 64, 665-677. Pennington. N., & Hastie, R. (1992). Explaining the evidence: Tests of the story model for juror decision making. Journal of Personality and Social Psychology, 62, 189-206. Pennington, N., & Hastie. R. (1993). Reasoiling in explanation-based decision making. Cognition, 49, 123-163. Pervin, L. A. (1976). A free-response description approach to the analysis of person-situation interaction. Journal of Personality and Social Psychology. 64, 665-677. Petty, R. E., & Wegener, D. T. (1993). Flexible correction processes in social judgment: Correcting for context-induced contrast. Journal of Experimental Social Psychology, 29, 137-165. Prentice, D. (1990). Familiarity and differences in self- and other-representations. Journal of Personality and Social Psychology, 59, 369-383. Read, S. J., & Cesa, I. L. (1991). This reminds me of the time when . . . : Expectation failures in reminding and explaining. Journal of Experimental Social Psychology, 27, 1-25. Read, S.J., & Miller, L. C. (1993). Rapist or "regular guy": Explanatory coherence in the construction of mental models of others. Personality and Social Psychology Bulletin, 19, 526-541. Read, S. J.. Vanman, E. J.. &Miller, L. C. (1997). Connectionism. parallel constraint satisfaction processes. and gestalt principles: (Re)introducing cognitive dynamics to social psychology. Personality and Social Psychology Review, I , 26-53.
188
SERENA CHEN AND SUSAN M. ANDERSEN
Rhee, E., Uleman, J. S., Lee, H., & Roman, R. (1995). Spontaneous self-descriptions and ethnic identities in individualistic and collectivistic cultures. Journal of Personality and Social Psychology, 69, 142-152. Rips, L. J. (1991). Similarity and the structure of categories. In D. J. Napoli & J. A. Keg1 (Eds), Bridges between psychology and linguistics: A Swarthmore festschriff for Lila Gleitman (pp. 35-53). Hillsdale, NJ: Erlbaum. Rips, L. J., & Collins, A. (1993). Categories and resemblance. Journal of Experimental Psychology: General, 122, 468-486. Rogers, C. (1951). Client-centered therapy. Boston: Houghton-Mifflin Rosch, E. (1978). Principles of categorization. In E. Rosch & B. B. Lloyd (Eds.), Cognition and categorization (pp. 27-48). Hillsdale, NJ: Erlbaum. Ross, B. H., & Bradshaw, G . L. (1994). Encoding effects of reminding. Memory and Cognition, 22, 591-605. Ross, B. H., Perkins, S. J., & Tenpenny, P. L. (1990). Reminding-based category learning. Cognitive Psychology, 22, 460-492. Safran, J. D. (1990a). Toward a refinement of cognitive therapy in light of interpersonal theory: I. Theory. Clinical Psychology Review, 10, 87-105. Safran, J. D. (1990b). Toward a refinement of cognitive therapy in light of interpersonal theory: 11. Practice. Clinical Psychology Review, 10, 107-121. Safran, J. D., & Segal, Z. V . (1990). Interpersonal processes in cognitive therapy. New York: Basic Books. Schank, R. C. (1982). Dynamic memory: A theory of reminding and learning in computers and people. New York: Cambridge University Press. Schneider, D. J. (1973). Implicit personality theory: A review. Psychological Review, 79, 294-309. Schneider, D. J., & Blankmeyer, B. L. (1983). Prototype salience and implicit personality theories. Journal of Personality and Social Psychology, 44, 112-722. Schwarz, N., & Bless, H. (1991). Constructing reality and its alternatives: An inclusion/exclusion model of assimilation and contrast effects in social judgment. Sedikides, C., & Skowronski, J . J. (1990). Toward reconciling personality and social psychology: A construct accessibility approach. Journal of Social Behavior and Personality, 5, 531-546. Sedikides, C., & Skowronski, J. J. (1991). The law of cognitive structure activation. Psychology inquiry, 2, 169-184. Seifert, C. M., McKoon, G., Abelson, R. P., & Ratcliff, R. (1986). Memory connections between thematically similar episodes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 12, 220-231. Shaver, P. R., Collins, N., & Clark, C. L. (1996). Attachment styles and internal working models of self and relationship partners. In G. J. 0.Fletcher & J. Fitness (Eds.), Knowledge structures in close relationships: A social psychological approach (pp. 25-61). Mahwah, NJ: Erlbaum. Shaver, P., & Rubenstein, C. (1980). Childhood attachment experience and adult loneliness. In L. Wheeler (Ed.), Review of personality and social psychology (Vol. 1, pp. 42-73). Beverly Hills, CA: Sage. Shoda, Y., & Mischel, W. (1993). Cognitive social approach to dispositional inferences: What if the perceiver is a cognitive social theorist'? Personality and Social Psychology Bulletin, 19, 574-585. Simpson, J. A., & Rholes, W. S . (Eds.). (1998). Attachment theory and close relationships. New York: Guilford. Simpson, J. A,, Rholes, W. S., & Nelligan, J. S. (1992). Support seeking and support giving within couples in an anxiety-provoking situation: The role of attachment styles. Journal of Personality and Social Psychology, 70, 893-912.
SIGNIFICANT-OTHER REPRESENTATIONS AND TRANSFERENCE
189
Singer, J. L. (1988). Reinterpreting the transference. In D. C. Turk & P. Salovey (Eds), Reasoning, interference, and judgnient in clinical psychology (pp. 182-205). New York: The Free Press. Smith, E. E., & Medin, D. L. (1981). Categories and concepts. Cambridge, MA: Harvard University Press. Smith, E. R., & Henry, S. (1996). An in-group becomes part of the self: Response-time evidence. Personality and Social Psychology Bulletin, 22, 635-642. Smith, E. R., & Zarate, M. A. (1990). Exemplar and prototype use in social categorization. Social Cognition, 8, 243-262. Smith, E. R., & Zarate, M. A. (1992). Exemplar-based model of social judgment. Psychological Review, 99, 3-21. Srull, T. K., & Wyer. R. S.. Jr. (1979). The role of category accessibility in the interpretation of information about persons: Some determinants and implications. Joirrnal of Personality and Social Psychology, 37, 1660-1672. Srull, T. K., & Wyer, R. S.. Jr. (1980). Category accessibility and social perception: Some implications for the study of person memory and interpersonal judgment. Journal of Personality and Social Psychology, 38, 84 1-856. Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimentalsocialpsychology (Vol. 21, pp. 261-302). New York: Academic Press. Sullivan, H. S. (1940). Conceptions of’niodern psychiatry. New York: Norton. Sullivan, H. S. (1953). The interpersonal theory ofpsychiatry. New York: Norton. Taylor, S. E. (1981). Categorization approach to stereotyping. In D. L. Hamilton (Ed.). Cognitive processes in stereotyping and intergroup behavior (pp. 83-1 14). Hillsdale, NJ: Erlbaum. Taylor, S. E.. & Brown, J. D. (1988). Illusion and well-being: A social psychological perspective on mental health. Psychological Bulletin, 103, 193-210. Taylor, S. E.. & Lobel, M. (1989). Social comparison activity under threat: Downward evaluation and upward contacts. Psyctiological Review, 96, 569-575. Tesser, A. (1988). Toward a self-evaluation maintenance model of social behavior. In L. Berkowitz (Ed.), Advances in experimental socinlpsychology (Vol. 21, pp. 181-227). New York: Academic Press. Thagard, P. (1989). Explanatory coherence. Behavioral and Brain Sciences, 12, 435-502. Tversky, A. (1977). Features of similarity. Psychological Review, 84, 327-352. Tversky, A,, & Gati, 1. (1978). Studies of similarity. In E. Rosch & B. B. Lloyd (Eds.), Cognition and categorization (pp. 79-98). Hillsdale, NJ: Erlbaum. Wachtel. P. L. (1981). Transference, schema. and assimilation: The relevance of Piaget to the psychoanalytic theory of transference. The Annual of Psychoanalysis, 8, 59-76. Westen, D. (1988). Transference an information processing. Clinical Psyctiology Review, 8, 161-179. Williamson, G . M., & Clark. M. S. (1992). Impact of desired relationship type on affective reactions to choosing and being required to help. Personcdify and Social Psychology Bulletin, IS, 10-18. Williamson, G. M., Clark, M. S., Pegalis, L. J.. & Behan, A. (1996). Affective consequences of refusing to help in communal and exchange relationships. Personality and Social Psychology Bulletin, 22, 34-47. Wisniewski, E. J., & Medin, D. L. (1994). On the interaction of theory and data in concept learning. Cognitive Science, 28, 221-281. Wittenbrink, B., Gist, P. L., & Hilton, J. L. (1997). Structural properties of stereotypic knowledge and their influences on the construal of social situations. Journal of Personality and Social Psychology, 72, 526-543.
190
SERENA CHEN AND SUSAN M. ANDERSEN
Wittenbrink, B., Hilton, J. L., & Gist, P. L. (1998). In search of similarity: Stereotypes as naive theories in social categorization. Social Cognition, 16, 31-55. Wyer, R. S., & Jr., Srull, T. K. (1986). Human cognition and its social context. Psychological Review, 93, 322-359. Zukier, H. (1986). The paradigmatic and narrative modes in goal-guided inference. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 465-502). New York: Guilford.
THE PUZZLE OF CONTINUING GROUP INEQUALITY: PIECING TOGETHER PSYCHOLOGICAL, SOCIAL, AND CULTURAL FORCES IN SOCIAL DOMINANCE THEORY Felicia Pratto
Surely social psychologists research prejudice, discrimination, stereotypes, and intergroup violence because we are concerned that these may sustain social inequality. Yet none of the extant work in these areas directly addresses the question of why inequality persists. If our research and theories do not explain persistent inequality, they leave a gaping hole in scientific understanding of the human condition, for group-based dominance hierarchies are the most common form of social organization among human societies, including all empires and former empires and nearly all nonindustrialized societies (Murdock, 1956). In fact, all societies that have economic surplus are group-based dominance hierarchies in which one ethnic group, race, caste, or class enjoys more power, status, and resources than at least one other and in which men dominate women (Brown, 1991, pp. 137-139; Lenski, 1984). Answering the question of why group dominance persists is practically important and theoretically central if we are to understand the context and consequences of stereotypes, group identity, prejudice, and discrimination. Fortunately, the considerable work on these topics enables me to take a larger perspective than do most theories designed to address one phenomenon. In fact, to encompass the range of phenomena relevant to group dominance, from individual psychologies to societal organization, I use a systems-dynamic approach. Rather than restricting myself to one research method, level of analysis, explanatory style, or phenomenon, then, I expand upon previous work to fit together the pieces of the puzzle of group dominance in a dynamic, integrative theory. Examined in isolation, the various pieces of this puzzle are, well, puzzling: identifiable as distinct in form, yet unclear what they mean because they fit no pattern and are jagged around the edges. Once this puzzle is assembled, however, the significance of each ADVANCES IN EXPERIMENTAL SOCIAL PSYCHOLOGY. VOL 31
191
Copyright 0 1Y?Y by Academic P r e s All rights of ruproduclioii in anv lorm rcrerved m s - ~ m i / ? s9i o 00
192
FELICIA PRATTO
piece will become clear because of how it fits with the surrounding pieces, and the whole dynamic picture of group dominance will be revealed. The aesthetic of social dominance theory is thus its integration: the assembly of different parts into a working whole. Before assembling the puzzle, I will first take the pieces out of the box and examine them so that I can show how they fit together. The next sections summarize robust and important findings from individual prejudice, social ideologies, and institutional discrimination, emphasizing their relationships. After introducing a new piece of the puzzle, social dominance orientation, I will describe research derived from social dominance theory that shows how these pieces fit together.
I. Individual Prejudice The first pieces of the puzzle are psychological orientations that feed prejudice and discrimination. These pieces usually fit between “cultural beliefs” and “intergroup politics.” For example, in the context of a struggle for group dominance, namely the Nazi takeover of Europe, Adorno, Frenkel-Brunswik, Levinson, and Sanford (1950) postulated that there was a general psychological orientation or “mentality” that would unify individuals’ social, economic, and political convictions, including group prejudices, political-economic conservatism, and right-wing political party support. Considerable research has borne out this postulate. Adorno et al. (1950) found that Americans who were prejudiced against “Negroes” also expressed prejudice against Jews, “Okies” (poor Oklahomans fleeing the dust bowl), Japanese, and other denigrated groups. Further, they found that the motivation to denigrate such groups more parsimoniously explained the correlations among separate prejudices than adherence to particular principles did. For example, two subscales of the anti-Semitism scale assess contradictory ideas: that Jews are too uninvolved in society, and that Jews have too much influence in society. However, these two subscales correlated highly and positively with one another ( r = .75, Adorno et al., 1950, p. 75). Such results were initially dismissed for methodological reasons: high correlations within and among prejudice scales might be due to “yeasaying” to unbalanced scales (see Christie & Cook, 1958; Duckitt, 1989; Kirscht & Dillehay, 1967, for reviews). However, Altemeyer (1981, 1988) developed balanced scales to measure Right-Wing Authoritarianism (RWA) that are highly internally reliable, valid among a number of kinds of respondents, and also correlate strongly with balanced measures of preju-
SOCIAL DOMINANCE THEORY
193
dice against many groups. A different, essentially ideological criticism of authoritarian research was leveled by Shils (1954), who decried authoritarian personality theory (APT) for exploring authoritarianism on the right but not on the left. Rokeach (1960) postulated that dogmatism described the attitudinal and cognitive rigidity underlying authoritarianism and that it would be found among extremists on the left and on the right. But neither Shils’s (1954) nor Rokeach’s (1960) claims can be substantiated. Adorno et al. (1950, p. 275) did postulate that authoritarianism and rigidity might be found among leftists, but empirically, the APT team found that conservatism but not liberalism correlated with fascist tendencies. Likewise, Altemeyer (1981,1988,1996) has repeatedly tried to develop measures of Left-Wing Authoritarianism, but he finds exceedingly few people who can be classified as Left-Wing Authoritarians, and even they do not behave like Right-Wing Authoritarians in being intolerant of out-groups, deferential to their own authorities, or in believing that their political opposition should be suppressed (see Altemeyer, 1996, pp. 216-234). In a major review, Stone (1980) also found no hint of dogmatism among leftists, though again, rightists were consistently dogmatic (see also Stone & Smith, 1993). Further, Rokeach’s dogmatism scale usually correlates positively with measures of authoritarianism and conservatism (see Ray, 1973, for a review). A separate, large international body of empirical research on social and political attitudes has substantiated both general findings found in the original authoritarian research: (1) that prejudice against various stigmatized groups tends to correlate, and (2) that group prejudice tends to correlate with conservative political affiliations and support for conservative political policies, such as religious orthodoxy, harsher treatment of prisoners, capital punishment, and going to war (e.g., Eysenck, 1944, 1951, 1971, 1976; Kerlinger, 1984; McClosky, 1958; Nias, 1972; Ray, 1983; Sidanius, 1987; Sidanius & Ekehammar, 1979; Sidanius, Ekehammar, & Ross, 1979; Sirgo & Eisenman, 1990; G. D. Wilson & Bagley, 1973, G. D. Wilson & Patterson, 1968). Such attitudes amount to more than feelings: they pertain directly to political activity. For example, elected North American legislators from more conservative political parties are higher on RWA than legislators from liberal parties (Altemeyer, 1996, pp. 258-298). Such research linking individual attitudes and their psychological substrates to political activity helps us join individual prejudice to political power in the puzzle of group dominance. Milton Rokeach worked toward that end by researching the values that underlie both sociopolitical attitudes and political party activism. Rokeach (1973) hypothesized that communism values equality but not freedom, socialism values both equality and freedom, fascism values neither, and
194
FELICIA PRAITO
capitalism values freedom but not equality. Content analyses of large samples of writings of apologists for each political ideology (e.g., Hitler, Goldwater, Lenin) showed that the rankings of freedom and equality values followed the predicted pattern, and that these two values distinguished among the writings better than many other values (Rokeach, 1973, pp. 171-179). Similarly, Cochrane, Billig, and Hogg (1979) found that membership and activism in the four types of political parties among the British public corresponded to their relative rankings of freedom and equality. Among British Members of Parliament, Searing (1979) found that 85% of capitalists (Conservative Party members) ranked freedom above social equality, whereas 79% of socialists (Labour Party members) ranked social equality above freedom. As in research on authoritarianism and conservatism, Rokeach’s two-value theory joins individual psychological motivations to the structural and political aspects of group dominance. This linkage between individual motivation and aggregate phenomena is essential for constructing a dynamic theory of society. Moreover, without it, personality variables can make no contribution to explaining a society’s level of intergroup discrimination if that society has normatized and institutionalized discriminatory social practices and beliefs (e.g., see Pettigrew, 1958; van den Berghe, 1967, pp. 20-21). Of the two terminal values Rokeach’s theory identified, equality has proven more significant in understanding political attitudes and behavior. In the American public, political differences such as self-identified liberalismconservatism, political party, and presidential candidate preference vary only along the equality dimension (Rokeach, 1973, pp. 180-184). Freedom is more consensually valued, perhaps because it is construed to mean different things v i s - h i s equality across the political spectrum (see Rokeach, 1973, p. 184).’ Likewise, in the British public, only rankings of equality statistically distinguish among the four parties (Cochrane et al., 1979). Among British members of Parliament (MPs), equality and freedom both appear to distinguish the political parties, but freedom means different things to the two parties (Searing, 1979, p. 173). Several other social theories also hint that one must examine whether social relationships are egalitarian or hierarchical. However, none of them have fully developed how this dimension relates to intergroup relations. APT came closest to considering hierarchical relationships among social groups in its concept of ethnocentric ideology (Adorno et al., 1950, p. 150). However, most of the theory and empirical focus of APT concerns relations I Rokeach argued that for conservatives, freedom, or lack of constraints on individuals, is necessary for them to achieve superior status, wealth, and power; hence freedom is incompatible with equality. For socialists, however, Rokeach argued that freedom for all, and thus equality, could not be achieved without some constraints on individual initiative.
SOCIAL DOMINANCE THEORY
195
between an individual and an authority within a family or nation, that is, intragroup rather than intergroup phenomena (cf. Duckitt, 1989, but see Altemeyer, 1998). Similarly, several cultural analyses of human relations note the importance of hierarchical power relations in organizing human societies, but these also focus on relations between individuals rather than between groups (e.g., Hofstede, 1982, on power-distance; A. P. Fiske, 1991, on authority relations; Schwartz, 1994, on personal power). Glenn Wilson (1973, p. 22) mentioned group dominance as underlying the correlations among conservative attitudes, but his theory and research focused on individuals' fear of uncertainty. McClosky (1958) showed that people holding conservative social and political attitudes were also interpersonally dominant, but this finding has received little further attention. Because social dominance theory attempts to explain group-based dominance, it examines individuals' psychological orientations towards group dominance versus group equality. This dimension is called social dominance orientation, and some of the research I review below will show how this individual orientation toward prejudice relates to the other pieces of group dominance. A. SOCIAL IDEOLOGIES To assemble the puzzle of group dominance, one must link individual psychological processes and behavior to institutional discrimination, social structure, and culture. Social dominance theory does this through social ideologies. Our theory of ideologies melds several disparate traditions: the elite theory (Michels, 1911/1962;Mosca, 1896/1939;Pareto, 1901/1979) and Marxist (Marx & Engels, 1848/1970) functional view of ideology, cultural analyses linking people's behavioral patterns and social structure to their beliefs (Johnson. 1994; Sanday, 1981), theories of social representations and discourse analysis that emphasize the collective and communicative aspects of social ideologies (e.g., van Dijk, 1987; Wetherell & Potter, 1993), and the social cognitive tradition, which identifies the mutual influences of basic psychological processes and the social environment. To demonstrate that social ideologies link individual psychological processes to institutional discrimination, social structure, and culture, my review need only focus on one arena: stereotyping. This is because stereotypes can lead to discrimination, can stem from social structure, and reflect the social relationships and practices that constitute cultures.* By concentrating this review on the 'For reviews of this literature with different theoretical agendas, see Augoustinos and Walker (1996); S. T. Fiske (1997); Hilton and von Hippel (1996); Leyens, Yzerbyt, and Schadron (1 994).
196
FELICIA PRA’ITO
social-cognitive approach to stereotyping inspired by Allport (1954), I will illustrate how basic psychological processes and the social environment of group dominance create and sustain each other. Following this truncated review, I elaborate social dominance theory’s more general theory of ideology. A number of social-cognitive theories describe how aspects of the social context of group dominance instigate psychological processes that cause people to form, use, and validate stereotypes. For example, Hamilton and Gifford (1976) showed that the process of forming illusory correlations can cause people to “learn” negative stereotypes because people make associations between unusual (negative) features and unusual people, such as ethnic min~rities.~ Note, however, that the psychological perceptual process that underlies illusory correlations is not sufficient in and of itself to explain stereotypes; one must assume that group membership is a salient feature of the social environment. Given salient group memberships, Hamilton and Gifford’s (1976) experiment showed how stereotypes could be formed even if the groups do not have different features. When groups do have different features, the process of learning real covariations, which people do readily (Lewicki, Hill, & Czyzewska, 1992; see also Jussim & Eccles, 1995) could also cause people to form stereotypes. A child who sees that most child caregivers are women is likely, then, to learn to expect such features to go together, forming a rudimentary stereotype of women or of caregivers. Another basic psychological process that acts upon social context is attribution. In contexts where groups occupy different social positions, the penchant to attribute to individual dispositions, ironically, may lead to the belief in group stereotypes (see Hamilton, 1979, Pettigrew, 1979). For example, Levine and Campbell (1972) suggested that outgroups who, when excluded from other economic roles play the role of merchants and financial middlemen, as have Jews in Europe and Asians in Africa, would be stereotyped in negative terms relevant to their roles, such as “miserly.” Eagly and Steffen (1984) showed that when participants read about a fictitious society in which one group did nurturing tasks and the other group did agentic tasks, participants formed group stereotypes whose contents were remarkably similar to modern American gender stereotypes. These findings Illusory correlation best explains the formation of stereotypes of minority groups, but obviously there are stereotypes about stigmatized nonminority groups as well (e.g., women, South African Blacks, Indian untouchables). The illusory correlation bias could help explain the existence of stereotypes about such groups if one assumes that whatever is salient is perceived as associated because social stigma producespsychological salience (Goffman, 1959). Of course, such stigma may be partly due to existing group stereotypes, so stereotypes also help create illusory correlations (Hamilton & Rose, 1980).
SOCIAL DOMINANCE THEORY
197
do not mean that role-stereotypes are necessarily truthful. Rather, when people make inherent attributions for actions, such as by asking themselves, “What kind of person would do this role?” they may well come up with a stereotype they already know as an explanation. Even when people are not thinking very hard or paying much attention to other people, simple cues such as sex or race easily activate stereotypes, leading people to presume stereotypic qualities in others (e.g., Branscombe & Smith, 1990, Pratto & Bargh, 1991). Because people’s cognitive resources, particularly their attention and time to decide how to behave, are quite limited, people rely on stereotypes heuristically in interacting with others (cf. Bodenhausen & Wyer, 1985; S. T. Fiske & Neuberg, 1989; S. T. Fiske & Taylor, 1991). Group markers and group differences in the social environment, then, interact with the processes of perceiving and explaining to contribute to stereotyping. Other features of group dominance societies also contribute to stereotyping. Group dominance is characterized by group segregation, and under those conditions, one’s personal experience teaches one so much more about in-groups than out-groups that it is easy to assume that out-groups are more homogenous and stereotypic than in-groups (Linville, Fischer, & Salovey, 1989; Sherif & Sherif, 1966). Many of the social conditions that have been found to reduce stereotyping are atypical within group dominance societies. For example, Erber and S. T. Fiske (1984) showed that people can get to know others in outgroups and rely less on stereotypes when people’s outcomes are positively interdependent with those others. However, members of dominant and subordinate groups are usually independent or negatively interdependent (competitive) with one another, and people persist in their group prejudice under these conditions (Stephan, 1987). S. T. Fiske and Neuberg (1989) showed that given greater attentional resources, people can individuate more and stereotype less. But Deprkt and S. T. Fiske (1993) argue that people in positions of power, which is more typical of dominant group members, are unlikely to have more attention to pay, and so are especially likely to stereotype. The goal to be accurate (Neuberg & Fiske, 1987; Tetlock & Kim, 1987) can reduce stereotyping, but it is less necessary for dominant group members because they possess greater power (S. T. Fiske, 1993). Even “tokens”-exceptions to stereotypes such as a woman in a maledominated profession, do not eliminate stereotyping, because people pay more attention to tokens and make more internal attributions for their actions. This can lead to greater stereotyping of such tokens and to discrimination in line with the stereotypes (Taylor, Fiske, Close, Anderson, & Ruderman, 1977). Further, being a token can lead to more consciousness of one’s stigmatized group membership (Frable, 1993), and thus to under-
198
FELICIA PRA’ITO
performance (Steele & Aronson, 1995), which in merit-judgment situations, provides the grounds for discrimination. Thus, tokens may not provide counter-examples to stereotypes and make inroads against segregation as much as they help reconfirm stereotypes. This review shows that the social conditions characteristic of group dominance societies-segregation, intergroup competition, group power differentials, and even tokenism-interact with social-cognitive processes to contribute to stereotyping. This mutual compatibility between the social environment and socialcognitive processes seems to give stereotypes a life of their own. In fact, stereotypes reperpetuate themselves through cognitive and behavioral channels. Cognitively, stereotypes filter the acquisition of information consistent with stereotypic expectancies through information-seeking strategies (e.g., Cantor & Mischel, 1977) and through selective attention (e.g., Bargh & Thein, 1985). Stereotypes also bias memory towards recollection of stereotype-confirming “evidence” (e.g., Snyder & Uranowitz, 1978). When participants are presented with both stereotypic consistent and inconsistent information about a target person, most studies have found greater recall of stereotype-inconsistent information. But Hilton and von Hippel (1996) point out that this bias may lead to greater attributions for “explaining away” the incongruity, thereby maintaining the stereotype in the face (and recollection) of inconsistent information. Stereotype-inconsistent information can be used to form a subtype that also functions as a stereotype (Andersen & Klatzky, 1987),and the coexistence of contradictory subtypes (e.g., madonna vs. whore, Ashmore, 1981;Jezebel vs. Mammy, White, 1985) only seems to ensure that a stereotype is available for every occasion. Most importantly, stereotypes can influence people’s behavior in ways that confirm the stereotype (cf. Merton, 1948). For example, Word, Zanna, and Cooper (1974) showed that when White interviewers had negative expectations of Black job candidates, those expectations led them to treat job candidates in interpersonally distant ways. In response to this uncomfortable situation, the job candidates appeared more flustered and unprepared, giving worse interviews than when they were treated more respectfully. Likewise, Steele and Aronson (1995) showed that the stereotype of Blacks, which includes the notion of intellectual inferiority, can cause Blacks to underperform in standard intelligence-testing situations, providing evidence confirming the stereotype. This mutual perpetuation of social conditions and psychological biases contributes to the stability of group dominance societies. My review has emphasized that psychological processes (e.g., categorization, attribution) coupled with aspects of common social context (e.g., group differences in power and behavior, segregation) together form a selfperpetuating system in which thoughts, behavior, and social patterns each
SOCIAL DOMINANCE THEORY
199
contribute to the others. Stereotypes and other social beliefs, then, live not just in individuals’ minds, but in patterns of behavior, and in the social structure and meanings that constitute culture (cf. Johnson, 1994; Sanday, 1981). To understand how they sustain social inequality, we cannot just examine stereotyping and discrimination by individuals, but also how ideologies pertain to more general social practices, such as social discourse and institutional discrimination.“ I define social ideologies as consensually shared beliefs about how people ought (and ought not) to behave. Ideologies therefore include social stereotypes, principles of resource allocation (e.g., equity, capitalism, socialism), role prescriptions, origin myths, citizenship rules, and other stories or ideas that define groups. In saying that ideologies are consensual, I do not mean that everyone agrees with them, but rather that within a given society, ideologies are commonly understood and presumed to be relevant. This definition of social ideology, then, necessarily deals with the social meaning of beliefs, and that must be understood in social context. For example, people can make much of whether “appropriate” racial beliefs should presume races are different, and whether “inferior” races should be excluded from society or simply given appropriate (subordinate) roles, all of which have been considered both “liberal” (progressing towards equality) and “conservative” (maintaining of status and power distinctions) at different historical points (Turner & Singleton, 1978). The same kinds of debates have raged regarding gender, even among social psychologists (e.g., Eagly, 1995; Kitzinger, 1994). Likewise, whether individual privacy, support for the free market, government activism, political freedom, and a number of other particular ideological principles are called “liberal” or “conservative” depends on the social context and the effects of the policies these principles are garnered to support. Thus, arguments over the particular contents of social and political attitudes that do not contextualize those arguments in terms of meaning and power, nor relate them to theoretical constructs, make much ado about little of theoretical interest (see also Rokeach, 1973, pp. 166-167). Clearly, the social and political power of ideologies depends as much on their current social legitimacy and what they have been made relevant to as on their content. The “ought” part of the definition emphasizes that ideologies do not just influence individual behavior, but also establish guidelines for social judgment. Ideologies therefore legitimize or de-legitimize persons, groups, and social practices, both in individual minds and in social discourse (e.g., van Dijk, 1987). Thus, ideologies explain and justify personal behavior and cultural practices. Ideologies are like cultural “genes”-they act on 4 S e e also Parker (1992), van Dijk (1987). Wetherall and Potter (1993).
200
FELICIA PRATTO
individuals and institutions so that people take actions to produce and reproduce the social relationships that comprise the culture and its ideologies. This linkage between psychological processes and social context, between thought and behavior through ideology, is one reason group-based inequality tends to be so stable even in the face of economic, technological, and cultural change. Anti-black racism in the United States, both in rhetoric and in practice, is an unfortunate example of this stability. The specific discriminatory policies that help prevent Black intellectual achievement have changed from laws prohibiting the education of Black slaves into the 19thcentury, to the refusal of universities to admit Black students into the early 20thcentury, to the current egregious differences in funding for public education for Black and White students (Kozol, 1991).Yet all such discriminatory policies are conveniently justified by the persistent stereotype (e.g., Devine & Elliott, 1995) that Blacks are intellectually inferior to Whites. Such practices also create racial differences in behaviors people consider evidence of “intellect.” Like viruses, social ideologies such as the myth of Black intellectual inferiority appear capable of just enough mutation and environmental influence to permit survival in slightly different social environments, but with their germ and social potency intact (cf. Hernnstein & Murray, 1994). Stereotypes are far from the only social ideologies that act this way. Pratto, Tatar, and Conway-Lanz (1999) argue that because social ideologies are formed and applied within particular social contexts, even those allegedly “general” ideologies that do not name particular groups, such as meritocracy, will have differential implications for social groups. For example, in contrasting themselves with British subjects, American revolutionaries idealized independence, a concept that is central to Americans’ social identity. But this “general” ideal of independence, and the legal definition of American identity the revolutionaries created, clearly applied more to groups with more power and resources, such as land-owning men versus indentured servants, slaves, women, and children. As Sears’s (1993) “symbolic politics” approach argues, social context can link attitudes towards groups with beliefs and attitudes that ostensibly are not about the group, but are symbolically linked to the group. That Blacks are often perceived by Whites t 3 violate the Protestant work ethic or other cultural virtues is a case in point (e.g., Katz & Hass, 1988; Kluegel & Smith, 1983; Sears, 1988). A symbolic analysis helps explain a robust finding that analyses concerned only with the content of conservative principles (e.g., Sniderman & Piazza, 1993) fail to account for, namely, that the ostensibly nonracial ideology of conservatism correlates robustly with anti-Black racism and
SOCIAL DOMINANCE THEORY
201
other forms of ethnic prejudice (see Sidanius, Pratto, & Bobo, 1996, for a review and discussion). By supposing that ideologies that actually privilege one group (e.g., capitalism, meritocracy) apply to everyone, other groups can be systematically disadvantaged or even denigrated. For example, the Supreme Court twice ruled that disability insurance that excludes pregnancy (even while including male-only medical benefits such as hemophilia and prostatectomies) does not discriminate on the basis of gender (Gedulgig v. Aiello, Electric Co. V. Gilbert). Sandra Bem (1993, pp. 74-77) argued that this is because the Court assumed that the situation men, but not women, are in is normal. Other research indeed substantiates psychological androcentrism in showing that people’s standard image of a person is of a mule person (e.g., Eagly & Kite, 1987; Miller, Taylor, & Buck, 1991). In one of the most profound insights as to why ideologies that privilege dominants become society’s “general” ideologies that everyone ought to live up to, Dahrendorf (1968) postulated that what societies define as meritorious are those features that are more easily exhibited by elites, Collier (1988) and Humphrey (1985) have provided ethnographic and experimental evidence, respectively, for this proposition. Among the Comanche, Kiowa, and Cheyenne, Collier (1988) documented that men with high rank were identified by their greater number of war honors and horses, but that it was the life-style afforded by high rank that enabled such men to acquire war honors and horses. Humphrey’s (1985) participants spent 2 hours roleplaying office work as either managers or clerks. Even though participants were told that they were randomly assigned to their role, both managers and clerks rated the managers as superior on leadership, assertiveness, and other role-related v i r t ~ e sSuch . ~ studies suggest that people use elite status as a heuristic for deciding what is good, virtuous, or socially valued, which results in societies establishing social standards that are more easily met by elites than nonelites. This pattern is one of the main reasons that consensual social ideologies-even those ostensibly neutral with respect to groups-help to maintain group dominance. Social dominance theory’s subtheory of ideology has clearly been influenced by Marx and Engels’s (e.g., 1846) thesis that ideology is aided by and justifies power differentials and the social, legal, and economic practices that create them. But our analysis of ideology differs from Marxism in two key ways. First, social dominance theory does not presume that social ideologies that privilege elites are necessarily false. Rather, it presumes that their truth lies not in some objective ascertainment, but in their social The exception to this pattern was that clerks did not rate themselves, but rather, the other clerks, as inferior to managers.
202
FELICIA PRATTO
efficacy. Legitimizing ideologies or myths, as we call them (e.g., Pratto, Sidanius, Stallworth, & Malle, 1994), make themselves come true, as Merton (1948), Pettigrew (1979), and Sanday (1981) have cogently argued and as much stereotyping research shows. Second, though social dominance theory recognizes that many ideologies legitimize social hierarchy, it also recognizes that some social ideologies attempt to delegitimize social hierarchy (e.g., early Christian principles, democratic principles, feminism, socialism), and it posits that to the extent those are enacted in social practice, the degree of inequality will attenuate (see Pratto et al., 1994; Pratto et al., 1998; Sidanius, Pratto, Martin, & Stallworth, 1991, for further discussions of ideology). As I describe my research below, I will explore further how ideologies justify and influence social practices such as social policies, voting, and resource allocation. B. SOCIETAL PATTERNS OF DISCRIMINATION Research on individual prejudice and stereotyping informs the psychological reasons for inequality, but group-inequality is most immediately created and maintained by systematic discrimination favoring dominants and disfavoring subordinates. Discriminatory action, then, is another piece of our puzzle. Much of this discrimination is performed by institutions, which allocate resources, privileges, and punishments on much larger scales than individuals do (see Feagin & Feagin, 1978). Summarizing research on institution discrimination comprises four chapters of a book on social dominance theory (Sidanius & Pratto, 1999), so I will provide only a few examples to indicate how systematic such discrimination is and how it is linked to cultural and psychological processes. Although education is supposed to be universal and publicly funded in the United States, far more funding and other resources are available to the socially dominant race (Whites) than to socially subordinate races (Blacks, Latinos; Kozol, 1991). These funding differences are written into law at the local level by being linked to local property taxes, and at the federal level by two Supreme Court decisions (San Antonio Zndependent School District v. Rodriguez; Milliken v. Bradley) that hold that states are required neither to provide equal access to education nor to desegregate schools-contrary to what many thought that the 1954 school desegregation case, Brown v. Board of Education, had established. Thus, educational discrimination results from the collusion of legislatures, schools, and the courts, aided by residential segregation. Systematic group discrimination is also typical of financial and housing authorities. For example, in a study of all loans considered by Federal
SOCIAL DOMINANCE THEORY
203
Reserve banks, Latinos (22%) and Blacks (34%) were substantially more likely to be denied loans than Whites (14%), even when they submitted equivalent loan applications (Munnell, Brown, McEneaney, & Tootell, 1992). Likewise, field experiments have shown that housing agents discriminate against members of subordinate ethnic groups even when their financial assets are exactly the same as members of dominant groups (Yinger, 1986, 1991a, 1991b, 1995, 1996). Field experiments by Ayres (1991, 1995; Ayres & Siegelman, 1995) have also shown financial discrimination in the purchase of another big-ticket item: automobiles. Compared to White men, women were overcharged by hundreds of dollars and African-American men by over a thousand dollars. Employment discrimination is another major arena in which the economic and personal power, resources, and health of subordinated group members are systematically harmed. In the most compelling kind of study, pairs of job-seekers with identical skills, credentials, recommendations, and employment histories, apply for the same advertised jobs. Nineteen such field experiments in five nations showed that job-seekers from subordinate groups are less likely to have their applications accepted, to be interviewed, and to receive job offers, than job-seekers from dominant groups (see Sidanius & Pratto, 1999, for a meta-analysis). Even where such discrimination is illegal, a substantial proportion of employers admit to being categorically unwilling to hire members of subordinate groups (e.g., Langendijk, 1986; W. J. Wilson, 1996; Zegers de Beijl, 1990). Rather than being gainfully employed, a shocking proportion of members of subordinate groups, especially young men, are targets of the criminal justice system. The examples of Australia and the United States are extreme in degree, but not in form. The indigenous people of Australia are imprisoned at 15 times the rate of nonindigenous people (Cunneen, 1990). At present, almost one-third of African-American men between ages 20 and 29 are under some form of criminal supervision: either in prison, in jail, on probation, or on parole (Mauer & H u h , 1995). This disproportional punishment is not simply due to greater rates of criminality among subordinate groups. Studies have also shown that discrimination in a variety of forms contributes to this phenomenon. For example, police especially target neighborhoods with high proportions of subordinates (Mauer, 1990). Once arrested, subordinates are more often found guilty and are more severely punished-even when controlling for numerous legally relevant variables such as plea, severity of crime, prior criminal record, and bail (e.g., Mazzella & Feingold, 1994; Sweeney & Haney, 1992). Another form of legal discrimination is to proscribe acts that are disproportionately committed by subordinates. As a case in point, U.S. federal law, in the 1988 AntiDrug Abuse Act, now provides that one has to possess 100 times as much
204
FELICIA PRATTO
powder cocaine as crack cocaine to receive the same federal prison term. Users of powder cocaine, the less-punished drug, are overwhelmingly EuroAmericans (75% of users), whereas crack cocaine is disproportionately used by Blacks (38% of users) and Latinos (10% of users; see Mauer & Huling, 1995). This differential legal sanctioning of comparable behavior by dominant and subordinate people contributes directly to the enormous disproportion of subordinates under criminal supervision (Mauer & Huling, 1995). In summary, discrimination against members of subordinate groups targets them at every turn: in employment, housing, purchasing, education, and punishment. This discrimination by virtually every social institution is systematic not only because it is consistent across form and across societies, but also because each form of institutional discrimination augments other forms. For example, discrimination in housing and lending tends to segregate subordinate groups and contribute to their crime victimization. Housing segregation and discrimination also contributes to underemployment and to putting subordinates in inferior schools, which also leads to future underemployment. Institutional discrimination shows that to understand inequality we must consider the collective nature of discrimination. Social dominance theory analyzes discrimination as one of the cultural patterns of group dominance societies and describes how it is wedded to the social beliefs, structural relationships, and psychological biases of the people in those societies. Foremost, the theory postulates that social ideologies are central in prescribing the goals of social institutions, and the behavior of people in those institutions. Thus, when individuals and institutions have adopted social ideologies that privilege dominants over subordinates, individuals and institutions are more likely to discriminate in favor of maintaining group hierarchy. Such discrimination may take the form of maintaining group boundaries in the face of change or challenges, or of actively channeling desirable resources (e.g., jobs, medical benefits, tax breaks, public works) more towards dominants than subordinates, or of channeling undesirable things (e.g., poor housing, dangerous jobs, prison terms, toxic waste) more toward subordinates than toward dominants. Conversely, when individuals and institutions have adopted social ideologies that privilege subordinates over dominants, individuals and institutions are more likely to discriminate against existing group hierarchy by channeling more positive resources toward the disadvantaged. This kind of behavior is also discriminatory in that it treats people in different groups differently, but it serves to attenuate rather than enhance social hierarchy. In considering whether individual and institutional action supports or attenuates social inequality, social dominance theory considers what social ideologies are prevalent.
SOCIAL DOMINANCE THEORY
205
11. The Integration: Research on Social Dominance Theory
We have now examined three basic pieces of the puzzle of group dominance: psychological orientations towards prejudice, social ideologies that justify social practices, and group discrimination. After explicating the construct of social dominance orientation (SDO), I will show how these pieces and SDO fit together to perpetuate group dominance. Each study will join at least two pieces of the puzzle of group dominance.
A. SOCIAL DOMINANCE ORIENTATION Social dominance orientation (SDO) is one’s general psychological motivation toward group dominance and is measured on a continuum from “high” (positive) to “low” (negative) support and preference for groupbased dominance in general. Whether people are low or high in any absolute sense is irrelevant to our research methods and to some extent unknowable because such standards always depend on some comparison point. Social dominance theory predicts that within a population, people higher on SDO will be the most supportive of the social ideologies, policies, and practices that help to reify group-based dominance, and people lower on SDO will be the most supportive of social ideologies, policies, and practices that reduce group inequality and instead promote group inclusion and equality. By examining how individual and group differences in SDO pertain to group status, social ideologies, social and institutional roles, we will see how individual prejudice works in conjunction with other components of systems of group dominance. Pratto et al. (1994) developed two reliable, balanced scales to measure SDO, which we now call SDOS(the 14-item scale) and SD06 (the 16-item scale; see items in Table I). Pratto et al. (1994) validated S D 0 5 more extensively, but S D 0 6is more face-valid. The studies presented here, unless otherwise mentioned, used the S D 0 6 scale, further validating it. By 1995, in my data, SDOS was averaging an internal reliability (Y of .83 and crosstime average reliability of r = 3 3 . The S D 0 6 scale was averaging an (Y of .92 and a cross-time r of 36. Before getting on to the predictive validity of these scales, I summarize evidence that SDO is not simply repackaged interpersonal dominance, conservatism, authoritarianism, or individual egalitarianism. Pratto et al. (1994) showed that people low on SDO are higher on openness to experience, but SDO related to none of the other Big Five personality dimensions. We found repeatedly that SDO was independent of Rosenberg’s (1965)
206
FELICIA PRATTO TABLE I ITEMS AND INSTRUCTIONSFOR SDOS and SD06 Scales“
SDOJ scale items 1. Some groups of people are simply not the equals of others. 2. Some people are just more worthy than others. 3. This country would be better off if we cared less about how equal all people were. 4. Some people are just more deserving than others. 5. It is not a problem if some people have more of a chance in life than others. 6 . Some people are just inferior to others. 7. To get ahead in life, it is sometimes necessary to step on others. 8. Increased economic equality. 9. Increased social equality. 10. Equality. 11. If people were treated more equally we would have fewer problems in this country. 12. In an ideal world, all nations would be equal. 13. We should try to treat one another as equals as much as possible. 14. It is important that we treat other countries as equals. SDOh scale items 1. Some groups of people are simply inferior to other groups. 2. In getting what you want, it is sometimes necessary to use force against other groups. 3. It is OK if some groups have more of a chance in life than others. 4. To get ahead in life, it is sometimes necessary to step on other groups. 5. If certain groups stayed in their place, we would have fewer problems. 6. It is probably a good thing that certain groups are at the top and other groups are at the bottom. 7. Inferior groups should stay in their place. 8. Sometimes other groups must be kept in their place. 9. It would be good if groups could be equal. 10. Group equality should be our ideal. 11. All groups should be given an equal chance in life. 12. We should do what we can to equalize conditions for different groups. 13. Increased social equality. 14. We would have fewer problems if we treated people more equally. 15. We should strive to make incomes as equal as possible. 16. No one group should dominate in society.
“ Items are intermixed arbitrarily when administered. Instructions read “Which statements do you feel positive or negative toward? Write a number from the scale below to indicate your feeling about each statement.” The scale reads: 7 very positive, 6 positive, 5 slightly positive, 4 neither positive nor negative, 3 slightly negative, 2 negative, 1 very negative. On SDOS,items 8-14 and on SD06, items 8-16 are reversed.
self-esteem measure. We used two measures of interpersonal dominance (CPI, Gough, 1981, and JPRF, Jackson, 1965). one of which McClosky (1958) had found to correlate with conservative attitudes (Gough, 1981), but neither correlated at all with SD05.I believe these kinds of dominance measures, which are not gender-typed and are socially desirable (see Gough,
SOCIAL DOMINANCE THEORY
207
1981), assess an organized, efficacious orientation that does not capture the desire for conquest, control, and superiority over others that thrills people high on SDO and disgusts people low on SDO. Some indications of this taste appear in an exploratory 1995 study of Stanford students ( N = 201). I wrote a “Could You?” questionnaire asking people to imagine a number of life events and rate how easy or difficult it would be for them to do each. The subscales that correlated with SDO concerned separating from others, being harsh to one’s family, commiting violence in war, and participating in derogatory group jokes (see Table 11). Altemeyer (1998) more extensively tested the association of SDO with interpersonal dominance and power drives. He found that SDO correlates strongly with Machiavellianism, power values, and the desire to humiliate and triumph over others in what high SDO people presume to be an omnipresent zero-sum TABLE I1 “COULDYou?” SUBSCALE ITEMS,RELIABILITIES, A N D CORRELATIONS WITH SDO“ Separate from others ( a = .78) r = .28** Be physically separated from my small child for over a month. Live apart from my husband or wife for over 6 months. Live without spending regular time with close friends. Go over a year without seeing my parent(s). Give main custody of my child(ren) to my spouse. Go indefinitely without talking to or writing to my siblings. Move somewhere that I wouldn’t see my parents for the rest of my life. Be harsh to family (a = .58) r = .23** Spank my child for discipline. Make harsh remarks to my child. Hit my child in anger. Make harsh remarks to my spouse. Make war ( a = .78) r = .34** Engage in combat. Help torture an enemy for strategic purposes. Kill someone to defend myself or others. Kill someone in warfare. Derogatory jokes (a = .67) r = .34** Go along with a sexist joke or remark, even if I didn’t like it. Tell a joke that made fun of a minority group. N = 201. Instructions read in part: “Adults are often confronted with difficult situations that they had not anticipated. . . . Think about what you know about yourself deep down. Then rate whether you could not, could with difficulty, or could easily, do the following things.” The rating scale read 1, “I absolutely could not do this”; 2, “I would find it extremely difficult to do this”; 3, “I would find it hard to do this”; 4, “I could do this with minor struggles”; 5 , “I could do this fairly easily”; 6 , “This would be very easy for me to do.” Items from different subscales were intermixed. **p < .01.
208
FELICIA PRATTO
contest for power and dominance. Such results confirm data of mine showing that SDO is related to desire for triumphal prestige, assessed in nine of the samples studied by Pratto et al. (1994; see items in Table 111). My prestige scale correlated fairly highly with SD05in each sample (see bottom of Table 111); moreover SDO was related linearly to participants’ aspired income levels. Because so many studies have found conservatism to correlate with sexism and racism, we view conservatism as a hierarchy-enhancing legitimizing myth and so expect it to correlate with SDO (see, e.g., Pratto, Liu, et al., 1998; Pratto, Stallworth, & Conway-Lanz, 1998; Pratto, Stallworth, & Sidanius, 1997; Pratto et al., 1994; Sidanius & Pratto, 1993a; Sidanius, Devereux, & Pratto, 1992; Sidanius, Pratto, & Bobo, 1994; Sidanius, et al., 1996). However, SDO and conservatism are conceptually distinct, and SDO does not correlate highly with conservatism (see Pratto et al., 1994; Sidanius et al., 1996 for discussions). Furthermore, the predictive effects of SDO cannot be reduced to conservatism (see Pratto et al., 1994; Pratto Stallworth, & Sidanius, 1997, Pratto, Stallworth, & Conway-Lanz, 1998). SDO is conceptually distinct from authoritarianism as traditionally operationalized in RWA because RWA concerns submission to authorities in one’s own group, whereas SDO concerns the desire for one’s own group
TABLE 111 PRESTIGE SCALEITEMS, INTERNAL RELIABILITIES, AND CORRELATIONS WITH SDOSa
1. Winning is more important than how the game is played. 2. Getting ahead in life by almost any means necessary. 3. Competition. 4. I feel very good after beating someone at a game. 5. One of the best feelings in the world is when your favorite sports team has just won. 6. If you got it, flaunt it. 7. I admire people who are able to spend money on luxuries. 8. In considering buying a car, I would think about what kind of car would impress people. 9. If I won the lottery, I’d be able to show people who I really am. 10. It is important to me that others respect and admire me. 11. It is important to get ahead in this world, no matter what. 12. I’d give almost anything to get rich. Sample 1 2 3a 3b 4 5 6 7 8 9 a reliability
.74
.77
.77
.78
35
.80
r with SDO
.44
.49
.39
.37
.29
.53
30 .55
.77
.78
.72
.43
.55
.46
“ Sample numbers match those used by Pratto et al. (1994). All correlations differed from zero at p < .01.
SOCIAL DOMINANCE THEORY
209
to dominate another group (Duckitt, 1989, but see Altemeyer, 1998). This distinction holds up empirically. S D 0 5 correlated only slightly with RWA and with Goertzel’s (1987) personality measure of authoritarianism (Pratto et al., 1994). Likewise, Altemeyer (1998) finds small to moderate correlations between SDO and RWA in Canada, as McFarland and Adelson (1996) have in the United States. In all these studies, SDO correlates more strongly with numerous forms of group prejudice than RWA does, excepting prejudice against gays and lesbians, perhaps because RWA but not SDO is related to religious orthodoxy (Altemeyer, 1998). Also, SDO and RWA predict different components of group prejudice, showing that they are not redundant with one another. Altemeyer’s (1998) extensive research also confirms earlier findings that SDO has a gender difference (men are higher), whereas RWA does not. Finally, although our concept of SDO is indebted to Rokeach’s illumination of social equality as a central social-political value, SDO specifically concerns group-based dominance rather than general or individual equality. To empirically test this distinction, in 1995 207 Stanford University undergraduates completed the S D 0 6 scale and Braithwaite and Law’s (1985) improvements on Rokeach’s values scales, which use rating rather than ranking formats. The Individual Rights scale corresponds to American notions of individual equality, but it did not correlate with SDO, r = -.03, n.s. Rather than individual equality, the two Rokeach values I believe are more conceptually related to SDO are the International Harmony and Equality (IHE) scale, reflecting the desire to achieve a more cooperative and humanistic social order, akin to low SDO, and the National Strength and Order (NSO), reflecting the desire to attain economic and political might and internal order, akin to high SDO. In fact, SDO correlated most strongly with the IHE scale ( r = -.40, p < .Ol), and then the NSO scale ( r = .14, p < .05). All three results further validate our measure of SDO, while showing that SDO is not redundant with relevant Rokeach political values. In terms of Rokeach personal values, SDO correlated r = -.14 with Personal Growth, and r = .17, p’s < .05, with Social Standing. There is a socioemotional temperament that mitigates against high SDO, namely being communal, hinted at in the negative correlation with the ease of separating from others in Table 111. This makes theoretical sense because people who include others in themselves and who are empathic and generally concerned about others, regardless of the others’ group membership, should be less group-prejudiced. As expected, Pratto et al. (1994) found that people low on SDO were more concerned with other people; S D 0 5 correlated an average of - .46 with concern for others, - .33 with communality, and -.30 with tolerance, and -.28 with altruism.
210
FELICIA PRATTO
B. SOCIAL DOMINANCE ORIENTATION AS THE PSYCHOLOGICAL MEDIUM FOR TRANSMI’MTNG GROUP DOMINANCE Studying SDO is useful for understanding group dominance societies because psychological biases like SDO are both an outcome and a cause of human environments. In other words, by studying the predictors and consequences of SDO, a psychological medium that helps to transmit social dominance from one particular social context to another, we will see how people are both influenced by and create their own environments. By seeing SDO as a means of transmission rather than a root cause, social dominance theory reflects both the learning tradition of psychology and evolutionary theory. C. SOCIAL DOMINANCE ORIENTATION AND REPRODUCTIVE STRATEGIES
I have pursued research showing this transmission process in two ways: one to reflect the evolutionary tradition, and one to reflect the social learning or culture-and-personality tradition. The essential evolutionary postulate is that features that provide an adaptive advantage by increasing one’s chance of reproductive success are likely to be selected. This postulate implies that behavioral and psychological predilections that help one obtain a mate or mates and raise children to maturity have been and will be selected. Because the sexes differ in how much to “invest” in each child to maximize reproductive success (Trivers, 1972), several theorists have postulated gender differences in certain reproductive strategies (see Buss, 1992; Feingold, 1992, for reviews).6 Men can increase their reproductive success by mating with more than one woman. To further provide for the children they “invest” substantially in, women are expected to seek mates with more resources to invest, notably, social status. In response, men try to insure that they are not contributing their status and other resources to children not their own, by being resistant to taking care of “other people’s’’ children and by being sexually jealous. Thus, evolutionary theory posits that the reproductive strategies of one sex will influence the behavior of the other. For example, if women are attracted to high-status mates, then those men are expected to try to monopolize social status, taking it away from women and also from other men. In group dominance systems in This gender difference stems from internal fertilization in the female, lactation, and the fact that men can produce simultaneous pregnancies, under the right social conditions, more easily than women can.
SOCIAL DOMINANCE THEORY
2s 1
which women are prevented from controlling resources and in which men do vary in the resources they control, selection of a powerful, high-status mate should be especially adaptive for women. If these behaviors seem undemocratic, it is because they rely on power differentials, both between men and women, and among men or male-dominated groups (cf. Betzig, 1993; Gailey, 1987; Pratto, 1996a). For reasons such as these, I have theorized that the predilection for SDO may have been selected in part from human reproductive strategies (see Pratto, 1996a; Pratto, Sidanius, & Stallworth, 1993). Moreover, the use of these reproductive strategies provides a basis in family structure for the two sociostructural features of group dominance societies: men dominating women, and male-dominated ethnic or class in-groups exploiting and oppressing out-groups (see Pratto, S996a). In such societies, higher levels of SDO may well be adaptive in facilitating these reproductive strategies. If so, then men who are especially prone to engage in the “masculine” reproductive strategies of controlling women sexually and economically, polygyny, dominating other men, and who resist taking care of “others”’ children, should be higher on SDO. Conversely, women who engage in the “feminine” reproductive strategy that feeds the status-and-power race among men, namely seeking high-status mates, should be especially high on SDO. Pratto and Hegarty (1998) tested predictions that reproductive strategies would be related to SDO in specific ways in four samples of young unmarried people who identified themselves as heterosexual. Samples 1, 2, and 3, used the SDOSscale; Sample 4 used the SD06 scale. The samples were San Jose State University and Stanford University undergraduates, and had N s of 462, 397, 108, and 142 with 47%, 43%, 43%, and 61% men, respectively. We hypothesized that SDO would be more strongly related among men than among women to four reproductive strategies that parental investment theory hypothesizes are especially adaptive for men: polygyny, guarding against paternity uncertainty, sexual jealousy, and having more status and resources than one’s mate. Similarly, we predicted that among women more than among men, SDO would be related to the strategy hypothesized to be especially adaptive for women: acquiring a high-status mate. We tested these hypotheses by comparing the size of the regression slopes relating preference for these reproductive strategies to SDO between men and women using a t-test.’ We predicted that the slopes would be larger among men for the “masculine” strategies, and larger among women for the “feminine” strategy. That is, we did not simply predict the gender differences in preferences for these sexual strategies as previous researchers have, but gender differences in how strongly each strategy would relate to SDO.
212
FELICIA PRATTO
We used several different ways of operationalizing each reproductive strategy and tested hypotheses in multiple samples. Some measures were based on Buss’s (1989) mate preferences questionnaire, which asks people to rate how important certain qualities are in their “ideal mate”; other items we added. To assess potential for polygyny, we asked Samples 3 and 4 to indicate the probability they would have at least one affair while married. As predicted (see Table IV), the slopes on SDO were reliably larger among men than among women. In other words, SDO was more strongly associated with self-predicted likelihood of polygyny among men than among women. The higher men were on SDO, the more unwilling they were to be parents to children they might not consider their “own” (adopted children, stepchildren): but this was untrue of women. Several sexual jealousy measures were also more strongly related to SDO among men than among women as predicted. The higher men were on SDO, the more they expected their ideal mate to have an extramarital affair, but the less willing they were to “release” her through divorce following such an affair. Although SDO was unrelated to sexual jealousy among women, in women, we found a complementary phenomenon: the higher women were on SDO, the more they expected their ideal mates to be especially sexually jealous. In all four samples, we found that high SDO men, especially, wanted to make more money than their spouses. This “strategy” allows husbands more power within the marriage (Okin, 1989). Among women, desire for a high-status, high-earning mate was much more characteristic of those high on SDO. In summary, we found that SDO was more strongly related to four “masculine” sexual strategies among men: polygyny, sexual jealousy, guarding against paternity uncertainty, and wanting a higher income than one’s mate. We also found that SDO was more strongly related to a complimentary “feminine” sexual strategy among women: obtaining a high-status, well-off mate. There are three main implications of these results. First, people high and low on SDO may select different marriage environments for themselves. As with most personal features, people desire similar mates to themselves. In Sample 3, participants completed the S D 0 5 scale as their ideal mates would; this score (a = .71) and own score correlated positively, r = .31, p < .01. Furthermore, Altemeyer (1998, p. 85) found that husbands’ and wives’ SDOSscores correlated r = .45.If there is any predictive validity to people’s expressed preferences about reproductive strategies (an important issue that requires further research), then these data suggest that people higher on SDO will experience more gender domination within their sexual
’See Pratto (1996a) on the critical importance of social definitions of “own” children.
SOCIAL DOMINANCE THEORY
213
relationships than people lower on SDO. A marriage of two people high on SDO may have a husband who provides financial and status resources, but who is also sexually and economically controlling and who is more likely to be sexually unfaithful. Given that the high SDO wife also desires status and high earnings and expects sexual jealousy in her husband, they may both be relatively satisfied. But their marriages and the nature of gender within these relationships would seem, not surprisingly, much less egalitarian than our data suggest would be true of low SDO people. A marriage of two low SDO people may have less gender differentiation in terms of sexual promiscuity, sexual jealousy, occupational status and earnings, and child-care. These speculations are of course tentative and require further research. But these results are consistent with the view that SDO may be integral to family patterns that reify both gender- and group-based inequality (see Pratto, 1996a). Second, these results make it plausible that the propensity for being high on SDO may have been selected. Evolutionary psychology emphasizes that people’s behavior in the past, especially social behavior that influenced reproductive success, may have influenced contemporary psychological features. This is part of how, I believe, social behavior and conditions have been “transmitted” into SDO. Third, these results show that the parental investment theory of gender differences in mate preferences (e.g., Buss, 1992) best describes people high on SDO. That fact has ramifications for parental investment theory and how it is to be interpreted. Although adaptive sexual strategies may accompany being high on SDO, it would be wrong to assume that high SDO is the main social psychological feature that is adaptive or successful. In fact, the available data suggest that being “low” on SDO may be more typical of human beings: the distribution of SDO among high-status U.S. college students is positively skewed. This begs for an expansion of evolutionary psychology to incorporate the psychological features that enable people to affiliate and identify with others, features which are also necessary to reproductive success and to the formation of groups (e.g., Brewer & Caporael, 1990). As current theories of parental investment only describe high SDO people well, it is not surprising that evolutionary theory has been described as a justification for gender inequality and social stratification (e.g., Shields, 1975). D. PERSONALITY AND CULTURE
The social learning tradition suggests that when people are surrounded by particular forms of group dominance in their societies, they learn enough
TABLE IV COMPARISONS BETWEEN MENAND WOMEN ON REGRESSION SLOPESOF SEXUAL STRATEGIES ON SDO" Strategy and specific measure
3 P
POlYgYnY Percentile I will be more faithful than Probability of own extramarital affair(s) Probability of own extramarital affair(s) Guarding against paternity certainty Comfort with same-ethnic adoption Willing to adopt if infertile (yesho) Ideal mate has no children from previous relationships Ideal mate has no children from previous relationships Ideal mate has no children from previous relationships Comfortable marrying if woman pregnant and paternity in doubt Sexual jealousy Probability spouse will have affair(s) Percentile that spouse is more faithful than Probability I would divorce unfaithful mate Probability I would divorce unfaithful mate Probability I would divorce unfaithful mate Percentile that ideal mate is more sexually jealous than
Sample
B women
3 4
-4.13 1.93 1.62 - .28 - .04
.oo .01 .30 -.15
- 1.46
-3.92 0.24 -0.97 -3.11 13.96*
B men
t
df
P
-8.00 5.09 8.81***
22.02 -4.60 -40.99
85 125
.0005 ,0005 .0005
-.59* -.19** .21* -.lo .62* -.38*
2.42 0.84 -2.21 0.98 -2.02 1.35
194 85 405 321 124 124
.01 .10 .01 n.s. ,025 n.s.
5.28** -1.76** -2.51 -3.54 -8.06* -5.97
-43.67 3.93 28.62 26.48 31.69 105.57
125 86 405 321 125 92
.0005 ,0005 .0005 ,0005 b005 .0005
84
Comparative income Difference between own desired income and ideal mate's (in dollars) Same Same Difference between own desired income and ideal mates' at age 30
k ! VI
High status mate Ideal mate has favorable social status, respectable family, is well educated Same Ideal mate has high social status, respected family, good provider. makes lots of money Ideal mate's income level Same Same Same
1
-2.87
5.05
-82.50
2 4 3
-0.82 - 10.84 -44.19*
-1.19 18.21** -36.87
- 188.99
1
2 3
.43*** .69*** 4.81 7.20 7.49 10.93
3.66 -5.97
405
b005
321 123 80
,0005 ,0005 .0005
.12*
1.73
405
.05
.15 .36*
2.89 2.21
321 126
,005 ,025
28.20 56.23 7.32 35.86
405 237 97 123
.0005
2.16 0.69 0.75 5.19
,0005 ,0005 .0005
" Asterices indicate p-value testing whether each slope differs from zero: * p < .05, **p <.01. ***p < .001.p-values in table correspond to the t-test comparing the size of the slopes between men and women.
216
FELICIA PRATTO
about say, racism and sexism, to develop a general psychological orientation toward group dominance. That general orientation then influences how they think about a broad variety of intergroup relations and issues, whether they behave in ways that reify or reduce group inequality, and whether the contexts they end up in enable or discourage them from promoting group inequality. Accommodating to or rejecting a specific form of group dominance (e.g., racism) may prepare people to accept or reject other specific forms of group dominance (e.g., nationalism). At the individual level, this implies that people who dislike racism will also dislike nationalism, sexism, and other kinds of group dominance, so that SDO accounts for the correlations that typically exist among different group prejudices and ideologies (see Sidanius et al., 1992; Sidanius & Pratto, 1993b, for evidence). At the cultural level, it implies that group-dominance cultures create psychological biases that encourage people to re-create group dominance. This psychological inertia for social dominance may help explain why group dominance is so long-lived even in the face of attempts at greater social equality (e.g., Johnson, 1994), why societies attempting to rid themselves of group dominance often change one form of it for another (e.g., the Bolshevik Revolution), and why immigrants from one group-dominance culture adopt previously irrelevant prejudices of their new group-dominance culture (e.g., anti-Black racism among Asian immigrants to the United States).
E. SOCIAL DOMINANCE ORIENTATION, CULTURAL CONTEXT. AND IDEOLOGICAL USAGE Social dominance theory's symbolic and cultural frame implies that support for any beliefs that justify inequality or equality in a given context should correlate with SDO. In other words, it is not the particular contexts of ideologies that matter, but their meaning in terms of power and intergroup relations. For example, SDO correlates strongly and positively with antiBlack racism in the United States, with prejudice against Asian immigrants and native Canadians in Canada, with prejudice against aborigines in Taiwan, and with prejudice against Palestinians and Jews of lower ethnic status (Mizrachim) in Israel (see Pratto et al., 1994; Pratto, Liu, et al., 1998). The reasons justifying such prejudice read as differently as the histories that created the forms of ethnic domination these societies have. For example, internal attributions about Blacks (e.g., their supposed lack of motivation or ability) undergirds anti-Black racism in the United States, but in Taiwan, poverty is explained with external attributions in religious ideas blaming fate o r the gods-yet both ideologies correlate highly with SDO in their respective contexts. Similarly, SDO correlates with different local forms of
SOCIAL DOMINANCE THEORY
217
sexism. In the United States, beliefs that men and women are psychologically and temperamentally different justify particular divisions of labor in the home and the paid workplace. In Taiwan, sexism is based in Confucian beliefs about hierarchy, which justifies the subordination of daughters and wives to fathers, sons, and mothers-in-law, as well as sexual harassment and unequal treatment of women on the job. In Israel, sexism stems from religious beliefs that women’s sacred role is to raise children and that virtuous women hide their sexuality, beliefs that also justify excluding women from certain occupations. We have found SDO to correlate with these different forms of sexism in each culture (see Pratto et al., 1994; Pratto, Liu, et al., 1998). Our general thesis that ideologies justify particular forms of group dominance and correlate with a general psychological orientation toward group dominance is shown in such numerous specifics. Such results illustrate an important point about social dominance theory. The theory is neither devoid of the shared meanings that describe culture, as social identity theory and attribution theory are, nor is it culturally specific, as symbolic racism theory is. Rather, social dominance theory describes group-dominance societies in general nomothetic terms by describing the way social practices, social belief systems, and psychological processes are linked through shared ideologies. Like language, social ideologies are both human universals and culturally specific. And like language, ideologies enable the people who share them to understand meaning, to know who they are, and to relate to one another (see Pratto, Liu, et al., 1998, for a discussion). Understanding group dominance as both universal and cultured (e.g., Pratto, Liu, et al., 1998; Sidanius & Pratto, 1993b; Sidanius et al., 1992) provides a general theoretical way of understanding specific social and cultural phenomena without overly particularizing them. F. IDEOLOGICAL JUSTIFICATION O F SOCIAL PRACTICE This discussion leads us from SDO to the next big piece of the puzzle: social ideologies. As our cultural studies imply, social dominance theory postulates that specific ideologies justify specific social practices. Hence, more than the stereotyping literature, our research has attended to how ideologies justify institutional discrimination and other general social policies. If hierarchy-enhancing ideologies are perceived as relevant and potent, then the more people believe in them, the more they should favor hierarchyenhancing social practices and the more they should oppose hierarchyattenuating social policies. Conversely, if hierarchy-attenuating ideologies are perceived as relevant and potent, then support for them should predict support for hierarchy-attenuating social policies and opposition to
218
FELICIA PRATI’O
hierarchy-enhancing social policies. My colleagues and I have tested these hypotheses in numerous samples, employing many kinds of ideologies and attitudes towards many kinds of social policies (Pratto, Stallworth, & Sidanius, 1997; Sidanius et al., 1992; Sidanius & Pratto, 1993a, 1993b). I present two examples from a paper by Pratto, Stallworth, and Conway-Lanz (1998). Our first example examines attitudes toward “guns and butter” policies. We theorized that the “butter” programs (government social programs like unemployment insurance, affirmative action, and public libraries) attenuate social inequality because they either serve everyone or assist downtrodden people especially. As such, these policies should appeal to people low on SDO and be bolstered by hierarchy-attenuating ideologies. We argued that the “guns” programs, including military programs and going to war to maintain or regain dominance, are hierarchy-enhancing, and so will be supported more by those high on SDO, especially when justified by hierarchy-enhancing ideologies, In February 1992, we measured 115 college students’ S D 0 5 scores, and their attitude toward the ongoing war against Iraq and toward a variety of social welfare programs (see Pratto, Stallworth, & Conway-Lanz, 1998, for details). The legitimizing ideologies we measured were noblesse oblige, that those with more should help those with less, nationalism, political-economic conservatism, and anti-Black racism. We tested the fit of a model in which SDO predicted support for ideologies, and ideologies predicted support for social policies. We also allowed participant gender to predict SDO because we hypothesize that men are higher on SDO than women are (see Pratto, Stallworth, & Sidanius, 1997; Pratto et al., 1994; Sidanius, Cling, & Pratto, 1991; Sidanius, Pratto, & Bobo, 1994, Sidanius, Pratto, & Brief, 1995). The gender difference on SDO was expected to account for gender differences in the policy attitudes (cf. Pratto, Stallworth, & Sidanius, 1997; Pratto, Stallworth, & ConwayLanz, 1998). The standardized regression coefficients from our path model and the fit statistics are presented at the top of Figure 1. The social dominance theory model described the data accurately. People higher on SDO were more racist, more nationalistic, and more conservative than people lower on SDO, but people lower on SDO believed more in noblesse oblige. These ideologies, in turn, predicted support for both social policies. As expected, people who supported noblesse oblige supported social welfare programs, whereas the more racist and conservative participants were, the more they opposed these programs. Further, both conservatism and nationalism predicted support for the war against Iraq. Note that not all ideologies related to both policies. We anticipated this because we believe that social discourse in the United States has made social welfare programs both “conservative” and “racist” issues, but it has
A
Participant sex
SDT Model x'(18)=26 13
p = 10 GFI = 941 AGFI= 882
B
1-71 Partlclpant
26"
Alternative model x'(18)= 30.85
p < 03 GFI= 933 AGFI = .866
Fig. 1. Social dominance theory (SDT) model and alternative model of support for social programs and war. SDO, social dominance orientation; GFI, goodness-of-fit index; AGFI. adjusted goodness-of-fit index. N = 105; *p < 0.05: **p < 0.01, and ***p < 0.001.
220
FELICIA PRA’ITO
not made war a racist issue, nor one relevant to noblesse oblige. Ideologies must be made relevant to social practices and policies in order to legitimize or delegitimize them. As is typical in our studies, alternative models with conservatism rather than SDO as the root value failed to fit the data (see bottom of Figure 1). We have found comparable results concerning SDO, conservatism, ideologies, and social policies in several other samples (see Mitchell & Sidanius, 1993; Pratto, Stallworth, & Conway-Lanz, 1998; Sidanius & Pratto, 1993a; Sidanius et al., 1992). Another survey we conducted showed that SDO can orient people toward new legitimizations for social policies in contemporary social discourse. When President Bush nominated Clarence Thomas to the Supreme Court in spring of 1991, he declared that this nominee was the best-qualified person for the job, invoking the meritocracy ideology, although the American Bar Association did not give Thomas its highest qualification rating usually conferred upon Supreme Court nominees. Clarence Thomas’s nomination could reasonably be considered a hierarchy-enhancing move because Thomas, an outspoken opponent of affirmative action, was nominated to replace Justice Thurgood Marshall, whom we would classify as a hierarchyattenuator because of his legal work (e.g., as a civil rights attorney for the NAACP). The nomination proceedings progressed with fairly little controversy until reporter Nina Totenberg discovered that Thomas had been accused of sexual harassment while he chaired the Equal Employment Opportunity Commission. In response, the Senate Judiciary Committee subpoenaed a former subordinate of Thomas’s at the EEOC, law professor Anita Hill, and numerous other witnesses over televised weekend hearings. Under oath, Hill testified about several specific harassing actions that Thomas had taken toward her, but a prominent member of the Judiciary Committee, Sen. Arlen Specter, suggested that Hill either was lying or was psychologically incapable of distinguishing fantasy from reality. The committee proceeded with its previously scheduled vote, recommended Thomas, and the Senate immediately confirmed Thomas as a Supreme Court Justice. Having assessed our participant pool’s SDOS scores and general support for meritocracy 6 weeks earlier, we telephoned each member and asked four questions concerning the nomination during the following 3 days. Our response rate of 100%for the 149 people we reached demonstrates the high level of interest this confirmation had then garnered. Participants were asked to indicate how much they agree or disagreed with four statements: (1) After Thurgood Marshall’s retirement from the Supreme Court, it was good that George Bush appointed a Black person to the Court; (2) After Thurgood Marshall’s retirement from the Supreme Court, it was good that George Bush appointed a conservative to the Court, (3) If I were in the Senate, I would have voted in favor of Clarence Thomas’s
SOCIAL DOMINANCE THEORY
221
confirmation to the Supreme Court; and (4) Anita Hill was telling the truth in claiming that Clarence Thomas sexually harassed her. We considered their answers to question 3 their policy attitudes. Though the other three questions are not full-blown ideologies, they represent the kind of legitimizations for positions that social discourse can create. We modeled support for Thomas’ nomination as shown in Figure 2. Reminiscent of George Bush’s invocation of meritocracy, those who supported this ideology favored Thomas’s Confirmation. Perhaps as Senator Specter feared, those who thought Anita Hill was telling the truth opposed Thomas’s confirmation. Interestingly, as a Black conservative, Thomas was favored by people favoring a Black nominee and by people favoring a conservative nominee. That these are not usually the same set of people is illustrated in that people low on SDO favored the appointment of a Black and people high on SDO favored appointment of a conservative. Our model fit the data. This study documents both our theoretical propositions concerning SDO, ideologies, and social policies. One, people’s SDO levels orient them toward policies with implications for social inequality, including political appointments, and two, social discourse can give people reasons to justify their policy stances. Pratto, Stallworth, and Sidanius (1997) found support for a similar model concerning a political appointment in a survey of voters in
SDT Model
x 2 (11) = 17.35 p = .10 GFI = .967 AGFI = .917
Fig. 2. Social dominance theory (SDT) model of support for Clarence Thomas’s nomination to the U S . Supreme Court. GFI, goodness-of-fit index. N = 149; *p < 0.05, **p < 0.01; and ***p < 0.001.
222
FELICIA PRA’ITO
the 1992 presidential election. In that sample, SDO oriented people toward many of the social policies discussed in the election, and voters’ perceptions of the candidates’ positions on those policies was strongly related to their presidential vote. The kinds of studies previously described have the virtue of examining real and consequential social policy attitudes and political behavior in viva But how compelling they are theoretically also hinges on whether one accepts our assertion of what makes policies seem hierarchy-enhancing or hierarchy-attenuating. There is considerable policy analysis research that backs up our assumptions about the kinds of policies or programs that enhance or attenuate social hierarchy (see Mitchell & Sidanius, 1993;Pratto, Stallworth, Sidanius, & Siers, 1997; Sidanius, Liu, Pratto, & Shaw, 1994), but people can always argue about the meaning of social policies. In fact, shared social discourse is what gives policies meaning. To illustrate this point, Pratto, Orton, Cathey, and Stallworth (1998) conducted persuasion experiments in which we varied the meaning of social policies using persuasive arguments. We expected that if a policy was framed in a hierarchyenhancing way, it would appeal more to high SDO participants, whereas if the same policy was framed in a hierarchy-attenuating way, it would appeal more to low SDO participants. Moreover, we predicted this whether the persuasive message argued for or against the policy. In Experiment 1, we recruited participants in the top or bottom third of a student distribution of S D 0 6 scores, which were highly reliable (a = .91) and highly correlated with the SD06 scores assessed at the end of the experiment ( r = .92). We gave a brief description of our new and ambiguous social policy, that “the U.S. would require all new immigrants to go through an intensive learning program before entering mainstream American society,” and then measured opinions about this policy on a 7-point (strongly oppose to strongly support) scale. This initial policy attitude was uncorrelated with previously measured SDO, r = .15, n.s. Participants then read a more detailed, one-paragraph description of the policy, followed by one of four essays about the policy. They then indicated their current opinion of the policy on the same scale. The hierarchy-enhancing (HE) essays played on differentially benefiting the in-group and maintaining group distinctions and status boundaries. Specifically, the HE essay favoring the policy argued that the policy would reduce the tax burden on Americans created by immigrants using welfare and second-language programs, and prevent the weakening of the United States by allowing in only immigrants capable of success and who would not “dilute” American culture. The HE essay opposing the policy argued that the program would teach immigrants to be dependent on Americans,
SOCIAL DOMINANCE THEORY
223
that the funds for this program would be better spent on Americans, and that the program would give an unfair advantage to immigrants. The hierarchy-attenuating (HA) essays played on differentially benefiting the outgroup, reducing group distinctions and status differences. Specifically, the H A essay favoring the policy argued that the policy would benefit the low-status outgroup, immigrants, and consequently the whole country, by increasing their job opportunities and by providing the aid necessary in dire economic circumstances. The H A essay opposing the policy said that the program would take away the unique contributions immigrants make to the United States, that the funds for this program should be used to help immigrants through other means, and that the program would harm immigrants by stigmatizing them. The contents or principles of the arguments were similar across type of essay (HE or HA) and persuasion direction (for or against). For example, both opposing essays questioned the efficiency in using funds for the policy, and both favoring essays argued that the policy would decrease unemployment among immigrants. We analyzed attitude toward the policy after the experiment with analysis of variance (ANOVA) using initial policy opinion as a covariate ( r = .61, p < .001).9We obtained a two-way interaction between type of essay (HE or HA) and direction of persuasion (for or against), F (I, 75) = 4.56, p < .04. This interaction is an essay effect: participants liked the policy best when they had read H A arguments favoring the policy o r HE arguments opposing the policy, rather than H A arguments against or H E arguments in favor. In addition, we obtained the expected 3-way interaction between SDO level (low/high), type of essay, and direction of persuasion, F (1, 75) = 5.55, p < .02. When participants had read essays favoring the policies, the H A essay garnered stronger favoritism towards the policy among low SDO participants than high SDO participants, whereas the H E essay persuaded high SDO participants to like the policy more than low SDO participants. Conversely, for essays against the policy, the H A essay garnered more opposition to the policy among low than among high SDO participants, whereas the HE essay garnered more opposition among high than among low SDO participants (see adjusted means in Figure 3). Followup tests showed that the low SDO participants were most persuaded by the essays in the expected directions. Among low SDO participants, the two-way interaction between essay type and essay direction was reliable, F (1, 75) = 10.34, p < .005. Not surprisingly, low SDO participants liked the policy more when they had read H A arguments in favor of it rather than opposed to it, F (I, 75) = 8.45,p < .005. But low SDO participants also appeared to react against the H E arguments, liking the policy better The same results obtained without using the covariate in Experiments 1 and 2.
224
FELICIA PRATTO Essays favoring policy
Essays opposing policy
4 -
LowSDO HighSDO
d 2
3
-a
.-s 0
3 -
2 -
Hierarchyenhancing
Hierarchyattenuating
Hierarchyenhancing
Hierarchyattenuating
Fig. 3. Postexperimental attitude toward ambiguous immigration policy by message frame, direction of persuasive message, and participants’ social dominance orientation (SDO) level.
when they read HE arguments against it rather than in favor of it, F (1, 75) = 2.88, p < .05, one-tailed. Among high SDO participants, there were no reliable effects. We replicated these results in Experiment 2 in which each participant read essays for and against the policy. The order of whether the favoring or opposing essay was read first was counter-balanced and did not influence the results. Again, preopinion did not correlate reliably with SDO, r = .17, n.s. We again covaried initial opinion from the postexperiment attitude ( r = .67, p < .01) and found essentially the same essay effect indicating that the participants liked the policy better when they had read the HA support/HE oppose essay set rather than the H E support/HA oppose essay ~ .003. In addition, we found the expected interaction set, F(1,114) = 8 . 9 4 , < between SDO level and essay set, F (1,114) = 4 . 7 4 , ~= .03. Simple effects tests confirmed that low SDO participants liked the policy better when they read the HA support/HE oppose essay set (adjusted M = 4.79), than
22s
SOCIAL DOMINANCE THEORY
when they read the HE support/HA oppose essay set (adjusted M = 3.47) F(1, 111) = 13.43, p < .001, whereas high SDO participants were uninfluenced by the essay set, F < 1. In Experiment 2, participants also rated how strong a summary of each argument was (see Table V for items and reliabilities). We analyzed their mean ratings of the S HE arguments and S H A arguments they had read
TABLE V A N D HIERARCHY-An.ENUA-rlNCARGUMENTS FOR AND AGAINST NEW HIERARCHY-ENHANCING IMMIGRATION POLICY ANDSCALERELIABILITIES" Hierarchy-enhancing arguments for policy, (Y = .hl This policy helps to decrease welfare dependency among immigrants. This policy will lower costs created by second-language requirements. This policy will prevent the spread of foreign disease from other countries to the United States. This program would ensure that only those who are willing to adjust to American culture and learn the English language will make it in our country. In order for our country to be a world power, we need all U S . residents to share common beliefs, traditions, and love of country. Hierarchy-attenuating arguments for policy, (Y = .h7 The education provided in this program will create increased job opportunities for immigrants. This program will give immigrants a foothold to further education and the ability to cope in a foreign culture. This policy would help to reduce current inequalities existing between immigrants and native-born U S . citizens. This program will provide knowledge and skills that will help immigrants feel more at home in a foreign land. This program will reduce the poverty and hopelessness that is prevalent among immigrants. Hierarchy-enhancing arguments against policy, (Y = .80 Successful immigrants can make it on their own and we shouldn't make it easier for those who don't have what it takes to be successful here. The funds used for this program would be better spent on native-born U.S. citizens. American society gives each individual a chance to succeed using one's own merit and determination. This program may cause immigrants to become overly dependent on government support. If immigrants are given free handouts, they may never learn to take care of themselves. Hierarchy-attenuating arguments against policy, 01 = .82 This program limits our opportunities to learn from other cultures. This policy wrongly implies that American culture is better than that of immigrants. This program stigmatizes immigrants by implying that they are unfit for American society. It is inhumane to require people to undergo a learning program against their will. The limited funds would be better spent on helping immigrants with their basic needs. ~
~~
-
~~
~
~
" Participants rated how strong each argument was from 1 (very poor argument) to 5 (very good argument).
226
FELICIA PRATTO
in the essays in a mixed ANOVA by order of arguments, SDO level, essay set, and argument type rated. The ratings for the HA arguments ( M = 3.33) were more positive than those for H E arguments ( M = 2.43), F (1, 111) = 63.74, p < .001, regardless of which essay set they read. However, this effect interacted with SDO level, F (1, 111) = 15.46, p < .001, as expected (see Figure 4). Simple effects showed that low SDO participants rated the HA arguments higher ( M = 3.54) than high SDO participants ( M = 3.111, F (1, 111) = 7.14, p < .01, and high SDO participants (M = 2.62) rated the HE arguments higher than low SDO participants ( M = 2.24), F (1, 111) = 5.58, p < .025. No other effects obtained, indicating that the SDO X argument type interaction was not influenced by order nor even by which essay set was read. This pair of experiments shows that it is not simply the policy itself that appeals differentially to high and low SDO people, but rather, that policies can be given meaning through social discourse, and that meaning mediates the policy’s reception in different audiences.
Arguments favoring policy
Argumeuts opposhg policy
LowSDO
HighSDO
Hierarchyenhancing
Hierarchyattenuating
Hierarchyenhancing
Hienrchyatteuuating
Fig. 4. Ratings of hierarchy-enhancing and hierarchy-attenuating arguments for ambiguous immigration policy by participants’ social dominance orientation (SDO) level.
SOCIAL DOMINANCE THEORY
227
Experiment 3 demonstrated the ecological validity of this claim in August 1998, by using a policy proposed by Congressional legislation (HR 2646), public funding for school vouchers, and H E and H A arguments derived from the floor debate and testimony of expert witness. Similar to Experiment 2, participants were provided with a description of the proposed policy, then read an essay favoring and an essay opposing the policy, in counter-balanced order. Participants were randomly assigned to essay condition, either reading a HE-for and a HA-against essay or a HA-for and a HE-against essay. We used eight basic principles in these essays.’” These principles were balanced across the HE/HA dimension and direction of essay so that every participant read one argument based on each principle. Participants rated how their opinion about their congressional representative would change if they learned their representative had voted for and had voted against the bill on scales from much more negative (1) to much more positive (7). The difference between these two ratings was our primary persuasion measure. Participants next rated the strength of each of the eight arguments in the essays (four for and four against) on scales from very weak (1) to very strong (4), completed the SD06scale, and described themselves in demographic terms. We expected SDO to influence acceptance of these arguments, which would in turn influence policy opinions. The ratings of arguments in a given essay were highly intercorrelated (a’s ranged from .70 to .85 for the 4 arguments), so we averaged them. The 58 participants were adult members of charitable clubs in rural Arkansas who ranged in age from 25 to 87, with a median age of 50. Most were well-educated professionals. Their SDO scores were internally reliable ( a = 3 2 ) . As expected, men were higher on SDO ( M = 3.09) than women (M = 2.52), F (1, 57) = 5.39, p = .02, and those who came of age during the most recent civil-rights era (those aged 45 to 58) were substantially lower on SDO ( M = 2.55) than those older and younger ( M = 3.12), F (, 54) =8.22,~ = . O M . These participants differed from our college populations in that this group included more men, was higher on SDO, and the SDO distribution was not positively skewed (skew = .45). We used regression to analyze the results because we had not preselected participants based on SDO levels. We expected SDO to interact with essay condition in influencing judgments of the arguments, which we expected in turn to influence opinions about the congressional representative. These expectations were confirmed. Using the difference between the mean rating of the against arguments and the mean rating of the for arguments as the dependent variable, and SDO, essay condition, and their interaction as “I The principles were freedom of choice, rewarding ambition, reducing problems associated with welfare and crime, failing to address the real problem, maintaining U S . dominance. equality of opportunity, undermining a cherished American institution, and our nation’s traditional support for equality.
228
FELICIA PRATTO
predictors, we found no SDO effect, p = .23, only a marginal condition effect, /3 = 2 2 , t (53) = 1.74, p < .09, and the expected SDO X condition interaction, p = -.37, t (53) = -2.91, p = .005. The interaction effect indicates that the HE-for essay pair made the policy appeal more to people the higher they were on SDO, whereas the HA-for essay pair made the policy appeal more to people the lower they were on SDO. We then tested whether assessments of the arguments influenced opinions about the congressional representative. The difference between opinions about the representative voting for and against the bill was the dependent measure, and the regression equation used the following predictors: the mean ratings of the for and the against arguments, SDO, condition, and the interaction of condition with all other predictors. Results showed that the only reliable predictors of opinions of the representative were ratings of the for arguments, /3 = .39, t (49) = 2.95, p = .005, and ratings of against arguments, /3 = -.61, t (49) = - 4 . 1 2 , ~< .001. The more positive participants’ assessments of the arguments in favor of the policy were, the more their opinion of a representative voting for the policy improved, and conversely for their assessments of the arguments against the policy. Because SDO influenced assessments of arguments, but did not directly influence opinions of the representative, these results confirm that the argument frame is critical in mediating the influence of SDO on political opinions relevant t o the policy. Figure 5 illustrates this mediational process in path models in which assessments of the arguments mediate the influence of SDO on opinion following nearly complementary patterns in each condition. A t the top of Figure 5, the standardized regression weights show that in the HE-for/HAagainst condition, the higher people were on SDO, the more they liked the HE-for arguments and the less they liked the HA-against arguments. These evaluations largely determined their opinions about politicians, and SDO did not reliably predict opinions directly. As shown in the bottom of Figure 5, in the HA-for/HE-against condition, the higher people were on SDO, the more they liked the HE against arguments, although SDO was not reliably negatively related to evaluations of the H A for arguments as we had expected. However, again, evaluations of both sets of arguments determined opinion about the politicians, and SDO was unrelated. Like the survey about Clarence Thomas’s nomination, this study demonstrates that real policy debates can be framed in different ways to appeal to high or to low SDO people by appealing to their values concerning group equality or inequality. Such policy debate demonstrates that social practices are not inherently meaningful, but rather, that their meaning for intergroup relations is created through social discourse. The studies presented thus far suggest that when psychological orientations, social ideologies, and social conditions are compatible with
229
SOCIAL DOMINANCE THEORY HE-FodHA-againsfcondilion A
- ~ .
arguments for Bill Opinion change toward Representative
Social dominance orientation -.58***
49
85
HA-FodkLE-againsr condition
B
Fig. 5. Path models showing the influence of social dominance orientation (SDO) on evaluations of persuasive arguments for and against a bill and their influence on opinions of congressional representatives by essay condition. HE, hierarchy-enhancing; HA, hierarchyattenuating. N = 57: * p < 0.05; **p < 0.01; ***p < 0.001.
group dominance, group dominance is likely to be realized in social practice. That is, a general compatibility among group inequality, hierarchyenhancing social ideologies, and high levels of SDO, helps to reify group dominance. But group dominance systems also contain people low on SDO and hierarchy-attenuating ideologies and social practices that work against group dominance. These forces, however, do not always work as well against group dominance as they might because they do not always function in concert. I will illustrate this point in a series of studies showing asymmetries between how SDO and in-group identification produce group discrimination in high and low status groups. Obviously, not everyone in group dominance societies comes to approve of group dominance. In fact, certain positions may make people suspicious
230
FELICIA PRATTO
of the rightness of group dominance. For example, I once heard an interview with political reporter Molly Ivins in which she was asked how a White woman from Texas could become so ardent a liberal. Ivins explained that when she was growing up under segregation, she asked her mother why she couldn’t drink out of the “colored” water fountain, and her mother answered that the colored fountain was dirty. But Ivins could plainly see that the white fountain was far grimier. So her answer to the question was that “Once you figure out they’re lying to you about race, you figure they might be lying about everything else too.” I expect that people in subordinate groups are especially positioned to perceive legitimizing ideologies as lies (or at least as excuses). For example, it must be difficult for hard-working Black people to accept their own laziness, and smart women to accept their own stupidity, as reasons for their groups’ subordination. Moreover, their desire for positive self-esteem is at odds with subordinate group’s subordination and should motivate them to reject group dominance, especially if they identify with their in-group. People in dominant groups, though, will have identity motivations that lead them to be more accepting of legitimizing ideologies and of group dominance in general. Because of such processes, social dominance theory predicts that members of subordinate groups will have lower levels of SDO than members of dominant groups do (Sidanius, Pratto, & Rabinowitz, 1994). Table VI presents some of these status-group differences on SDO among U.S.college students surveyed by Choudhury and Pratto (1996).l’ For all comparisons, people in dominant groups, including men, Euro-Americans, and heterosexuals, were higher on SDO than people in subordinate groups, including women, Hispanic- and African-Americans, and gays, lesbians, and bisexuals, respectively. Such group differences appear robust: African-Americans have lower SDO levels than Euro-Americans both among Los Angeles Police officers (Sidanius, Liu, et al., 1994) and in a random sample of Los Angeles County (Sidanius & Pratto, 1999). Levin (1996) also found SDO levels to increase with ethnic group status in Israel: Ashenazi Jews had the highest levels, followed by Mizrachi Jews, mixed Ashkenazi-Mizrachi Jews, Arab-Israelis (ethnic Arabs with Israeli citizenship), and finally Palestinians living in the West Bank and Gaza. Childs, Woollaston, Mee, and Gallagher (1995), found that among New Zealand police officers, Whites had higher SDO levels than Maori and other subordinated ethnic group members. In such group status differences, we see that one’s position in the social hierarchy may influence one’s psychological orientation towards that hierarchy. Shana Levin (1996) has produced the most compelling evidence that I ’ This study used some new samples as well as some of those used by Sidanius, Pratto, and Rabinowitz (1994). but Choudhury and Pratto (1996) used the SDO scales subsequently validated by Pratto et al. (1994).
23 1
SOCIAL DOMINANCE THEORY TABLE VI MEANSOCIAL DOMINANCE ORIENTATION LEVELBY GROUP STATUS A N D SAMPLE" Sample
Group status designation
Subordinate
Dominant
~
I 2 3 4 5
Sexual orientation
1
Gender
2.4
2.8
2
Gender
2.5
3.0
Race
2.4
2.6
Race
2.4
2.6
Race
2.0
2.9
Race
1.7
2.3
1.9
2.3
3
Gender
2.2
2.7
4
Gender
1.9
2.3
5
Gender
2.1
2.6
6
Gender
1.6
2.0
7
Gender
2.0
2.4
All paired comparisons differed a t p < .05. For race, subordinates were African-Americans and Hispanic Americans and dominants were Euro-Americans. For sexual orientation, subordinates were self-identified as gay, lesbian, or bisexual, and dominants were self-identified as heterosexual. For gender, subordinates were women and dominants were men. Data from Choudhury and Pratto (1996).
it is the context of group status that creates these group differences in SDO levels. Within a survey of social attitudes among Israeli Jews, Levin positioned random halfs of the SD06 scale after questions about relations either between Ashkenazi and Mizrachi Jews or between Israeli Jews and Palestinians. Considering Palestinians put all these participants in a dominant position, whereas only Ashkenazis are in a higher status position when considering the AshkenadMizrachi distinction. Thus, Levin (1996) predicted that in the context of intra-Jewish group differences, Ashkenazi Jews would have higher SDO levels than Mizrachi Jews, but that in context of Jewish-Palestinian relations, all participants would express higher levels of SDO, and the difference between Ashkenazis and Mizrachis would disappear. Her results supported all three hypotheses, suggesting that being in a dominant group increases SDO levels.** These group differences in SDO levels that group hierarchy appears to create in turn help re-create social hierarchy. This is because the psychological motivators of out-group discrimination are both stronger and more
'*
Despite the experimental manipulation, individuals' SDO scores were consistent, with the two halves of the scale correlating reliably, r = S6, with each other.
232
FELICIA PRATTO
coordinated in dominant than in subordinate groups. Two motivators that increase discrimination in minimal groups experiments are (1) high SDO (Sidanius, Pratto, & Mitchell, 1994) and (2) identifying with one’s in-group (Perreault & Bourhis, in press). But people who are high on SDO and more identified with their in-group are especially group discriminatory (Sidanius, Pratto, & Mitchell, 1994). Choudhury and Pratto (1996) found that the double-barrel-high SDO coupIed t o high in-group identification-characterizes dominant but not subordinate groups. Among dominant groups, SDO correlates positively with differential in-group identification (i.e., greater identification for in-group than out-group). But for people in subordinate groups, SDO is unrelated or is negatively related to in-group identification (see Table VII). In subordinate groups, the barrels aren’t aimed in the same direction. So, the motivators of intergroup discrimination, namely higher levels of SDO, and higher levels of SDO linked to positive in-group identification, are more typically found in dominant than in subordinate groups. These studies suggest that people in dominant groups will discriminate to their own group’s advantage more than people in subordinate groups. When dominant groups behave in ways that advantage their TABLE VII REGRESSION SLOPES RELATING SOCIAL DOMINANCE ORIENTATION AND DIFFERENTIAL INGROUP AFFILIATION BY GROUP STATUS^ Sample dominant group Whites 1 2 4
5 6&7
6&7 5
Subordinate group
df
p-value
3.02 3.77 2.19
439 22s 472
,0005 .COO5 ,0005
2.52 2.73
249 501
.0025
3.00 7.03
567 247
.0025 ,0005
Blacks and Latinos
.16
-.15
.18
- .05
.15 Men
- .07 Women
.21 .12
-.18 -.17
Heterosexuals
Gays, lesbians, bisexuals
.I1 .13
2-test
- .25
- .70
,005
r-tests compare the size of the regression coefficients between dominant and subordinate groups. Betas were unstandardized regression coefficients from regressing social dominance orientation scores on composite group identification (in-group-out-group) scores. Data are from Choudhury and Pratto (1996).
SOCIAL DOMINANCE THEORY
233
in-group more than subordinate groups do, one clearly has a recipe for continued group domination. G. SOCIAL DOMINANCE ORIENTATION, GROUP CONTEXT, AND GROUP DISCRIMINATION The differences between dominant and subordinate groups in in-group identity, SDO, and their relation suggest that group discrimination is a joint function of the social context people are in and their psychological predispositions. A series of experiments by Pratto, Shih, and Orton (1998) demonstrate this point further by examining how SDO and group context predict group discrimination in two minimalist experimental paradigms. The first, developed by Perdue, Dovidio, Gurtman, and Tyler (1990, Experiment 2), measures implicit group discrimination and uses generic in-group and out-group labels (e.g., us and them) to automatically prime evaluations of good and bad trait words. Perdue et al. (1990) found an interaction between the group-prime and the valence of the target words, such that in-group pronouns facilitated good compared to bad evaluations, with the opposite (though unreliable) trend for outgroup pronouns. This suggests that people implicitly and automatically think of their in-groups as “good,” but not their outgroups. Part of the elegance of this paradigm is that there is nothing about the procedure that demands behavior, as it is devoid of the rich social-contextual cues about groups that characterize much realgroup behavior. Moreover, it does not invoke specific ideologies or stereotypes as racism measures do, and so it is ideal for exploring a general orientation towards group relations such as SDO. In Experiment 1, Pratto, Shih, and Orton (1998) selected undergraduate participants from the top or bottom third of their class distribution on S D 0 6 scores to participate. We replicated Perdue et al.3 (1990) group prime X target valence interaction, F (1,35) = 5.14, p < .03, but we found no hint of an interaction with participants’ SDO level, F < 1. Low and high SDO participants were implicitly in-group-biased to the same extent (see the amount of bias by prime, target, and SDO group on left side of Figure 6). On reflection, we suspected that the reason this experimental paradigm did not exhibit SDO differences is that the experimental situation was so minimal, having only abstract, quickly presented stimuli relevant to groupsthat there was scarcely any aspect of the context to engage the different psychological orientations towards groups that SDO represents. Other implicit priming paradigms have similarly failed to show a correspondence between participants’ expressed attitudes toward racism and their degree
234
FELICIA PRATTO
90
80
-
-
70
I
60
I
50 40 30
20
0In-group Prime
out-group Prime
Experiment 1
-
-
I
I
10 I 0 -10
-20
-30 -40
-50
-60 -70
-
I
Low SDO
High SDO
Low SDO
High SDO
Fig. 6. Implicit discrimination by group prime, social dominance orientation (SDO) level, and experiment.
of implicit race discrimination (Devine, 1989, Experiment 1;Fazio, Jackson, Dunton, & Williams, 1995, Experiment 1 for White participants). But related experiments that included stronger race cues have found a correspondence between explicit attitudes and implicit attitudes or discrimination (Lepore & Brown, 1997; Locke, MacLeod, & Walker, 1994; Wittenbrink, Judd, & Park, 1997). Such cues may simply provide a richer stimulus context to invoke different psychological orientations towards groups. In our second experiment, we retained the elegance of using the minimalist, automatic discrimination paradigm that did not demand behavior, but added enough group context to engage different orientations towards social dominance. We did this by having participants first read an essay threatening the status of a real high-status group to which participants belonged, namely Stanford University students. The essay, which was purportedly written by another Stanford student, discussed a number of scandals plaguing Stanford,
SOCIAL DOMINANCE THEORY
235
and reasoned that these scandals indicate that Stanford students were neither as intelligent nor as mature as they were reputed to be. After this ostensibly unrelated task, low and high SDO participants performed the same implicit group-discrimination task as in Experiment 1. In Experiment 2, there was a reliable 3-way interaction among group-prime, target valence, and SDO level, F (1, 41) = 4.97, p < .003. Among high SDO participants, there was a reliable and very large interaction between prime and target, F (1, 41) = 21.14, p < .001, and it was nearly symmetrical, with in-group pronouns facilitating evaluations of good words, and out-group pronouns facilitating evaluations of bad trait words (see right side of Figure 6). Among low SDO participants, however, the prime X target interaction was not reliable, F (1, 41) = 1.90, n.s., nor was any other effect. Experiment 2 therefore demonstrated that high SDO people were more automatically discriminatory than low SDO people-so long as there are more than minimal group cues to which they can respond. Although Experiment 2 shows that SDO is an orientation that is responsive to group context, such as group status concerns, it doesn’t show that people high and low on SDO actually behave in ways that influence groups differently. After all, implicit discrimination may or may not lead to explicit discrimination, and indeed a number of theorists have argued for a dissociation between the two (Devine, 1989; Fazio et al., 1995; Greenwald & Banaji, 1995). To test whether people low and high on SDO explicitly discriminate differently in an intergroup situation, but still in a paradigm that does not rely on particular social ideologies or behavioral demands, we used the minimal groups paradigm. In Experiment 3, we selected undergraduates high and low on SDO to participate using the same criteria as above, and told them that they had been selected to participate in a decision-making study exploring two kinds of cognitive style. The distinction between these two minimal groups were made highly ambiguous and indeed tenuous. For example, the instructions reminded participants that they were part of the same superordinate group (Stanford students), and said that the experiment would test whether there are any differences between the two cognitivestyle groups, and that the experimenter would not describe what is known about those styles until the end of the experiment so as not to influence the results. The participants then allocated points between a hypothetical person in their own cognitive-style group and the other group using allocation matrices used by Brewer and Silver (1978). This set of questions allows one to compute how many times participants use each of four orthogonal point allocation strategies: dividing points as equally as possible between groups (equality), giving the most points possible to one’s in-group (ingroup gain), allocating the most points possible to both groups (joint gain), and maximizing one’s in-group’s gain over the out-group’s (relative gain).
236
FELICIA PRATTO
Because this last strategy pertains most to group differentiation, this strategy was the one we expected to show SDO differences. Following Experiment 2, we expected the difference on use of the relative gain strategy between people high and low on SDO to be larger when the group context of this minimalist paradigm was made more salient. For this reason, there were two conditions of the experiment: low group salience and high group salience, to which participants were randomly assigned. In the high group salience condition, participants read an updated Stanford status-threat essay before allocating points. In the low group salience condition, participants read a different essay by a Stanford student about travel accomodations. True to expectations, there were no significant differences on the first three allocation measures except that low SDO participants used the strategy of maximizing joint gain (M = 2.16) more than high SDO participants (A4 = 1.62), F (1, 59) = 4.55, p < .04.On the relative gain strategy that differentiates the groups, the expected results were found; the model using the 3-way interaction among SDO level, condition, and use of the strategy fit the data, L2 (6) = 7.67, p = .26. There were no SDO differences in the low group salience condition, x2 (1) = .17, n.s. About two-thirds (69%) of both high and low SDO participants used the strategy. This result mimics the results of Experiment 1in which group discrimination appears normative across SDO level. In the high group salience condition, however, most high SDO participants used the relative gain strategy (72%), but few low SDO participants did (38%), x2 (1) = 4.14, p < .04,mimicking the results of Experiment 2. These results suggest that when the context has stronger group stimuli, the different orientations towards group dominance that low and high SDO people have will be reflected in their overt discriminatory behavior. Low SDO people will not discriminate based on group membership, but high SDO participants will do so to maximize group differences, even at the expense of absolute gain for their in-group. Finally, to show that it is not just something special about the Stanford group threat essay, but rather, how much the social context invokes people’s orientations towards groups that brings out these SDO differences, Pratto, Shih, and Orton (1998) conducted another minimal groups experiment using the same procedure to select high and low SDO participants. We reasoned in Experiment 4 that heightening the salience of the minimal group distinction, which is thought to increase group discrimination in this paradigm (Brewer, 1979), would make the group context salient enough to obtain SDO differences. The experimental procedure therefore made the “cognitive style” group distinction salient in several ways. For example, the experimenter’s instructions repeatedly mentioned the well-validated group differences we were studying, we placed bright neon yellow or orange stickers on participants’ forms to denote their group, and we induced partici-
SOCIAL DOMINANCE THEORY
237
pants to perseverate on the group distinction by having them form words from the letters in “COGNITIVE STYLE.” Given these and other procedural methods of making the minimal group distinction highly salient, we expected SDO differences in use of the relative gain strategy. Further, we reasoned that reducing the relative salience of the minimal groups distinction by having participants read the Stanford threat essay would reduce SDO differences on the relative group gain measure. And this is just what was found. In the high group salience condition, 90% of the high SDO participants used the strategy to maximize group differences, whereas only 22% of the low SDO participants did, a reliable difference, x2 (1) = 8.93, p < .001. In the low group salience condition, there were no SDO differences, x2 (1) = .09. About 35% of both groups used the strategy. Again, a model with the three-way interaction among SDO level, condition, and strategy use fit the data, L2 (6) = 7.98, p = .24. No effects were found on the other three measures. In total, the four experiments by Pratto, Shih, and Orton (1998) show that the social group context people are in is highly likely to bring out their psychological readiness to create or to attenuate group dominance through discriminatory b e h a ~ i 0 r . l ~
H. SOCIAL DOMINANCE ORIENTATION, SOCIAL IDEOLOGIES, AND HIERARCHY ROLES The theme not only of our minimal experiments, but also of our cultural studies, policy attitude surveys, and persuasion experiments is that SDO is most likely to result in group discrimination when people are in a social context that contains social ideologies or other cues that spur group discrimination. To understand the dynamics of institutional discrimination, my colleagues and I have explored what kinds of occupational contexts people low and high on SDO obtain. In general, social dominance theory expects people higher on SDO to obtain occupational roles that enhance social hierarchy and people lower on SDO to obtain occupational roles that attenuate social hierarchy. Occupations that generally channel things of positive social value, such as money, status, health, or education, more towards dominant than subordinate people, or channel things of negative social value such as punishments and lousy neighborhoods more towards subordinates than dominants, are classified as hierarchy-enhancing. Occupations that generally channel positive resources more towards subordinates than dominants, that is, discriminate against l 3 Such results have been independently corroborated by Kemmelmeir, Cameron, and Chaiken (1998). who found similar interactions between context and SDO on a race discrimination measure.
238
FELICIA PRATTO
social hierarchy, are classified as hierarchy-attenuating. Because of the considerable evidence of different forms of institutional discrimination, police, prosecutors, and business executives are considered hierarchyenhancing roles, and public defenders, charity workers, civil rights workers, and social workers are considered hierarchy-attenuating roles (see Pratto, Stallworth, Sidanius, & Siers, 1997). In one of our first studies testing whether people in such occupations do indeed have different levels of SDO, Sidanius, Liu, Pratto, and Shaw (1994) compared police officers with members of the general public (those called to jury duty) and with public defenders in Los Angeles. As expected, we found that SDO levels decreased from the police to the general public to the public defenders. In an even broader survey of the general public, Pratto, Stallworth, Sidanius, and Siers (1997) classified a sample of voters’ occupations into hierarchy-enhancers, middlers, and hierarchy-attenuators, and again found a linear trend relating these roles to their SDO levels (see Figure 7). Several processes conspire to sort people with “appropriate” SDO levels into hierarchy roles. The first is self-selection. Jobs like CIA agent and prosecutor appeal to high SDO people, whereas jobs like public defender and civil rights lawyer appeal to low SDO people (Sidanius, Pratto, Sinclair, & van Laar, 1996). Pratto, Stallworth, Sidanius, and Siers (1997, Experiment 1)tested in a more controlled fashion whether it is the hierarchy aspect of occupations that differentially appeals to those high and low on SDO, rather than other job features like the duties or salaries. In Experiment 1, college students, who are making decisions about their future careers, indicated what occupation they intend, and later, completed the S D 0 6 scale and numerous work values scales. Participants also chose between a HE job description and a hierarchy-attenuating one with the same salary and duties within 10 occupations (e.g., advertising, fundraising, law, librarian). The fact that their preference for HE versus H A jobs corresponded to our classification of their intended real occupations as HE, middling, or H A validates our classification of real occupations along those lines. In addition, SDO strongly predicted people’s preferences for H E versus H A occupations ( r = .58), and did so even when all the other work values were considered (Experiment 1).Experiment 2 replicated the match between high or low SDO level and selection of hierarchy role ( r = .70, see Figure 8). We have considerable evidence that self-selection is one process thai leads high SDO people into H E occupational roles and low SDO people into H A ones. But it is not the only one. Institutions also select people who have values compatible with the goals of the institution. Thus, people who appear low on SDO would be more likely to be hired into H A occupations and people high on SDO into H E ones. Pratto, Stallworth, Sidanius, and Siers (1997)
SOCIAL DOMINANCE THEORY
Hierarchyenhancers
Middlers
239
Hierarchyattenuators
Occupation as hierarchy role Fig. 7. Social dominance orientation level by occupational hierarchy roles among 400 1992 voters.
tested this by first having college students (Experiment 3 ) and then business people (Experiment 4)make hiring decisions. Provided with the same kinds of HE and H A job descriptions as above, they also were given resumis of four recent college graduates who pretests showed appeared to be either high or low on SDO. As expected, the high SDO appkants were recommended more often for H E jobs, and the low SDO applicants were recommended more H A jobs. The bias ratios were substantial: about 1.5 to 1 in
240
FELICIA PRATTO
200
4 1 LowSDO
175
High SDO
150
125
100
75
50
Hierarchyenhancing
Hierarchyattenuating
Hierarchy role within occupation Fig. 8. Selection of hierarchy-enhancingand hierarchy-attenuatingjobs within occupations by social dominance orientation (SDO) level. N = 50 Stanford students.
Experiment 3 and 2 to 1 in Experiment 4, showing fairly powerful assortment for applicants who were identical in credentials and experience, save for what connoted their SDO levels, like having been camp counselor for an elite summer camp (high SDO) versus for Head Start (low SDO). These hiring experiments also showed a third process by which higherSDO people get HE jobs and lower-SDO people get HA jobs, and that is sex discrimination. Even with the same resumks, men were more often hired for H E jobs and women for HA jobs at a ratio of over 1.5 to 1. Because in all known populations, men are slightly higher on SDO than women, this process would also contribute to assortment by SDO into hierarchy role. In addition, people whose group prejudice (e.g., racism) conforms to the normative attitude of their occupational context get more
SOCIAL DOMINANCE THEORY
241
positive feedback (van Laar, Sidanius, Rabinowitz, & Sinclair, 1999), so differential attrition may also contribute to this outcome. We suspect that the fact that several different processes lead to compatibility between institutional contexts and their members’ prejudice makes the effect quite robust. I. SOCIAL IDEOLOGIES AS MEDIATORS OF INDIVIDUAL PSYCHOLOGIES AND SOCIAL CONTEXT Once people low and high on SDO obtain different hierarchy roles, their different ideological habits guide them to allocate social resources in different ways. In addition to ideological biases employees bring with them, the context of hierarchy roles also promotes systematic discrimination because some hierarchy roles invoke and may even demand discriminatory ideologies. As a dramatic example, the Stanford prison experiment (Haney, Banks, & Zimbardo, 1973) showed that normal young men randomly assigned to play prison guards severely discriminated against other young men randomly assigned to play the role of prisoners by humiliating and beating them and by withholding food. People low and high on SDO discriminate in opposite ways with respect to hierarchy both because of the way their personal ideological habits guide their behavior and because their hierarchy roles invoke different ideologically scripted discriminatory behavior. Pratto, Tatar, and Conway-Lanz (1999) tested both these expectations in experiments in which we examined ideologies participants relied on to allocate resources. Our experiments reenacted a common ideological debate. In the United States several social debates relevant to group inequality, such as affirmative action programs, how public education funds are distributed, and who should get tax breaks, are framed by opposing the general ideal of rewarding merit against the general ideal of alleviating need. Asking people to chose between a party who is more meritorious and one who is needier therefore poses a dilemma between the two ideologies. Our allocation scenarios posed just this dilemma. For example, one scenario asked participants to imagine they were a professor at the end of the term, asked to write letters of recommendation for two good students: one was the best in the class, and the other a transfer student who did not know many professors from whom she could get a recommendation. With time to write only one letter, the professor had to choose which letter to write. We wrote five other scenarios which also pitted a more meritorious party against a needier party. In all, three scenarios asked participants to give a resource to one party, and three asked participants to remove a resource from one of the
242
FELICIA PRATTO TABLE VIII SUMMARY OF RESOURCE ALLOCATION SCENARIOS' Resource decision
______~
Merit party feature
Need feature (lack of)
__________~
Locate new federal office in one of two states
Better facilities
Higher unemployment
Allow new grocery store to be built in one of two neighborhoods
Higher tax revenues Better business risk
No grocery in neighborhood
Close one of two high schools
Better grades, test scores
No other community facilities
Which student to write recommendation for
Superior performance in course
Cancel newspaper service from one paperboy
Provides better service
Transfer student; knows no other professors Paper route supplements low-family income
Withdraw deposits from one of two banks
Better interest rate, more locations, less risky
a
Smaller assets
Participants rated which allocation they preferred from 1 to 6 .
parties (see Table VIII for a summary of each scenario). After rating which allocation they would prefer, from 1 to 6, participants described the reasons for their decisions. Because the ideology of rewarding merit implies that some people should be treated better than others, it is a HE ide~logy,'~ so we expected it to appeal especially to those high on SDO. On the other hand, being in need puts one in a subordinate position compared with people who aren't, and so alleviating need can be a HA act, and should therefore appeal to those low on SDO. Further, Deutsch (1975) argued that when people presume they are in communal relationships with others, something more characteristic of low SDO people (Pratto et al., 1994), they will want to alleviate need rather than promote merit. For these reasons, we expected SDO to predict how people allocated resources between more meritorious and more needy parties. In fact, participants preselected to be high or low on SDO did have different opinions about whether allocating to meritorious or needy parties was most fair (Experiments 1 and 2), violating the assertion that merit or l 4 When the Protestant ethic arose, this merit-ideology may well have been less hierarchyenhancing than the doctrine used to bolster the current but failing aristocratic regime, the Divine Right of Kings. Such changes in the historical context of this important ideology and its likely relation to SDO are discussed by Sidanius and Pratto (1993a) and Pratto, Tatar, and Conway-Lanz (1999).
SOCIAL DOMINANCE THEORY
243
equity is a general psychological principle (cf. Homans, 1976; Walster, Berscheid, & Walster, 1973), and demonstrating that meritocracy is an ideology linked to social values (see also Martin & Murray, 1983). Moreover, in Experiment 3, high and low SDO people allocated the resources to different parties: low SDO participants favored the needy party ( M = 2.76) and high SDO participants favored the merit party ( M = 4.09), F (1, 30) = 22.86, p < .001. The reasons participants gave for their allocations invoked the need or merit ideologies. For example, for the letter-writing scenario, a merit reason one participant wrote was “Jane did better (maybe worked harder) than Anne, so she deserves rec.” and a need reason was “If Jane can get the recommendation elsewhere and Anne cannot, Anne needs the recommendation more.” In total, high SDO participants wrote more merit reasons, F (1, 30) = 6.52, p < .02, and low SDO participants wrote more need reasons, F (1,30) = 4.24, p < .05. These results confirm our expectation that high and low SDO people do allocate real-not just minimal-resources differently, and suggest that this occurs because they have different ideological habits. In social contexts that reinforce or at least fail to override people’s ideological habits, these results suggest that people will allocate resources differently depending on their SDO level. But social contexts such as occupational roles can suggest or even demand the use of particular ideologies, which will influence how people playing those roles allocate resources. Moreover, practicing those ideologies in role may create the habit of using those ideologies out of role. To demonstrate this, in Experiment 4, participants practiced either the need or the merit ideology as part of an occupational task, and then allocated resources “as themselves” out of role. We expected people in the merit ideology condition to then favor the merit parties in the second task, and people in the need ideology condition to then favor the needy parties. In fact, we expected this ideological inculcation to override SDO differences in these two conditions. Only in the control condition, in which participants were not primed with any particular ideology, did we expect to replicate the association between resource allocation and SDO. In this experiment we were not able to preselect people based on extreme SDO scores, so we administered the SD06scale after the experiment. All participants were in a business decision-making study in which they read two pairs of loan applications, one for a small business loan and one for a large business loan, and decided which applicants should receive the loans. In the merit ideology condition, participants were told they were loan officers in a commercial bank in a local well-off city, and in the need ideology condition, participants were told they were loan officers in a community development agency in a local poor city. The role instructions
244
FELICIA PRATTO
further invoked each ideology. The bank loan officers were reminded of the general idea of rewarding merit when they were told that banks make money by lending money to people who can pay it back, and their job was to pick the loan applicant who was most credit-worthy and deserved the loan. The two loan decisions in that condition helped bring meritorious thinking to mind by pitting one meritorious loan against another. The community development agency loan officers were reminded of the ideology of alleviating need when told that the mission of the community lending agency is to lend money not just to help the loan recipient, but others in the community, and that their job was to pick the loan applicant who needs the loan to provide goods and services to the community. The two loan decisions in that condition pitted one loan designed to alleviate a need against another designed to alleviate a different need to invoke thinking along the need dimension. In the control condition, participants were not given either type of “ideological” instructions, but simply told that the job of a banker is decide how to lend money out, and then selected one loan applicant from the merit pair and one from the need pair. Following the loan decisions, participants were told that we were interested in how they make decisions on their own, and then were provided with the first three scenarios in Table VIII, which we averaged. As expected, participants in the merit condition favored the merit parties most ( M = 3.30), participants in the need condition favored the need parties most ( M = 2.77), with a reliable difference between them, F (1,85) = 4.24, p < .001, and the control condition in the middle ( M = 3.08). Clearly, then, both merit and need ideologies can be invoked through occupational agendas and are general enough to apply to very different decision scenarios. In this experiment, practice applying the ideologies during loan decisions carried over to decisions about school closing, city zoning laws, and locations of government agencies. We also showed that SDO differences in allocations were apparent only in the control condition; regression showed SDO to influence allocations strongly in that condition, /3 = .47, t (26) = 2.72, p = .01, but not in either other condition, p’s > .18. Similar results obtained for the number of merit and need reasons given by condition. The contrast between the two ideology conditions and the control condition shows that different SDO levels are associated with different ideological habits, but that social contexts such as hierarchy occupations can encourage the use of nonhabitual ideologies that will influence resource allocation. 111. The Completed Puzzle: How Group Dominance Assembles Itself
By summarizing the research presented here, I will show how social dominance theory integrates the major findings in the field, including social
SOCIAL DOMINANCE THEORY
245
and gender roles, stereotypes, prejudice, social identity, and discrimination, to illustrate a self-perpetuating dynamic system of group dominance. Most notably, I will reveal the heuristic or principle of social dominance theory that organizes the complex and interacting components of group dominance systems. Also, as I promised at the outset, fitting these pieces together provides a different understanding of them than work that examines each piece in isolation. In summarizing the present work, I will therefore revisit some questions and conclusions provided by previous research. Like any social psychologist, the way I addressed the question of why humans so often create group dominance societies was by considering the interplay between psychological states and social conditions. I postulated that living in the context of group dominance creates a general psychological orientation toward group dominance that then influences people to recreate group dominance. I also argued that how that general orientation, called SDO, is expressed, depends very much on the social condition one is in. This was shown in several different studies. First, the comparisons of SDO with reproductive strategies showed that in heterosexual men, SDO is associated with one set of behaviors and attitudes about sexual relations, whereas in heterosexual women, SDO is associated with a different but complementary set of behaviors and attitudes. In the “context” of each gender, then, SDO seems likely to facilitate different behaviors, but together, these behaviors contribute to intersexual dominance (limited sexual and economic power for women) and intrasexual dominance (stratification among men in status, power, and reproductive success). In this example, different expressions of SDO in people in complementary gender roles jointly contribute to the overall structure of group dominance societies (gender dominance and social stratification). This example also illustrates why group dominance societies must be considered dynamic systems: the behaviors of people at one level (e.g., interpersonal) influence the emergent properties of the system at another level (e.g., social structure), and vice versa. Further, my theory about reproduction links gender inequality and arbitrary group inequality psychologically, economically, politically, and sociostructurally, further illustrating that group dominance societies are dynamic systems with interactive components bound along different but interconnecting dimensions (e.g., Vallacher & Nowak, 1997). As such, my theory and data emphasize the active and political nature of sexual relationships, certain psychological gender differences, and family relationships (see Pratto, 1996a, for a fuller discussion), in contrast to theories that explain gender differences in reproductive strategies as “natural.” My second example of SDO being expressed differently in different contexts was the set of cultural studies relating ideologies to specific social practices and intergroup relations. There, SDO correlated with support for
246
FELICIA PRATTO
different local ideologies (e.g., different rationales for ethnic prejudice and sexism) that reify different particular instances of group inequality in different cultural contexts. Our theoretical postulate, that support for any consensually shared ideology that prescribes group relations will correlate with SDO, acknowledges the significance of both culture and ideology in establishing group dominance, but does not rely on culturally particular values or psychological agendas. In doing so, social dominance theory contrasts with numerous other theories about racism and sexism that hinge on the particular contents of cultural ideologies (e.g., Glick & Fiske, in press; Katz & Hass, 1988; Sears, 1988; Sniderman & Piazza, 1993), and also with structural theories of intergroup relations that do not incorporate shared meanings (e.g., Blumer, 1960; Tajfel, Billig, Bundy, & Flament, 1971). Similarly and third, social dominance theory recognizes that the way groups are defined (e.g., based on race, caste, nationality) varies culturally, but postulates that the relation of SDO to group status and to group affiliation follows a general pattern across type-of-group designation. I reviewed evidence that SDO is higher in higher-status groups than in lowerstatus groups across several group-status designations in the United States and in other countries. Further, I showed that the nature of the relation of SDO to in-group identification depends on group status. Among highstatus groups alone we saw the magic formula that leads to high levels of intergroup discrimination: high SDO coupled with high in-group identification. Among low-status groups, average SDO levels are not only lower, but even their high SDO members tend to identify more with dominant outgroups than with the ingroup. This pattern is one of the asymmetries between dominant and subordinate groups that contributes to the persistence of group inequality. In it, we again see that different and indeed complementary processes at one level contribute to the same emergent property of group inequality at another level. This research also showed that “basic” social identity processes are contingent on group status. Though social identity theory attempted to address this issue, its treatment has been problematic. Social identity theory posits that the motivation for positive social identity motivates in-group favoritism but not out-group denigration (Brewer, 1979). If any group is motivated to regain positive identity by discriminating, it should be the subordinate group, yet there is evidence that subordinate groups discriminate less in favor of their own group and even discriminate in favor of dominant outgroups (Hinkle & Brown, 1980). Our results suggest that the reason subordinate groups do not discriminate as much in favor of their own groups is because they are lower on SDO and even when high on SDO, are not identified with their in-groups. Another difficulty with social identity’s ap-
SOCIAL DOMINANCE THEORY
241
proach is that the motivation for positive identity does not explain why groups actively oppress and commit violence against each other (cf. Mummendey, 1995). By examining the dynamics of SDO, we uncovered at least part of a solution. In the implicit and explicit discrimination experiments, we found that for high SDO people, group discrimination is not simply motivated by positive feelings about the in-group, but by negative feelings about the out-group and a desire to triumph over it. Fourth, these experiments also illustrated that the way SDO is expressed depends on social context. In the most minimal contexts, high and low SDO people discriminated to the same extent. Other researchers have interpreted similar results as suggesting that individual differences on overt group prejudice measures do not really matter (Fazio et al., 1995), or that overt discrimination matters more because it alone is controllable (Devine, 1989). The distinction between implicit and explicit discrimination is critical to both points (see also Greenwald & Banaji, 1995), but we found that when the contexts had more salient cues about groups, SDO levels corresponded to people’s level of group discrimination. The similarity of our results across the implicit and explicit minimalist paradigms suggests a different interpretation: that the critical factor in predicting behavior is how strongly social context engages psychological orientations. We used two minimalist paradigms: implicit group discrimination and minimal groups, to make a point about the importance of context and of social cues to behavior. In fact, in most social contexts except minimalist ones, there are abundant cues about how people ought to behave with respect to intergroup relations. People’s physical appearance and behavior quite readily cue their socia1 status and group memberships. The social conditions typical of group dominance societies, such as group segregation and power differentials, invoke social ideologies that prescribe how people behave. Between social symbols and the social organization in which we live, and the group attitudes and stereotypes that live in us, the ideologies that justify and prescribe our social practices are everywhere. This brings us to the second major point about group dominance systems. What binds group dominance societies together are shared ideologies. Three parameters govern the effect of those ideologies on social inequality: (1) their meaning, particularly whether they are HE or HA; (2) their degree of consensuality; and (3) their accessibility. All three of these can be manipulated, but local cultural context usually fixes the first two. The accessibility of ideologies is influenced both by individuals’ mental habits, and their immediate social contexts (cf. Higgins & Bargh, 1987), most notably, aspects of the context of group dominance. As we saw in the review of stereotyping and several other studies, accessible, consensual ideologies contribute to both individual and institutional discrimination.
248
FELICIA PRATTO
In surveys of current events, we saw that ideologies appear to justify social policies that would influence group inequality. In hiring experiments, we saw that gender stereotypes contributed to gender discrimination in attainment of hierarchy roles. This discrimination not only influences men and women as individuals, but also contributes to institutional discrimination inasmuch as the higher-SDO gender obtains H E roles and the lowerSDO gender obtains H A roles. In the business role-playing experiment, we saw that practice playing H E or H A roles could invoke the use of H E or H A ideologies, which may not only lead to institutional discrimination while on the job, but also to personal discrimination outside the job. In the hypothesis that discrimination results from the influence of accessible ideologies, social dominance theory reconciles a theoretical dilemma. One set of theories has emphasized that resource allocations between groups are gestures towards the maintenance of symbolic group distinctions (e.g., Blumer, 1960;Sears, 1988; Tajfel &Turner, 1986). In contrast, realistic group conflict theory (e.g., Bobo, 1988; Campbell, 1965; Sherif & Sherif, 1966) emphasizes that group distinctions are fostered by competition over the allocation of material resources. Social dominance theory posits that through ideologically prescribed resource allocations, people transform symbolic ideals into material and social reality. Moreover, in our emphasis that such ideologies and resource allocation practices are embedded in cultural and institutional practices, we link what is often viewed only as an individual or dyadic process to institutional discrimination and social structure. I hope these comments make clear that putting the puzzle pieces of group dominance together can lead us to better understandings of the individual pieces. Now it is time to look at the whole puzzle. The assembled puzzle of group dominance is depicted in Figure 9, with arrows representing causal influences established with experiments, and interlocking nibs representing processes shown to co-occur. The model implies that group inequality does not exist merely because of cognitive biases, or because it is part of culture, or because of certain prejudiced personalities, or because of discrimination, but because all of these work together. Figure 9 does not show a causal chain with many links and some root cause, but rather, a dynamic model in which each process influences the others, either directly or indirectly. This means that there is some redundancy in the system. For example, we saw that high and low SDO people get assorted into different hierarchy roles through at least three separate processes. This overdeterminacy makes it unlikely that interrupting one process will break the whole system, adding to the system’s resiliency. The system is also robust because each process shown helps to perpetuate the others. Moreover, the processes feed back on themselves. For example, several theories posit that if people observe
SOCIAL DOMINANCE THEORY
249
Fig. 9. A dynamic model of group dominance.
group differences, they will form and use stereotypes, which can cause them to discriminate, which would result in group differences. Starting at any of these three stages (group differences, stereotypes, or discrimination) will start the whole feedback cycle, so a given component can assemble the system. This illustrates, fundamentally, why social dominance systems are so common and so robust, even in the face of social, technological, and economic change. In Caporael’s (1995) terms for described evolving social systems, social dominance systems “repeatedly assemble” themselves. In addition to generating new research that provides new interpretations of existing work, social dominance theory holds some lessons for social psychological theorizing. First, for complex phenomena (e.g., culture or group dominance or gender), reductionism is either misleading or only applicable once the problem has been severely truncated. Further, the linear thinking found in the pursuit of the root cause has theoretical costs
250
FELICIA PRATTO
because it presumes that causation only works in one direction. But as we have seen here, many co-acting processes do in fact mutually cause each other. Quite simply, researchers must not forget that an experiment that shows that A causes B does not rule out that B may cause A outside the experiment. Another problem with applying linear thinking to complex phenomena is that linear thinking only applies when one can assume that all relevant organisms and processes have a definitive and simultaneous starting point. But societies, species, and groups that gain and lose members continually have no definitive “start,” and there is no simplifying assumption one can make to get around this. The need to be able to incorporate history and development is fundamentally why social dominance theory has adopted a dynamic model. Reductionism, admittedly, has two appealing aesthetic properties. One is parsimony. But we must not confuse parsimony with reductionism. Dynamic models not only can but must have parsimony (see Vallacher & Nowak, 1997). For example, I analyzed the question of sexual politics, and considerable data pertinent to the subject, as hinging on a single heuristic: whether men’s and women’s child-rearing strategies diverge (Pratto, 1996a). Similarly, social dominance theory has many components, but they are all part of a mutually organizing and self-perpetuating system of social dominance. Dynamic theories therefore have a different parsimony than linear theories: recursion. In this, they lose the other aesthetic for which we admire some reductionistic theories: the “aha!” we get when the compelling, but heretofore hidden cause is revealed (e.g., Zajonc, 1965). But recursion is so elegant, and, as evolutionary theory exemplifies, so powerful, that we might forgive dynamic models for explaining how something causes itself. What is, of course, so parsimonious about recursive, dynamic models, is that they use the same heuristic at different levels of analysis (cf. Caporael, 1995). Like a fractal, the same principle or pattern operates at different levels and in all parts of the system. In social dominance theory, the principle is group inequality, which we have seen in SDO, in in-group favoritism, in adherence to legitimizing ideologies, in the acquisition of powerful roles, in the consequences of institutional discrimination, and even in how well the components of the system work together. Social dominance theory posits that group inequality, combined with people’s psychological nature, drives asymmetries in processes to favor dominant groups, which sustains group inequality. I hope that social dominance theory will encourage other social psychologists to develop dynamic models of other complex social psychological problems (see also Vallacher & Nowak, 1997) because the use of dynamic models may enable social psychologists to escape from their retreat into the “middle level” and the short-term. This would have several benefits.
SOCIAL DOMINANCE THEORY
25 1
It would enrich the intellectual contact our theories make with other disciplines like anthropology, sociology, and history, and with the ordinary lives of people. We are already seeing that using broader participant populations and examining culture and social structure can lead us to recognize when something we thought was a basic psychological principle is actually a legitimizing ideology (such as equity) o r a cultural artifact (such as the individualistic self). Taking larger perspectives would make us theorize about larger contexts. Theorizing about social context is, of course, the zeitgeist of social psychology, but social psychological work has also been criticized for failing to contextualize (e.g., Cherry, 1995; Sherif, 1994). The error in such theorizing is not simply in overgeneralizing from particular populations or conditions, but in failing to theorize what the relevant context is. It is with some irony, then, that I recall several social psychologists telling me that we (social psychologists) don’t theorize about social structure because that’s for so~iologists.’~ Psychology colleagues, both inside and outside of social psychology, have also implied that researching “political” topics like group dominance, sexual politics, and group differences is for unscientific disciplines like political science, feminist studies, and identity studies. Steering clear of researching political topics might seem to be more scientific and less political, but in fact it is equally political and less scientific because it obscures power differences. More importantly, the social structure of inequality is the context within which most humans live, and one humans create. If social psychologists don’t theorize about it, we aren’t doing our job. This is not to ignore concerns about whether theories about political topics may be politically biased. Far from it. In fact, because I take the notion of psychological ethnocentrism (cf. Bartlett, 1932) extremely seriously, I think it is essential that scientists consider how their theories may have been influenced by the normal social ideologies of their societies (Pratto, 1996b). Further, because I take the notion of legitimizing ideologies seriously, I think we have an obligation to consider what kind of ideologies our theories might become. I have been confronted with these problems in working on social dominance theory. I have found it a major ethical challenge to research group dominance without justifying it, although I did it because I felt that obfuscating group dominance was more unethical. In particular, I have worried that describing group dominance as an adaptive, well-functioning system (e.g., Sidanius & Pratto, 1993b) might seem to justify and even naturalize oppression. Our research has also confronted l5 Perhaps this only reaffirms what the feminist critique of science (cf. Harding, 1986) tells us: that we scientists are subject to human predilections such as ethnocentrism just like everyone else.
252
FELICIA PRATTO
us with the problem of explaining an apparently invariant gender difference (e.g., Sidanius, Cling, & Pratto, 1991; Sidanius, Pratto, & Brief, 1995; Sidanius, Pratto, & Bobo, 1996), without resorting to an inherency explanation, and evolutionary theory has confronted us with the problem of explaining how social hierarchy and dominance-striving, which clearly harm people, could be adaptive (see Pratto, 1996a). I hope that part of the solution to these political dilemmas is in the content of the theory. Another part is in its presentation, such as what facts we reveal and the symbolic and evaluative connotation of the terms used to describe those facts. Yet another part is in the data the theory generates. The objectivist tradition demands that one not reveal how one decided what to research, but on the optimistic assumption that the scientific community is by now too sophisticated to believe that hiding that process makes us objective, I will describe some of my decisions, in case they can serve as a good (or bad) example to other scientists grappling with similar concerns. The most obvious strategy my colleagues and I have taken is simply to name and draw attention to group dominance, even to the point of revealing some uncomfortable similarities between societies people think of as egalitarian (e.g., the Western democracies) and those they do not (e.g., dictatorships, empires, China; cf. Sidanius & Pratto, 1993a). Likewise, we (especially Sidanius, 1993; see also Sidanius, Liu, Pratto, & Shaw, 1994; Sidanius & Pratto, 1999, Chapter 8) have analyzed the operation of institutions popularly proclaimed to be the pinnacle of fairness as integral to group dominance, namely, the police and criminal justice system. Because group dominance is the modal form of social organization, our research has emphasized how normal psychologies and consensually approved social practices and beliefs sustain group dominance. For this reason, we have not theorized that people high on SDO are abnormal and fundamentally different (see Pratto et al., 1994). To do so would paint the psychological states that sustain group dominance as far more aberrant and unusual than they are. In all our studies, we have examined SDO in “normal” populations. In fact, the levels of SDO among even our “high” category (the top third of a sample) are not absolutely high on the SDO scales, nor are they statistically unusual-being no less common than the middle or low thirds. Extraordinarily high levels of SDO may not be necessary to the maintenance of group dominance. Rather, the everyday normal discriminatory social patterns of group dominance societies and tacit tolerance of oppression by everyone accomplishes much of the work that sustains group dominance. Just as we have not situated the cause of group dominance in a few deviant, abnormal, exceptionally high SDO people, we also have not examined really far-out, right-wing ideologies like the Unabomber’s manifesto
SOCIAL DOMINANCE THEORY
253
or the beliefs of the Ku Klux Klan as being central to the maintenance of group dominance. Rather, we have examined how very popular, socially acceptable, even standard cultural ideologies, like sexism, political conservatism, meritocracy, U.S. nationalism, and Israeli Zionism, justify H E social practices and group dominance. Such ideologies should, I believe, be seen in a new light when they are empirically identified as hierarchy-legitimizing myths. Another of our methods is to examine where power is situated. American individualism leads people to look for power in personal choices, for example, in the annoyingly perennial question, “What do women want‘?” Scientifically, it is easy to attribute group differences in outcomes solely to individual choice if one has evidence that a group psychological difference does cause some effect. For example, we showed that men and women chose different occupational roles because they differ in their SDO levels (Pratto, Stallworth, Sidanius, & Siers, 1997, Experiments 1 & 2). But to deliberately show that analyzing that outcome solely in terms of personal choice was inadequate, in the same paper, we also showed two other discrimination processes that led to the exact same pattern of gender segregation of hierarchy roles (Pratto, Stallworth, Sidanius, & Siers, 1997, Experiments 3 & 4). Likewise, we have examined both group differences in behavior and institutional discrimination as contributing to inequality (Sidanius & Pratto, 1999). Social dominance theory should remind people that the socialstructural contexts people live in alter the power of their controllable behavior to influence social hierarchy. If, for example, a well-educated adult wants to “share the wealth” with the next generation by tutoring neighborhood children, in all likelihood that adult will tutor White children because residential segregation typically links education levels to race (e.g., Zubrinsky & Bobo, 1996). Considering the intentions, values, and actions of this tutor without considering the tutor’s larger social context would indeed make the persistence of group inequality seem a puzzle. What I have done in this chapter is to fit together many psychological, social, and cultural forces that previous research suggested are part of the problem to show how these forces perpetuate group dominance in a systematic way. The processes illustrated in this research are socialpsychological, and the research methods highly familiar. But because social dominance theory integrates these processes into a dynamic system, it generates research and explains phenomena across levels that social psychology all too frequently ignores, including the society, family, group, and individual. At this point, social dominance theory describes the maintenance of group dominance, but needs formal expansion to incorporate and predict the chaotic changes that societies sometimes exhibit, such as the dissolution
254
FELICIA PRATTO
of Yugoslavia into an unstable, warring, re-ethnified disaster, and to discover what could help societies stabilize around greater equality. Neither agenda can be realized by ignoring, denying, or trivializing the problem of group dominance nor its systematic nature. I think the analysis we have developed thus far, examining emergent properties of societies by examining the social psychology of their members, holds considerable promise. In all, power is a complex system, one that deserves social psychology’s best, and not its least, attention.
Acknowledgment I thank Peter Hegarty, Tony Lemieux, Deborah G. Tatar, Christie Cathey, Reuben M. Baron, Jim Sidanius, Jill Santopietro. and David J. Pratto for their useful remarks on previous drafts of this chapter.
References Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). Theaufhorifarian personality. New York: W. W. Norton. Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg: University of Manitoba Press. Altemeyer, B. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco: Jossey-Bass. Altemeyer, B. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press. Altemeyer, B. (1998). The “other” authoritarian. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (Vol., 30, pp. 47-92). San Diego: Academic Press. Andersen, S. M.. & Klatzky, R. L. (1987). Traits and social stereotypes: Levels of categorization in person perception. Journal of Personality and Social Psychology, 53, 235-246. Ashmore, R. D. (1981). Sex stereotypes and implicit personality theory. In D. L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup behavior (pp. 87-81). Hillsdale, NJ: Erlbaum. Augoustinos, M., & Walker, I. (1996). Social cognition: An integrated introductiun. London: Sage. Ayres, 1. (1991). Fair driving: Gender and race discrimination in retail car negotiations. Harvard Law Review, 104, 817-872. Ayres, I. (1995). Further evidence of discrimination in new car negotiations and estimates of its cause. Michigan L a w Review, 94, 109-147. Ayres, I., & Siegelman, P. (1995). Race and gender discrimination in bargaining for a new car. American Economic Review, 85, 304-322. Bargh, J. A,, & Thein, R. D. (1985). Individual construct accessibility, person memory, and the recall-judgment link: The case of information overload. Journal of Personality and Social Psychology, 49, 1129-1 146.
SOCIAL DOMINANCE THEORY
255
Bartlett, F. (1932). Remembering. New York: Cambridge University Press. Bem, S . L. (1993). The lenses of gender. New Haven, CT: Yale University Press. Betzig, L. (1993). Sex, succession, and stratification in the first six civilizations: How powerful men reproduced, passed power on to their sons, and used power to defend their wealth, women, and children. In L. Ellis (Ed.), Social stratification and socioeconomic inequality: A comparative biosocial analysis (pp. 37-74). New York: Praeger. Blumer, H. (1960). Race prejudice as a sense of group position. Pacific Sociological Review, 1, 3-5. Bobo, L. (1988). Group conflict, prejudice, and the paradox of contemporary racial attitudes. In P. A. Katz & D. A. Taylor (Eds.), Eliminating racism: Profiles in controversy (pp. 85-114). New York: Plenum. Bodenhausen, G. V., & Wyer, R. S. (1985). Effects of stereotypes on decision-making and information-processing strategies. Journal of Personality and Social Psychology, 48, 267-282. Braithwaite, V. A., & Law, H. G. (1985). Structure of human values: Testing the adequacy of the Rokeach value survey. Journal of Personality and Social Psychology, 49, 250-263. Branscombe, N. R., & Smith, E. R. (1990). Gender and racial stereotypes in impression formation and social decision-making processes. Sex Roles, 22, 627-647. Brewer, M. B. (1979). In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-324. Brewer, M. B., & Caporael, L. (1990). Selfish genes vs. selfish people: Sociobiology as origin myth. Motivation and Emotion, 14, 237-243. Brewer, M. B., & Silver, R. (1978). Ingroup bias as a function of task characteristics. European Journal of Social Psychology, 8, 393-400. Brown. D. E. (1991). Human universals. New York: McGraw-Hill. Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12, 1-49. Buss, D. M. (1992). Mate preference mechanisms: Consequences for partner choice and intrasexual competition. In J. H. Barkow, L. Cosmides. & J. Tooby (Eds.), The adapted mind (pp. 249-266). New York: Oxford University Press. Campbell, D. T. (1965). Ethnocentric and other altruistic motives. In D. Levine (Ed.), Nebraska symposium on motivation (pp. 283-31 1). Lincoln, NE: University of Nebraska Press. Cantor, N., & Mischel. W. (1977). Traits as prototypes: Effects on recognition memory. Journal of Personality and Social Psychology, 35, 38-48. Caporael. L. (1995, February). Sociality: Coordinating bodies, minds, and groups [51 paragraphs]. Psycoloquy [on-line serial]. Available ftp: hostnarne:princeton.edu Directory: /cogsci.ecs.soton.ac.uk/pub/harnad/Psycoloquy/lY95/volume.6 File: psych.95.6.01.groupselection.l.caprorae1. Cherry, F. (1995). The stubborn particulars of socialpsychology: Essays on the research process. London: Routledge. Childs. M., Woollaston, S., Mee, T., & Gallagher, M. U. (1995). Socialization effects and gender differences in social dominance orientation within the New Zealand police. Unpublished manuscript. Christie. R., & Cook, P. (1958). A guide to published literature relating to the authoritarian personality through 1956. Journal of Psychology, 45, 1717-1799. Choudhury, P., & Pratto, F. (June, 1996). Sex, race, and sexual orientation: Status differences in attitudes about hierarchy. Paper at symposium on Social identity and group status: Implications for intergroup relations. Annual meeting of the American Psychological Society, San Francisco.
256
FELICIA PRATTO
Cochrane, R., Billig, M., & Hogg, M. (1979). British politics and the two-value model. In M. Rokeach (Ed.), Understanding human values (pp. 179-191). New York: Free Press. Collier, J. (1988). Marriage and inequality in classless societies. Stanford, CA: Stanford University Press. Cunneen, C. (1990). Aborigines and law and order regimes. Journal for Social Justice Studies: Special Edition Series, Contemporary Race Relations, 3, 37-50. Dahrendorf, R. (1968). Essays in the theory ofsociety. Stanford, C A Stanford University Press. DeprCt, E. F., & Fiske, S. T. (1993). Social cognition and power: Some cognitive consequences of social structure as a source of control deprivation. In G . Weary, F. Gleicher, & K. Marsh (Eds.), Control motivation and social cognition (pp. 176-202). New York: Springer-Verlag. Deutsch, M. (1975). Equity, equality, and need What determines which values will be used as a basis of distributive justice? Journal of Social Issues, 31, 137-149. Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled processes. Journal of Personality and Social Psychology, 56, 5-18. Devine, P. G., & Elliot, A. J. (1995). Are racial stereotypes really fading? The Princeton trilogy revisited. Personality & Social Psychology Bulletin, 21, 1139-1 150. Duckitt, J. (1989). Authoritarianism and group identification: A new view of an old construct. Political Psychology, 10, 63-84. Eagly, A. H. (1995). The science and politics of comparing men and women. American Psychologist, 50, 145-158. Eagly, A. H., & Kite, M. E., (1987). Are stereotypes of nationalities applied to both women and men? Journal of Personality and Social Psychology, 53, 451-462. Eagly. A. H., & Steffen, V. J. (1984). Gender stereotypes stem from the distribution of men and women into social roles. Journal of Personality and Social Psychology, 34, 590-598. Erber, R., & Fiske, S. T. (1984). Outcome dependency and attention to inconsistent information. Journal of Personality and Social Psychology, 47, 709-726. Eysenck, H. J. (1944). General social attitudes. Journal of Social Psychology, 19, 207-227. Eysenck, H. J. (1951). Primary social attitudes as related to social class and political party. British Journal of Sociology, 11, 198-209. Eysenck, H. J. (1971). Social attitudes and social class. British Journal of Social and Clinical Psychology, 10, 210-212. Eysenck, H. J. (1976). Structure of social attitudes. Psychological Reports, 39, 463-466. Fazio, R. H., Jackson, J. R., Dunton, B. C., & Williams, C. J. (1995). Variability in automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of Personality and Social Psychology, 69, 1013-1027. Feagin, J. R., & Feagin, C. B. (1978). Discrimination American style: Institutional racism and sexism. Englewood Cliffs, NJ: Prentice-Hall. Feingold, A. (1992). Gender differences in mate selection preferences: A test of the parental investment model. Psychological Bulletin, 112, 125-139. Fiske, A. P. (1991). Strucrure of social life. New York: Free Press. Fiske, S. T. (1993). Controlling other people: The impact of power on stereotyping. American Psychologist, 48, 621-628. Fiske, S. T. (1997). Stereotyping, prejudice, and discrimination. In D. T. Gilbert, S. T. Fiske, & G . Lindzey (Eds.), The handbook of socialpsychology (4th ed.) (pp. 3.57-411). New York: McGraw-Hill. Fiske, S. T., & Neuberg, S. L. (1989). A continuum of impression-formation, from categorybased to individuating processes: Influences of information and motivation on attention and interpretation. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 23. 1-108). New York: Academic Press. Fiske, S. T., & Taylor, S. E. (1991). Social cognition. New York: McGraw-Hill.
SOCIAL DOMINANCE THEORY
257
Frable, D. S. (1993). Being and feeling unique: Statistical deviance and psychological marginality. Journal of Personality, 61, 85-1 10. Gailey, C. W. (1987). Evolutionary perspectives on gender hierarchy. In B. B. Hess & M. M. Ferree (Eds.), Analyzing gender (pp. 32-67). Beverly Hills, CA: Sage. Glick, P., & Fiske, S. T. (1999). Sexism and other “isms”: Interdependence, status, and the ambivalent content of stereotypes. In W. B. Swann, Jr., J. Langlois, & L. A. Gilbert (Eds.), Sexism and stereorypes in modern society: The gender science of Janet Taylor Spence (pp. 193-221). Washington, DC: American Psychological Association. Goertzel, T. G. (1987). Authoritarianism of personality and political attitudes. Journal of Social Psychology, 127, 7-18. Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday. Gough, H. (1981). The California Personality Inventory. Palo Alto: Consulting Psychologists Press. Goertzel, T. G. (1 987). Authoritarianism of personality and political attitudes. Journal of Social Psychology, 127, 7-18. Greenwald, A. G., & Banaji, M. B. (1995). Implicit social cognition: Attitudes, self-esteem, and stereotypes. Psychological Review, 102, 4-27. Hamilton, D. L. (1979). A cognitive-attributional analysis of stereotyping. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 12, pp. 53-84). New York: Academic Press. Hamilton, D. L., & Gifford, R. K. (1976). Illusory correlation in interpersonal perception: A cognitive basis of stereotypic judgments. Journal of Experimental Social Psychology, 12, 392-407. Hamilton, D. L., & Rose, T. L. (1980). Illusory correlation and the maintenance of stereotypic beliefs. Journal of Personality and Social Psychology, 39, 832-845. Haney, C., Banks, C.. & Zimbardo, P. (1973). A study of prisoners and guards in a simulated prison. Naval Research Reviews, September 1-17. Washington, DC: Office of Naval Research, Department of the Navy. Harding, S. (1986). The science question in feminism. Ithaca: Cornell University Press. Herrnstein, R. J.. & Murray, C. A. (1994). The bell curve: Intelligence and class structure in American life. New York: Free Press. Higgins, E. T., & Bargh, J. A. (1987). Social cognition and social perception. Annual Review of Psychology, 39, 369-425. Hilton, J. L., & von Hippel, W. (1996). Stereotypes. Annual Review of Psychology, 47,237-271. Hinkle, S., & Brown, R. (1980). Intergroup comparisons and social identity: Some links and lacunae. In D. Abrams and M. Hogg (Eds.), Advances in socialidentity theory. New York Harvester Wheat Sheaf. Hofstede, G. (1982). Dimensions of national cultures. In R. Rath, H. S. Asthana, D. Sinha, & J. P. P. Sinha (Eds.), Diversity and unity in cross-cultural psychology (pp. 173-187). Swets & Zeitlinger B. V. Lisse. Homans, G. C. (1976). Commentary. In L. Berkowitz & E. Walster (Eds.), Advances in Experimental Social Psychology (Vol. 9, pp. 231-244). New York: Academic Press. Humphrey, R. (1985). How work roles influence perception: Structural-cognitive processes and organizational behavior. American Sociological Review, 50, 242-252. Jackson, D. N. (1965). Personality Research Form. Goshen, NY: Research Psychologists Press. Johnson, W. R. (1994). Dismantling apartheid: A South African town in transition. Ithaca, New York: Cornell University Press. Jussim, L., & Eccles, J. (1995). Naturally occurring interpersonal expectancies. In N. Eisenberg (Ed.), Social development. Review of personality and social psychology (pp. 74-108). Thousand Oaks. CA: Sage Publications.
258
FELICIA PRATTO
Katz, I., & Hass, R. G. (1988). Racial ambivalence and American value conflict: Correlational and priming studies of dual cognitive structures. Journal of Personality and Social Psychology, 55, 893-905. Kemmelmeier, M., Cameron. J., & Chaiken, S. (July, 1998). A n experimental test of social dominance theory: The power 4 prejudice hypothesis. Paper presented at the annual meeting of the International Society of Political Psychology, Montreal. Kerlinger, F. N. (1984). Liberalism and conservatism: The nature and structure of social attitudes. Hillsdale, NJ: Erlbaum. Kirscht, J. P., & Dillehay, R. C. (1967). Dimensions of authoritarianism: A review of research and theory. Lexington: University o f Kentucky Press. Kitzinger, C. (Ed.). (1994). Should psychologists study sex differences? Feminism & Psychology, 4, 501-546. Kluegel, J. R., & Smith, E. R. (1983). Affirmative action attitudes. Effects of self-interest, racial affect, and stratification beliefs on Whites’ views. Social Forces, 61, 797-824. Kozol, J. (1991). Savage inequalities: Children in America’s schools. New York: Harper Perennial. Lagendijk, J. (1986). Opinie onderzoek, het FEM Lagendijk ondernemerspanel, Raadpleging over het tweede kwartaal. Apeldoorn, Lagendijk’s Opinie Onderzoek. Lenski, G. E. (1984). Power and privilege: A theory of social stratifcation. Chapel Hill: University of North Carolina Press. Lepore, L., & Brown, R. (1997). Category and stereotype acivation: Is prejudice inevitable? Journal of Personality and Social Psychology, 72, 275-287. Levin, S. L. (1996). A social psychological approach to understanding intergroup attitudes in the United States and Israel. Doctoral Dissertation, Department of Psychology, UCLA. Levine, R. A., & Campbell, D. T. (1972). Ethnocentrism: Theories of conflict, ethnic attitudes and group behavior. New York: Wiley. Lewicki, P., Hill, T., & Czyzewska, M. (1992). Nonconscious acquisition of information. American Psychologist, 47, 796-801. Leyens, J. P., Yzerbyt, V., & Schadron, G. (1994). Stereotypes and social cognition. Thousand Oaks, CA: Sage. Linville, P. W., Fisher, G., & Salovey, P. (1989). Perceived distinctiveness of the characteristic of ingroup and outgroup members: Empirical evidence and a computer simulation. Journal of Personality and Social Psychology, 57, 165-188. Locke, V., MacLeod, C., & Walker, I. (1994). Automatic and controlled activation of stereotypes: Individual differences associated with prejudice. British Journal of Social Psychology, 33, 29-46. Martin. J., & Murray, A. (1983). Distributive justice and unfair exchange. In D. M. Messick & K. S . Cook (Eds.), Equity theory: Psychologicaland sociologicalperspectives (pp. 169-205). New York: Praeger. Mauer, M. (1990). Testimony of Marc Mauer, Assistant Director, the Sentencing Project, before the Subcommittee on Legislation and National Security and the Subcommittee on Government Information, Justice and Agriculture of the House Government Operations Committee, May 2, 1990, Washington, D.C. Mauer, M., & Huling, T. (1995). Young Black Americans and the criminal justice system: Five years later. The Sentencing Project. October. 918 F. St. NW, Suite 901, Washington, DC 20004. Marx, K., & Engels, F. (1970). The German ideology. New York: International Publishers. (Original work published 1848) Mazzella, R., & Feingold, A. (1994). The effects of physical attractiveness, race, socioeconomic status, and gender of defendants and victims on judgments of mock jurors: A metaanalysis. Journal of Applied Social Psychology, 24, 1315-1344.
SOCIAL DOMINANCE THEORY
259
McClosky, H. (1 9%). Conservatism and personality. American Political Science Review, 52, 27-45. McFarland, S. G.. & Adelson, S. (July. 1996). A n omnibus study of personality, values and prejudices. Paper presented at the Annual Convention of the International Society for Political Psychology, Vancouver, British Columbia. Merton, R. K. (1948). The self-fulfilling prophecy. Antioch Review, 8, 193-210. Michels, R. (1962). Political parties: A sociological study of the oligarchical tendencies of modern democracy. New York: Free Press. (Original work published 1911) Miller, D. T., Taylor, B., & Buck, M. L. (1991). Gender gaps: Who needs to be explained? Journal of Personality and Social Psychology, 61, 5-12. Mitchell, M., & Sidanius, J. (1993). Group status and asymmetry in the relation between ideology and death penalty support: A social dominance perspective. National Journal of Sociology, 7, 67-93. Mosca, G. (1939). The ruling class: Elements of political science. New York: McGraw-Hill. (Original work published 1896). Mummendey, A. (1995). Positive distinctiveness and social discrimination: An old couple living in divorce. European Journal of Social Psychology, 25, 657-670. Munnell, A,, Browne, L. E., McEneaney, J., & Tootell, G. M. B. (1992). Mortgage lending in Boston: Interpreting HMDA dam. Federal Reserve Bank of Boston. Working paper 92-07. Boston: Federal Reserve Bank of Boston, October. Murdock, G. (1956). Social structure. New York: Macmillan. Neuberg, S. L., & Fiske, S. T. (1987). Motivational influences on impression formation: Outcome dependency, accuracy-driven attention, and individuating processes. Journal of Personality and Social Psychology, 53, 431-444. Nias, D. K. B. (1972). The structuring of social attitudes in children. Child Development, 43, 21 1-219. Okin, S. M. (1989). Justice, gender, and the family. New York: Basic Books. Pareto, V. (1979). The rise and fall of the elites. New York: Arno. (Original work published 1901). Parker, I. (1992). Discourse dynamics: Critical analysis for social and individual psychology. New York: Routledge. Perdue, C. W., Dovidio, J. F., Gurtman, M. B., & Tyler, R. B. (1990). Us and them: Social categorization and the process of intergroup bias. Journal of Personality and Social Psychology, 59, 475-486. Perreault, S., & Bourhis, R. Y.(in press). Social identification, ethnocentrism, and discrimination. Personality and Suciaf Psychology Bulletin. Pettigrew. T. F. (1958). Personality and sociocultural factors in intergroup attitudes: A crossnational comparison. Journal of Conflict Resolution, 2, 29-42. Pettigrew, T. F. (1979). The ultimate attribution error: Extending Allport’s cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5, 461-476. Pratto, F. (1996a). Sexual politics: The gender gap in the bedroom, the cupboard and the cabinent. In D. M. Buss & N. M. Malamuth (Eds.). Sex, power and conflict: Evolutionary and feminist perspectives (pp. 179-230). New York: Oxford University Press. Pratto, F. (1996b). Ignoring inequality at our (or their?) peril: Overnorming and theories as ideologies. Paper presented at symposium on Right and Wrong: Social Psychological Insights into Ethics and Morality, at the annual meeting of the American Psychological Society, San Francisco. Pratto, F., & Bargh, J. A. (1991). Stereotyping based on apparently individuating information: Trait and global components of sex stereotypes under attention overload. Journal of Experimental Social Psychology, 27, 26-47.
260
FELICIA PRA'ITO
Pratto, F., & Hegarty, P. (1998). The political psychology of reproductive strategies. Unpublished manuscript, University of Connecticut, Storrs, CT. Pratto, F., Liu, J. H., Levin, S., Sidanius, J., Shih, M., Bachrach, H., & Hegarty, P. (1998). Social dominance orientation and the legitimization of inequality across cultures. Unpublished manuscript, University of Connecticut, Storrs, CT. Pratto, F., Orton, J., Cathey, C., & Stallworth, L. M. (1998). Issue-framingandsocial dominance orientation in policy persuasion. Unpublished manuscript, University of Connecticut, Storrs, CT. Pratto, F., Shih, M., & Orton, J. (1998). Social dominance orientation and group context in implicit and in explicit group discrimination. Unpublished manuscript, University of Connecticut, Storrs, CT. Pratto, F., Sidanius, J., & Stallworth, L. M. (1993). Sexual selection and the sexual and ethnic basis of social hierarchy. In L. Ellis (Ed.), Socialstrat@cationand socioeconomic inequality: A comparative biosocial analysis (pp. 111-137). Westport, C T Praeger. Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741-763. Pratto, F., Stallworth, L. M., & Conway-Lanz, S. (1998). Social dominance orientation and the legitimization of policy. Journal of Applied Social Psychology, 28, 1853-1875. Pratto, F., Stallworth, L., & Sidanius, J. (1997). The gender gap: Differences in political attitudes and social dominance orientation. British Journal of Social Psychology, 36,49-68. Pratto, F., Stallworth, L., Sidanius, J., & Siers, B. (1997). The gender gap in occupational role attainment: A social dominance approach. Journal of Personality and Social Psychology, 72, 37-53. Pratto, F., Tatar, D., & Conway-Lanz, S. (1999). Who gets what and why? Determinants of social allocations. Political Psychology, 20, 127-150. Ray, J. J. (1973). Conservatism, authoritarianism, and related variables: A review and an empirical study. In G. D. Wilson (Ed.), The psychology of conservatism (pp. 17-35). London: Academic Press. Ray, J. J. (1983). A scale to measure conservatism of American public opinion. Journal of Social Psychology, I 19, 293-294. Rokeach, M. (1960). The open and closed mind. New York: Free Press. Rokeach, M. (1973). The nature of human values. New York: Free Press. Rokeach, R. (1979). Understanding human values. New York: Free Press. Rosenberg, M. ( f 965). Society and the adolescent self-image. Princeton: Princeton University Press. Sanday, P. R. (1981). Femalepower and male dominance. Cambridge, UK: Cambridge University Press. Schwartz, S. H. (1994). Are there universal aspects in the structure and contents of human values? Journal of Social Issues, 50, 19-45. Searing, D. D. (1979). A study of values in the British House of Commons. In M. Rokeach (Ed.), Understanding human values (pp. 154-178). New York: Free Press. Sears, D. 0.(1988). Symbolic racism. In P. A. Katz & D. A. Taylor (Eds.), Eliminating racism: Profiles in controversy (pp. 53-84). New York: Plenum. Sears, D. 0. (1993). Symbolic politics: A socio-psychological theory. In S. Iyengar & W. J. McGuire (Eds.), Explorations in political psychology (pp. 113-149). Durham, NC: Duke University Press. Sherif, M., & Sherif, C. W . (1966). Groups in harmony and tension: A n integration of studies on intergroup relations. New York: Octagon Books.
SOCIAL DOMINANCE THEORY
261
Sherif, C. W. (1994). Bias in psychology. In A. S. Herrmann & A. J. Stewart (Eds.), Theorizing feminism: Parallel trends in the humanities and social sciences (pp. 117-133). Boulder, CO: Westview Press. Shields, S. (1975). Functionalism, Darwinism. and the psychology of women. American Psychologist, 30, 739-754. Shils, E. A. (1954). Authoritarianism: “Right” and “left.” In R. Christie & M. Jahoda (Eds.), Studies in the scope and method of’“The Authoritarian Personality.” Glencoe, IL: Free Press. Sidanius, J. (1987). Social attitudes and political party preferences among Swedish youth. Scandinavian Political Studies, 10, 111-124. Sidanius, J. (1993). The psychology of group conflict and the dynamics of oppression: A social dominance perspective. In S. Iyengar & W. McGuire (Eds.), Explorations in political psychology (pp. 183-219). Durham, NC: Duke University Press. Sidanius, J., Cling, B. J., & Pratto, F. (1991). Ranking and linking as a function of sex and gender role attitudes. Journal of Social Issues, 47, 131-149. Sidanius, J., Devereux, E., & Pratto, F. (1992). A comparison of symbolic racism theory and social dominance theory: Explanations for racial policy attitudes. Journal of Social Psychology, 132, 377-395. Sidanius, J., & Ekehammar, B. (1979). Political socialization: A multivariate analysis of Swedish political attitude and preference data. European Journal of Social Psychology, 9,265-279. Sidanius, J.. Ekehammar. B., & Ross, M. (1979). Comparisons of socio-political attitudes between two democratic societies. International Journal of Psychology, 14, 225-240. Sidanius, J., Liu, J., Pratto, F., & Shaw, J. (1994). Social dominance orientation, hierarchyattenuators and hierarchy-enhancers: Social dominance theory and the criminal justice system. Journal of Applied Social Psychology, 24, 338-366. Sidanius, J., & Pratto, F. (1993a). Racism and support of free-market capitalism: A crosscultural analysis. Political Psychology, 14, 383-403. Sidanius, J., & Pratto, F. (1993b). The dynamics of social dominance and the inevitability of oppression. In P. Sniderman, P. E. Tetlock, & E. G. Carmines (Eds.), Prejudice, politics, and race in America today (pp. 173-211). Stanford, CA: Stanford University Press. Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory of social hierarchy and oppression. New York: Cambridge University Press. Sidanius, J., Pratto, F., & Bobo, L. (1994). Social dominance orientation and the political psychology of gender: A case of invariance? Journal of Personality and Social Psychology, 67, 998-1011. Sidanius. J., Pratto, F., & Bobo, L. (1996). Racism, conservatism, affirmative action, and intellectual sophistication: A matter of principled conservatism or group dominance? Journal of Personality and Social Psychology, 70, 476-490. Sidanius, J., Pratto, F.. & Brief, D. (1995). Group dominance and the political psychology of gender: A cross-cultural comparison. Political Psychology, 16, 381-396. Sidanius, J., Pratto, F., Martin, M., & Stallworth, L. M. (1991). Consensual racism and career track: A social dominance perspective. Political Psychology, 12, 691-720. Sidanius, J., Pratto, F., & Mitchell, M. (1994). Ingroup identification, social dominance orientation, and differential intergroup social allocation. Journal of Social Psychology, 134, 151-167. Sidanius, J., Pratto, F., & Rabinowitz, J. (1994). Gender, ethnic status, ingroup attachment and social dominance orientation. Journal of Cross-Cultural Psychology, 25, 194-21 6. Sidanius, J., Pratto, F., Sinclair, S., & van Laar, C. (1996). Mother Teresa meets Genghis Khan: The dialectics of hierarchy-enhancing and hierarchy-attenuating career choices. Social Justice Research, 9, 145-170.
262
FELICIA PRA’ITO
Sirgo, H. B., & Eisenman, R. (1990). Perceptions of governmental fairness by liberals and conservatives. Psychological Reports, 67, 1331-1334. Sniderman, P. M., & Piazza, T. (1993). The scar of race. Cambridge, MA: Harvard University Press. Snyder, M., & Uranowitz, S. W. (1978). Reconstructing the past: Some cognitive consequences of person perception. Journal of Personality and Social Psychology, 36, 941-950. Steele, C. M., & Aronson, J . (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69, 797-811. Stephan, W. G . (1987). The contact hypothesis in intergroup relations. In C. Hendrick (Ed.), Group processes and intergroup relations. Review of personality and social psychology (pp. 13-40). Beverly Hills: Sage. Stone, W. F. (1980). The myth of left-wing authoritarianism. Political Psychology, 2, 3-19. Stone, W. F., & Smith, L. D. (1993). Authoritarianism: Left and right. In W. F. Stone, G . Lederer, & R. Christie (Eds.), Strengths and weaknesses: The authoritarian personality today (pp. 144-156). New York: Springer-Verlag. Sweeney, L. T., & Haney, C. (1992). The influence of race on sentencing: A meta-analytic review of experimental studies. Behavioral Sciences and the Law, 10, 179-195. Tajfel, H., Billig, M., Bundy, R., & Flament, C. (1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 5, 5-43. Tajfel, H., & Turner, J. C. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G . Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago: Nelson-Hall. Taylor, S. E., Fiske, S. T., Close, M., Anderson, C., & Ruderman, A. (1977). Solo status as a psychological variable: The power of being distinctive. Unpublished manuscript, Harvard University. Tetlock, P. E., & Kim. J. I. (1987). Accountability and judgment processes in a personality prediction task. Journal of Personality and Social Psychology, 52, 700-709. Trivers, R. L. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man (pp. 136-179). Chicago: Aldine. Turner, J. H., & Singleton, R. (1978). A theory of ethnic oppression: Toward a reintegration of cultural and structural concepts in ethnic relations theory. Social Forces, 56, 1001-1018. Vallacher, R., & Nowak, A. (1997). The emergence of dynamical social psychology. Psychological Inquiry, 8, 152-160. van den Berghe, P. L. (1967). Race and racism. New York: Wiley. van Dijk, T. (1987). Communicating racism: Ethnic prejudice in thought and talk. Newbury Park, CA: Sage. van Laar, C., Sidanius, J., Rabinowitz, J. L., & Sinclair, S. (1999). The three R’s of academic achievement: Reading, ‘Riting, and Racism. Personality and Social Psychology BuNetin, 25, 139-151. Walster, E., Berscheid, E., & Walster, G . W. (1973). New directions in equity research. Journal of Personality and Social Psychology, 25, 151-176. Wetherell, M., & Potter, J. (1993). Mapping the language of racism: Discourse and the legirimation of exploitation. New York: Columbia University Press. White, D. G . (1985). Jezebel and mammy: The mythology of female slavery. In Ar’n’t I a woman? (pp. 27-61). New York: Norton. Wilson, G. D. (1973). The psychology of conservatism. New York: Academic Press. Wilson, G . D., & Bagley, C. (1973). Religion, racialism, and conservatism. In G. D. Wilson (Ed.), The psychofogy of conservatism (pp. 117-128). London: Academic Press. Wilson, G . D., & Patterson, J. R. (1968). A new measure of conservatism. British Journal o j Social and Clinical Psychology, 7, 264-269.
SOCIAL DOMINANCE THEORY
263
Wilson. W. J. (1996). When work disappears: The world of the new urban poor. New York: Knopf. Wittenbrink, B., Judd, C. M., & Park. B. (1997). Evidence for racial prejudice at the implicit level and its relationship with questionnaire measures. Journal of Personality and Social Psychology, 72, 262-274. Word, C. 0.. Zanna, M. P., & Cooper. J. (1974). The nonverbal mediation of self-fulfilling prophecies in interracial interaction. Journal of Experimental Social Psychology, 10, 109- 120. Yinger, J. (1986). Measuring discrimination with fair housing audits: Caught in the act. American Economic Review, 76, 881-893. Yinger, J. (1991a). Housing discrimination study: Incidence and severity of unfavorable treatmenf. Washington, DC: U.S. Department of Housing and Urban Development. Office of Policy Development and Research. Yinger, J. (1991b). Housing discriminafion study: Incidence of dkcrimination and variation in discriminatory behavior. Washington, DC: U.S. Department of Housing and Urban Development. Office of Policy Development and Research. Yinger, J. (1995). Closed doors, opportunities lost: The continuing costs of housing discrimination. New York: Russell Sage Foundation. Yinger, J. (1996). Discrimination in mortgage lending: A literature review. In John Goering and Ron Wienk (eds.), Mortgage lending, ruciui discrimination, and Federal Policy (pp. 29-73). Washington, DC: The Urban Institute Press. Zajonc, R. B. (1965). Social facilitation. Science, 149, 269-274. Zegers de Beijl, R. (1990). Discrimination of migrant workers in Western Europe. Geneva, Switzerland: International Labour Office. Zubrinsky, C. L., & Bobo, L. (1996). Prismatic metropolis: Race and residential segregation in the City of the Angels. Social Science Research, 25, 335-374.
This Page Intentionally Left Blank
REPRESENTATION
Charles G. Lord Mark R. Lepper
I. Introduction
“Attitude,” as Gordon Allport (1935) saw it in his classic chapter in the first Handbook of Social Psychology, was the “most distinctive and most indispensable concept in contemporary social psychology” (p. 798). Although one might quarrel today with the extremity of his claim, there is no doubt that the attitude concept has remained ubiquitous in, and characteristic of, social psychology in the intervening 60 years. From the early days of Allport and his colleagues through the glory days of Hovland’s Yale Attitude Change Program (e.g., Hovland, Janis, & Kelley, 1953; Kiesler, Collins, & Miller, 1969; McGuire, 1969) to the more complex and sophisticated models of more recent years (e.g., Ajzen & Fishbein, 1980; Eagly & Chaiken, 1993; Fishbein & Ajzen, 1975; Olson & Zanna, 1993; Petty & Cacioppo, 1981, 1986; Zanna & Rempel, 1988), attitudes have indeed remained a central topic in social psychology. Crucial to the overwhelming success of the attitude concept were two key ingredients. The first was the implied stability of most important social attitudes, the second the presumed generality of the evaluative tendencies these attitudes involved. Thus, basic social attitudes were commonly seen as likely to persist, at least in the absence of concerted efforts to change them, over time. Attitudes were therefore distinguished by early theorists from other more ephemeral influences on a person, such as passing mood states or the momentary salience of particular stimuli in a person’s thoughts or surroundings. Indeed, measures of temporal stability were often explicitly sought in judging the adequacy of various potential measures of social and political attitudes. ADVANCES IN EXPERIMENTAL SOCIAL PSYCHOLOGY, VOL. 31
265
Copyright 0 1999 by Academic Press. All rights of reproduction in any form reserved. 0065-2601/99130.00
266
CHARLES G . LORD AND MARK R. LEPPER
Similarly, although people were clearly recognized to have attitudes about individual objects and persons, the prototypic social attitudes that gave the concept such breadth and utility were those concerning large and diverse groups of people or objects or policies-attitudes towards Blacks or Jews, Russians or communists, trade unions or integration. Not only were these attitudes presumed to indicate stable evaluative tendencies toward a single identical object at different times, they also implied cross-situationally consistent evaluative tendencies toward different members of a large class of specific attitude objects. In short, the essential power of the attitude concept lay in the possibility that a person’s reactions to various different members of an object class, observed at different times and in an array of different settings, might all be influenced-and predicted-by that person’s global attitude toward that class of objects. Indeed, some early authors sought, through concepts like ethnocentrism and authoritarianism, to suggest that people’s attitudes on these issues might prove even more general-that attitudes towards Blacks and Jews, communists and Japanese, might all share some common elements (e.g., Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950; Allport, 1954). At the same time that these features helped to make attitudes a central and indispensable concept for social psychologists, they also set the stage for two early, and enduring, challenges to the attitude concept-challenges that were crucial in shaping the thinking that has gone into most of the research reported in this chapter. Let us turn to these two challenges, in turn. A. THE ATTITUDE-BEHAVIOR PROBLEM At roughly the same time that Allport (1935), at Harvard, was writing that first, influential Handbook chapter in praise of the attitude concept, Richard LaPiere, a sociologist at Stanford, was preparing one of the first important attacks on the generality of the attitude concept (LaPiere, 1934). His paper, the first to point to what has since become well-known as the problem of “attitude-behavior inconsistency,” began simply enough, with a series of automobile trips that LaPiere had taken, across the United States and up and down the West coast. On those trips, LaPiere had driven over 10,000 miles, a substantial itinerary in those early days of the automobile. In his travels, he had stayed at 66 hotels and motels and had eaten at 184 restaurants and cafes. Only at one hotel had he been refused lodging. Such a record would, of course, have been entirely unremarkable had it not been the case that LaPiere had been accompanied on this trip by a young Chinese student and his wife, who were also offered food and lodging
ATTITUDE REPRESENTATION THEORY
267
at all but one of the 251 establishments they approached. What made this observation of particular theoretical interest was the seeming inconsistency between the warm welcome his Chinese traveling companions received and the large research literature on social distance and racial prejudice that showed the powerful and pervasive negative attitudes that most white Americans held in those days about “Orientals.” To document this inconsistency further, LaPiere subsequentiy wrote to all of the establishments he had frequented, asking explicitly whether those establishments would “accept members of the Chinese race as guests in your establishment.” Of the 47 hotels and 81 restaurants that replied to his letter, only one answered with a straightforward “yes,” whereas over 90% answered with an unequivocal “no.” Among several alternative hypotheses that LaPiere considered, in attempting to understand the striking apparent inconsistency of the proprietors of these businesses, one stood out to him-that the actual, well-dressed, highly educated, and clearly cultured Chinese couple who had accompanied LaPiere on his journeys may have borne little resemblance to the exemplars or stereotypes that may have come to the minds of the proprietors when they were asked an abstract question about purely hypothetical “members of the Chinese race.” Perhaps, he suggested, people’s responses to global attitude questions may fail to predict their overt actions toward specific members of the attitude category, when those specific representatives of the class fail to match the stereotypic representations called to mind by abstract and general attitude questions. LaPiere’s (1934) demonstration of the attitude-behavior problem was followed by many similar critiques, each advancing a different explanation or explanations for the phenomenon (e.g., Ajzen & Fishbein, 1977; Campbell, 1963; Deutscher, 1973; Festinger, 1964; Kutner, Wilkins & Yarrow, 1952; McGuire, 1976; Wicker, 1969). This abiding theoretical interest in the attitude-behavior problem attests to how strongly it violates the layperson’s assumptions about being able to predict many facets of an individual’s behavior toward a stimulus or a class of stimuli merely from knowing that one crucial piece of information-the individual’s global attitude. B. T H E ATTITUDE-OBJECT PROBLEM
If the appearance of inconsistency in people’s thoughts and actions toward different members of an attitude category offered one central challenge to the attitude concept, an equally important threat was posed by a second, quite different set of experiments reported by Solomon Asch (1940) just a few years after Lapiere’s study. Again, the basic phenomenon was
268
CHARLES G. LORD AND MARK R. LEPPER
simple and straightforward. In particular, Asch asked college students to rank-order ten different professions in terms of characteristics like intelligence, character, and idealism. For one specific profession, politics, he also led the students to believe that 500 of their peers had previously rated this particular category as either the first, or the last, on the list of ten professions. This normative information had a large impact on students’ rankings of politics. Those who had been led to believe that their peers had ranked politics first themselves ranked politicians an average of 4 to 5 ranks higher than those who had been led to believe that their peers had ranked politics last-a finding that seemed sharply at variance with the presumed stability of social attitudes. As Asch himself was quick to note, however, this effect was not simply the result of mindless conformity or of instant attitude change. Indeed, when he interviewed his participants, they strongly denied both of these accusations. They did, however, plead guilty, as it were, to a considerably more subtle and more interesting process. What Asch argued, on the basis of these interviews, was that the students in the two groups had come to interpret the term “politics” rather differently-to have had quite different politicians in mind when they answered Asch’s questions. Those who thought their peers had ranked politics at the top of the list, on the one hand, were more likely to have thought of great statesmen, such as Franklin D. Roosevelt, Churchill, Lincoln, or Washington when they considered the profession of politics. Those who thought their peers had ranked politics at the bottom of the list, on the other hand, were more likely to have thought of corrupt political bosses and party hacks, of Boss Tweed and Tammany Hall, when they considered the same profession of politics. It was as if, in Asch’s elegant and enduring phrase, there had been a “change in the object of judgment, rather than in the judgment of the object” (p. 458).
11. Attitude Representation Theory
It is these two challenges to the utility of the concept of general social attitudes, then, that have formed the primary stimuli leading to the program of collaborative research we report upon in this chapter. Indeed, it is in part from the speculations that both LaPiere (1934) and Asch (1940) offered in accounting for their respective findings that our own thinking about these problems has sprung. Both Asch and LaPiere offered, in slightly different ways, persuasive arguments for thinking about issues of attitude consistency and stability in terms of the construals and interpretations of the social perceiver.
ATTITUDE REPRESENTATION THEORY
269
A. THE REPRESENTATION AND MATCHING POSTULATES Our own model, referred to here as Attitude Representation Theory, formalizes and extends these theoretical ideas. The model seeks to help us to understand, in particular, when and where to expect consistency versus inconsistency in people’s responses to what is objectively the same class of stimuli-as a function of variations in the specific members of that class, their characteristics, the types of responses involved, and the contexts in which those responses are observed. It does so in terms of two basic postulates. 1. The Representation Postulate
Our first postulate, then, makes a simple point-that a person’s response to any attitude-relevant stimulus will depend not only on the perceived properties of that stimulus and the situation surrounding it, but also on the subjective representation of that stimulus by the person. This point, of course, has been made by other authors in a number of specific contexts (e.g., Schwarz, 1994; Tesser, 1978; Wilson & Hodges, 1992). We seek, however, to present a more general version of this argument that we call the “representation postulate.” Like LaPiere and Asch, we will argue that generating more concrete representations of abstract categories is a widespread response when people are asked to make global verbal evaluations of various attitude classes. In addition, however, we will also argue that the manner in which people construct or construe attitude stimuli (Griffin & Ross, 1991; Ross, 1987; Ross & Ward, 1995) may be important not only when they are asked to answer abstract and general questions about broad categories of people or objects, but even when they are asked to respond to specific situations that involve particular instances or exemplars of these categories. 2. The Matching Postulate Our second postulate makes an equally simple point-that the closer the match between the subjective representations and perceived immediate stimuli to which a person is responding in one situation and the subjective representations and perceived immediate stimuli to which the person is responding in a different situation, the more consistency there will be in that person’s responses. Again, this is a point that has been made by other authors, including Asch and LaPiere, in various specific contexts (e.g., Ajzen & Fishbein, 1980; Millar & Tesser, 1992), but once again we will propose a more general version of this claim that we call the “matching
270
CHARLES G. LORD AND MARK R. LEPPER
postulate.” Thus, consistency across various opportunities for attituderelevant responses should depend on the degree of match across an array of dimensions-not just the identity of the specific instances that come to mind, but also their assumed characteristics, the types of responses they are associated with, and the contexts in which they occur. The more two situations activate assumptions and perceptions that match across all of these relevant dimensions, the more response consistency or generality our model will predict. In the sections to follow, we will begin by describing a process model of evaluative responses that incorporates both the representation postulate and the matching postulate. We will then review research that addresses each of the postulates as it pertains to this process model. Finally, we will examine issues raised by thinking about the relationship between Attitude Representation Theory and other theoretical and empirical contributions.
B. PROCESSES IN ATTITUDE REPRESENTATION THEORY Figure 1 presents a process model of how people make evaluative responses to different members of an object class, or even to the exact same attitude object at two separate times, or in two separate situations. According to the Attitude Representation Theory (ART) model, the same processes should occur at both times, but these processes can yield very different outcomes, depending upon the individual’s subjective representation of the stimulus, as well as his or her perceptions of the immediate stimulus. These assumptions and perceptions, each weighted by its importance, are then combined to produce an overall evaluation that can range from very negative to very positive. Unless other factors extraneous to the model intervene, the individual will respond on the basis of that overall evaluation. The two responses will be more similar when the overall evaluations, based on their assumptions and perceptions, match than when they mismatch. As shown in Figure 1, the first step after identifying the situation as relevant to an attitude object is often to activate both a subjective representation of the stimulus and, if appropriate, perceptions of its immediate properties. At Time 1 in LaPiere’s (1934) study, when the couple requested service, the attitude object was objectively the same as it would be later on the questionnaire-namely, “accepting members of the Chinese race as guests.” At Time 1, the proprietors might have noticed that the couple were Chinese and activated a subjective representation of “Chinese people” that carried with it various assumptions or “default values” about members of this category. In the 1930s, for instance, a typical proprietor might have
271
ATTITUDE REPRESENTATION THEORY
attitude-relevant situation at Time 2
attitude-relevant situation at Time 1
ACTIVATE assumptions from subjective representation
perceptions from immediate
assumptions from subjective representation
perceptions from immediate stimulus
Tla TCa ARa2 CFa2
TIP 2 TCP 2 ARp2 CFD2
ARa, CFa
COMBINE
COMBINE
assumptions with perceptions to reach an overall evaluation
assumptions with perceptions to reach an overall evaluation
Matchl
2 wTla, + 2 wTlp, ZwTCa, + ZwARa, + C wCFa, +
2 wTla, +
+---b
z wTCp,
XwTlp,
2 wTCa, + 2 wTCP,
Mismatch
2 WARP,
z wARa,
Z
2 wCFa, + 2 wCFp,
1 +
wCFP,
RESPOND informed by the overall evaluation
ZwARp,
-~
Matchl
+---Mismatch
+
informed by the overall evaluation
Fig. 1. kttitude Representation Theory (ART) model of the process that underlies attitude-relevant evaluative responses. TIa and TIp = assumptions about and perceptions of a target’s identity or identities. TCa and TCp = assumptions about and perceptions of a target’s characteristics. ARa and ARp = assumptions about and perceptions of the types of responses associated with a target. CFa and CFp = assumptions about and perceptions of contextual factors that surround responses to the target. w = weight given to each of these four aspects when they are combined to form an overall evaluation.
272
CHARLES G. LORD AND MARK R. LEPPER
assumed that most Chinese people were coolie laborers (target identities), were ignorant and poor (target characteristics), were the “type of people I usually avoid and do not respect” (associated responses),’ and would be easy to eject because other patrons would applaud (contextual factors).2 These assumptions are represented in Figure 1 as TIal (assumed target identities at Time l), TCal (assumed target characteristics at Time l),ARal (other responses associated with the attitude object at Time l), and CFal (assumed contextual factors at Time 1). With an actual Chinese couple standing before them, however, the proprietors also had available to them a number of more direct perceptions. According to Lapiere’s (1934) account, the Chinese couple appeared to be academics or professionals (target identities). They were obviously well educated and well-to-do (target characteristics). They were in many respects the type of people the proprietors usually favored as paying customers (associated responses). In addition, they were accompanied by a Caucasian professor, and it may have appeared difficult to get rid of them without upsetting other patrons (contextual factors). These immediate perceptions are represented in Figure 1 as TIP,, TCpl, ARpl, and CFpI. To reach an overall evaluation of the attitude object (accepting these specific members of the Chinese race as guests), as Figure 1 shows, the proprietors combined the activated assumptions and perceptions, each of which carried both a valence and an importance weight.’ The overall evaluation, in turn, informed the proprietors’ responses. Of course, when we speak of the contrast between “subjective representations” and “assumptions” or “default values,” on the one hand, and “direct” or “immediate” perceptions, on the other, we must recognize that this distinction is actually a matter of degree, rather than a matter of kind (Bruner, 1957; McArthur & Baron, 1983). Certainly in the arena of social
‘
Assumptions about ‘‘associated responses” include cognitive, affective, and behavioral responses to the attitude object (Eagly & Chaiken, 1998) that the individual has had in the past (e.g., Bern, 1972) or imagines that he or she might have in the future (e.g., Ross, 1989). Perceptions of “associated responses” include cognitive, affective, and behavioral responses in the immediate situation. One type of affective response in the immediate situation can be an automatic, spontaneous affective reaction to the attitude object (Fazio el a]., 1986; Bargh et al., 1992). * “Contextual factors” are aspects of the social context that the individual assumes or perceives might facilitate or inhibit evaluating the attitude object with favor or disfavor (see Eagly & Chaiken’s, 1993, p. 1, definition of “attitude”). Attitude representation theory thus incorporates several distinct elements of the theory of planned behavior (Ajzen, 1985)-subjective norms, perceived behavioral control (contextual factors), and behavioral intentions (associated responses)-within the assumptions and perceptions that are activated and combined to reach an overall evaluation, which in turn informs an attitude-relevant response. (For details of possible combination rules and processes, see Eagly Chaiken. 1993, pp. 106-114.)
ATTITUDE REPRESENTATION THEORY
273
attitudes, there are no “immaculate perceptions” uncolored by our beliefs and expectations (Higgins, 1996; Neisser, 1967). Instead, we know that even our immediate and direct perceptions of specific social actors standing next to us may also be influenced by our stereotypes, our prior knowledge, and the context in which we encounter them (Griffin & Ross, 1991; Rock, 1984; Ross, 1987). Thus Lapiere’s proprietors did not actually “see” a cultured, well-educated, affluent couple asking for service; those presumed attributes had to be inferred from the dress, accents, appearance, and remarks of LaPiere’s companions. It is only in order to simplify our presentation, then, that we will describe the perceiver’s recalled representation of the attitude object as “assumptions,” but the perceiver’s interpretations of the immediate situation as “perceptions.” In these terms, then, at Time 1 in Lapiere’s (1934) study (shown on the left side of Figure l ) , the proprietors might have relied almost entirely on their immediate perceptions of the specific Chinese couple facing them, which clearly contradicted most of the assumptions from their subjective representations. The target’s identity was presumably different from what the proprietors might have assumed, as were the target’s characteristics and the context. Only the type of response (accepting them as guests) was similar to the attitude-relevant situation that would later be identified in the questionnaire at Time 2 (shown on the right side of Figure 1). As a result, they chose to serve the Chinese couple. By contrast, when the proprietors answered Lapiere’s attitude questionnaire at Time 2, they were forced to rely entirely on assumptions from their subjective representations of Chinese people, because there was no spec@ target stimulus to be directly perceived. Whatever assumptions were activated, whether about the target’s identity, characteristics, associated responses, or contextual factors, had to carry all the weight in determining the overall evaluation and evaluative response. LaPiere himself characterized the proprietors’ answers to his questionnaire as a “verbal response to a symbolic situation” (1934, p. 230). Given the negative stereotypes of the day, it is not surprising that when given three choices, the proprietors circled “no” rather than either “yes” or “uncertain; depend upon circumstances.” After all, how many reputable establishments would agree to serve a poor, ignorant, coolie laborer who could be easily ejected to loud applause? As the terms “Matchhlismatch” at the bottom center of Figure 1 indicate, evaluative responses to seemingly related, or even objectively identical stimuli will sometimes match; but sometimes, as in LaPiere’s (1934) study, they will not. When they do not match, barring extraneous factors such as coercion, the lack of a match may frequently be due to differences between the subjective default assumptions and more immediate and direct perceptions that were activated on the two occasions (and/or to differences in the
274
CHARLES G. LORD AND MARK R. LEPPER
weights that those assumptions and perceptions carried in determining the separate evaluations). In sum, laypeople in LaPiere’s (1934) and Asch’s (1940) era, as well as in the present, expect attitudes to be stable (expressed consistently across time) and general (expressed consistently across situations). The “attitudeobject problem” arose because empirical results suggested that attituderelevant self-reports are often unstable, leading Asch (1940) to suggest that even when an attitude object is objectively the same at two separate times, people might have in mind subjectively different instances or exemplars. Our more general “representation postulate” holds that evaluative responses of all types, not just answers to general attitude questions, can involve both perceptions of the immediate situation and representationbased assumptions. Similarly, the “attitude-behavior problem” arose because empirical results suggested that attitude-relevant responses often do not generalize from one situation to the next. LaPiere (1934) suggested that the consistency between verbal reports and overt actions may depend on a match between the assumed and the actual attitude object’s identity. Our more general “matching postulate” suggests that attitude generality depends not just on the matching of assumed and perceived identities, but also on the matching of representation-based assumptions and more immediate perceptions on a number of relevant dimensions. These two basic premises-the “representation postulate” and the “matching postulate”-were then incorporated into what we call Attitude Representation Theory-a general model of the process by which people make evaluative responses. The previous section provided a broad overview of this model. In the two sections to follow, we describe the specifics of this model, by concentrating first on research relevant to the representation postulate and then on research relevant to the matching postulate.
111. Research Relevant to the Representation Postulate
The starting point for ART, then, is the postulate that attitude-relevant responses typically involve the activation of a subjective representation of the stimulus. An attitude-relevant response, whether to a direct question or to a behavioral situation that concerns “members of the Chinese race,” for instance, might activate a default representation of a Chinese person or persons. Although ART allows for many different types of representational formats, such as associative networks, schemas, exemplars, and parallel distributed processing systems (Smith, 1998), much of the supportive evi-
ATTITUDE REPRESENTATION THEORY
275
dence comes from investigations of the activation of exemplars. Thus, in one set of studies, we tested whether exemplars are selectively activated when people evaluate a general stimulus given only its name. In a second set of studies, we tested whether people who activate the same exemplar to represent a stimulus on separate occasions also report more stable or consistent attitudes. Third, we tested whether consistent exemplar activation lends resistance to attitude change. Fourth, we tested whether attitudes can be altered by making different exemplars temporarily accessible. Fifth, we tested the role of abstract and specific exemplar representations in Allport’s (1954) social contact hypothesis. Finally, we tested the importance of exemplar representativeness in determining the generalization of attitude change that derives from equal status contact. Together, these studies provided convergent evidence for, and explored the implications of, the representation postulate.
A. SELECTIVE ACTIVATION OF REPRESENTATIONS After identifying a situation as relevant to an attitude object, as shown in Figure 1, some representation of the attitude object is activated. When responding to a questionnaire about attitudes toward “politicians,” for example, Asch’s (1940) participants might have activated a representation of the relevant category. Such cognitive representations will often entail category exemplars, such as Franklin Roosevelt or Winston Churchill (e.g., Smith, 1998). Indeed, Asch (1940) asked participants which politicians they had in mind when they made their attitude judgments and many of them named Roosevelt or other famous “statemen.” Unfortunately, we cannot necessarily assume from their self-reports that these participants were able to report accurately on their cognitive processes (Nisbett & Wilson, 1977). Considerable research has shown, however, that exemplars are frequently used for other, nonattitudinal social judgments. Judgments about social categories, for instance, often depend on the specific exemplars that an individual activates on that occasion (Smith, 1992; Smith & Zarate, 1992). In one relevant study, sportswriters and coaches rated very highly a college football player who happened to have the same hometown as a well-known professional star (Gilovich, 1981). Novel members of a social category may also be evaluated in terms of their similarity to the “default values” assumed for a typical category member (Rothbart & John, 1985). Black men are categorized primarily as “Black” and White women are categorized primarily as “women” in a culture that takes “White men” as the default value (Zarate & Smith, 1990).
276
CHARLES G. LORD AND MARK R. LEPPER
Moreover, exemplars can be activated and thus have an impact on judgment without being accessible to conscious recall or recognition (Smith, 1998). Participants in several studies have been better able to complete a fragmented word such as -SS -SS-- even though they could not recall or recognize having earlier read the word (assassin) on a list (e.g., Roediger, 1990; Tulving, Schacter, & Stark, 1982). Similarly, participants in another study avoided interacting with an experimental assistant who happened to have the same hairstyle as an experimenter who had previously insulted them, even though they staunchly denied being influenced by that earlier encounter (Lewicki, 1986). More generally, studies of repetition priming, word-fragment completion, and savings in relearning have all demonstrated types of implicit learning that seem qualitatively different from performance on semantic tasks such as defining or comprehending concepts (Carlston & Skowronski, 1994; Smith, Stewart, & Buttram, 1992;Tulving et al., 1982;for a recent review see Smith, 1998). Moreover, studies of halo effects, mere exposure, subliminal attitude conditioning, and context effects in survey research have likewise shown implicit social cognition processes that do apply more specifically to attitudes (Edwards, 1990; Greenwald, 1990 Schwarz & Clore, 1983; Thorndike, 1920; Zajonc, 1980; for a recent review see Greenwald & Banaji, 1995). If attitude judgments about social categories are similar to other types of judgments about social objects, then, attitude assessments might rely heavily on exemplars that are activated from a representation of the attitude object or stimulus, just as LaPiere (1934) and Asch (1940) suggested. Furthermore, people might use exemplars in assessing their attitudes even if they cannot accurately report having done so (Tesser & Martin, 1996). One way of gaining evidence about the process of attitude assessment that does not rely completely on the validity of self-reports is derived from a wellestablished principle of mental operations-that people selectivelyactivate, from what they know about a topic, those pieces of information that are most useful for the specific judgment at hand. Thus selective activation implies that a concept is used in a judgment process (Anderson, 1983; Anderson & Pichert, 1978; Higgins & King, 1981; Newell, Rosenbloom, & Laird, 1989; Wyer & Srull, 1986). Because it is often impossible (or inefficient) when making a judgment “to engage in an exhaustive search of all relevant information stored in memory, the outcome of a search will be influenced by the relative accessibility of the relevant information” (Strack, 1992, p. 256). More specific to attitudes, several models hold that “people often have a large, conflicting ‘data base’ relevant to their attitudes on any given topic, and the attitude they have at any given time depends on the subset of these data to which
ATTITUDE REPRESENTATION THEORY
277
they attend” (Wilson & Hodges, 1992, p. 38; see also Tesser, 1978). That is one reason why attitude measurement is so susceptible to subtle context and framing effects (Ottati, 1997; Schwarz, 1994; Tourangeau & Rasinski, 1988). Sia, Lord, Blessum, Thomas, and Lepper (in press) used three “process” measures to investigate Asch’s (1940) claim and the A R T prediction that people selectively activate relevant exemplars when asked to indicate their general social attitudes. One such process measure was based on the welldocumented principle that the activation of cognitive elements may be detected by asking participants to interpret ambiguous stimuli. In one classic study of construct accessibility, for example, students who had recently memorized positive trait words such as “adventurous” interpreted a fictional character’s exploits more positively than did students who had recently memorized negative trait words such as “reckless” (Higgins, Rholes, & Jones, 1977). Since then, dozens of studies have shown that recently activated constructs can color the interpretation of subsequently presented ambiguous information (Stangor, 1990; Sedikides, 1990; Srull & Wyer, 1979). Similarly, recently primed or activated constructs are more likely to be “found” in the jumbled letters of anagrams (Roediger, 1990), even when explicit measures such as recognition or free recall remain unaffected (Rajaram & Roediger, 1993). Assessing one’s attitude toward talk show hosts might thus increase the likelihood of solving the anagram HAPOR as OPRAH rather than HARPO. In their first experiment, therefore, Sia and colleagues (in press) had college students complete one of two booklets. Students who completed an attitudes booklet wrote out their attitudes toward several social categories, including politicians, basketball players, talk show hosts, and televangelists. Students who completed a definitions booklet wrote brief definitions, instead of attitudes, for the same social categories. Immediately after answering each question, participants in both groups were asked to solve an “anagram,” such as rearranging letters from CILRNOPPROTUN to form a seven-letter word or name. The only difference between the conditions, therefore, was that some participants wrote their attitude toward a category before solving each anagram, whereas other participants wrote a definition of the category before solving the same anagrams. Readers may have noticed that seven of the letters in CILRNOPPROTUN-which occurred directly after a question asking for the students’ attitudes toward politicians in the attitudes condition and directly after a question asking for a definition of politicians in the definitions conditioncould be rearranged to form either “Clinton” or “corrupt.” In fact, in a pretest in which participants merely free-associated to the word “politicians,” 37% had mentioned “Clinton” and 37% had mentioned “corrupt.”
278
CHARLES C . LORD AND MARK R. LEPPER
We know, therefore, that the nonexemplar associate “corrupt” and the exemplar “Clinton” were equally likely to be activated by the mere name of the category, when no judgment was required. Because using the concept to solve an ambiguous anagram is a measure of activation during a judgment, the A R T prediction was that students would find “Clinton” more often than “corrupt” after an attitude than a definition question about politicians. Similarly, they should find “Shaq” more often than “Shoe,” “Oprah” more often than “smile,” and “Baker” more often than “money” after attitude questions than after definition questions about basketball players, talk show hosts, and televangelists, respectively, even though in each case the alternatives had been mentioned by an equal percentage of pretest participants who were asked to free-associate to the category namee4 Figure 2 shows the significant Preceding Question X Type of Solution interaction for the number of anagram solutions (of a possible four) that students produced. After semantic, definition questions, students found an equivalent number of nonexemplar and exemplar solutions. After attitude questions about the same categories, they found significantly more exemplar than nonexemplar solutions. The result, derived from process measures rather than self-reports, clearly supported Asch’s (1940) contention and the A R T hypothesis that attitude assessment selectively activates exemplars from a representation of the stimulus, presumably because concrete exemplars are very useful for making an evaluative response when given only the name of the attitude object. One’s opinion of Newt Gingrich, for instance, seems more relevant to assessing one’s attitude toward “politicians” than does one’s opinion of “people who wear suits,” even though they might be equally associated with the category in a strictly semantic sense. Sia, and her associates’ (in press) second experiment yielded similar results, using another well-researched indicator of concept activation, namely word-fragment completions (see Tulving et al., 1982; Tulving & Schacter, 1990), instead of anagram solutions. A preceding question about attitudes toward, rather than definitions of, basketball players, for instance, increased the likelihood of completing --RD-N as “Jordan” rather than as “garden.” Likewise, in a third experiment Sia and associates (in press) borrowed a response latency technique that had previously been used to disconfirm a “common-sense’’ hypothesis about the concepts that are activated when people assess their own personality traits (Klein, Loftus, & Burton, 1989). The technique relies on the principle that cognitive processes occur faster if one of the pieces of information used in a judgment has recently been activated (e.g., Bargh, 1992; Collins & Quillian, 1969; Devine, 1989; Dovidio, Evans, & Tyler, 1986; Klein & Loftus, 1993; Klein et al., “Bakker” was misspelled by all participants.
REPRESENTATION THEORY
1.6
-
279
Nonexemplar solutions ~
1.2 1.4
.-:
-5:a
1 -
CI
0. 8
-
0.6
-
0.4
-
0.2
-
r
0
L
2
s
2
0 7
Semantic
Attitude
Question preceding the puzzle
Fig. 2. Mean number of exemplar and nonexemplar anagram solutions that participants used after answering either a semantic or an attitude question about each of four social categories. (Data from Sia, Lord, Blessurn, Thomas, & Lepper, in press, Experiment 1.)
1989; Lord, Desforges, Chacon, Pere, & Clubb, 1992; Macht & O’Brien, 1980; Macht & Spear, 1977; Niedenthal, 1992: Perdue & Gurtman, 1990; Smith et al., 1992). Indeed, for many purposes, response latency measures have proven more reliable than a perceiver’s (often erroneous) impressions of his or her own cognitive processes (Nisbett & Wilson, 1977). College undergraduates (Sia et al., in press, Experiment 3), then, were asked to answer yes or no to questions about various social categories on a computer screen. For eight of t h e categories, the students answered (at least 10 questions apart) both an attitude question and a definition question. Immediately following each of these two types of judgments, they answered a category-inclusion question such as “Is Bill Clinton a politician?” The hypothesis was that, if attitude assessment selectively activates exemplars, participants would be faster to answer the category-inclusion question after an attitude question than after a comparable definition question. Figure 3 shows mean response latencies for the category-inclusion question following either an attitude question or a definition question for a representative four of the eight categories. (The other four categories showed the same pattern.)
280
CHARLES G. LORD AND MARK R. LEPPER Semantic question Attitude question
2500
Q
2000
U C
0 0
.--t
‘i 1500
-6 .-c C
-#
1000
Q u)
B3
K
500
0 Politicians
Talk show hosts
Televangelists People with AIDS
Social category
Fig. 3. Mean response latency (in milliseconds) for a category-inclusion question that followed either a semantic or an attitude question about each of four social categories. (Data from Sia, Lord, Blessurn, Thomas, & Lepper, in press, Experiment 3.)
As predicted, students were significantly faster to judge whether an exemplar belonged to the category after an attitude question than after a definition question, as would be the case if the attitude question had indeed selectively activated exemplars. This result held when differences in response latency for the first question (attitude or definition) were statistically controlled. It held also whether answers to the category-inclusion question were “yes” or “no,” presumably because even a “no” category-inclusion answer (e.g., “Is Rochester the capital of New York?”) is facilitated by having correct exemplars activated in advance. The evidence regarding concept activation during attitude-relevant judgments thus supported the “activation” step in the ART model as depicted in Figure 1. Admittedly, actual exemplars might not be activated for social categories that have no named exemplars, such as “homeless people” or “student dissidents,” but both exemplar theory (Smith, 1998)and associated
AITITUDE REPRESENTATION THEORY
281
systems theory (Carlston, 1994) argue that exemplars are often imaginary, nameless mental images such as a bag lady we saw on the street or a young man standing in front of a column of tanks. Similarly, exemplars might also not be selectively activated for attitudes toward other types of attitude objects, such as specific persons (e.g., Clinton), social policies (e.g., capital punishment), abstract concepts (e.g., religion), or activities (e.g., going to the movies). Finally, some attitudes might be “known directly” from a spontaneous affective response rather than inferred from any other activated elements (Bargh, 1996; Bargh, Chaiken, Govender, & Pratto, 1992; Fazio, Sanbonmatsu, Powell, & Kardes, 1986), though response latencies of approximately 2000 ms, as shown in Figure 3, suggest that spontaneous affective reactions, which occur in less than 500 ms (Logan, 1989; Neely, 1977), might well be integrated with other pieces of information, such as default values from activated assumptions about likely exemplars, before people arrive at a final attitude-relevant response.
B. CONSISTENT ACTIVATION AND RESPONSE STABILITY One central concern of the ART model presented in Figure 1, is, of course, the relationship between attitude-relevant responses that occur at two separate times. In such cases, people’s responses sometimes remain consistent, but sometimes they do not. A R T addresses at least one reason for attitude stability versus instability. If no useful information has been added in the interval between two attitude-relevant responses, one might assume that the two responses should be identical. Suppose, for instance, that an individual answers the question “What is your attitude toward politicians?” at two separate times, one month apart. Because there is no immediate stimulus or specific attitude object present on either occasion, the individual is presumed to respond both times on the basis of attitude representation assumptions shown in Figure 1 as TIa, TCa, ARa, and CFa, which suggests that “Match/Mismatch” in the figure should automatically be “Match.” A R T predicts, in contrast, that the match is unlikely to be perfect, because, even with no intervening information and no change in the content of the underlying attitude representation from Time 1 to Time 2, it is possible that the assumptions activated at Time 1 might not be the same as the assumptions activated at Time 2. To understand this prediction, recall that representation-derived assumptions depend in part on which particular aspects of a representation are activated (Anderson, 1983; Strack, 1992; Tesser, 1978; Wilson & Hodges, 1992). An individual’s representation for politicians, for instance, might contain many exemplars, not just Bill Clinton. At any given time, 37% of
282
CHARLES G. LORD AND MARK R. LEPPER
respondents might activate Clinton as their first exemplar (Sia et al., in press), but not all of those respondents would necessarily activate Clinton first if they were to assess their attitudes again at a later date. According to data reported by Bellezza (1984), when people are asked twice, one month apart, to name an exemplar of a natural object category such as vegetables, they have only a .69 probability, not a 1.0 probability, of naming the same exemplar (e.g., carrots or corn) on both occasions. Sia, Lord, Blessum, Ratcliff, and Lepper (1997) reasoned that similar probabilities might pertain to social categories such as politicians or talk show hosts. An important tenet of ART, suggested by Figure 1, is that the stability of evaluative responses, even using the exact same response format, will depend on the match between the attitude representations and assumptions that are activated at the two separate times when people assess their attitudes. With all else held constant, people who activate the same exemplar when assessing their attitudes at two separate times have a higher probability of making the same evaluative response than do people who activate different exemplars. To test this prediction, Sia and her colleagues (1997, Experiment 1) asked students, for each of seven social categories, to assess their attitudes toward the categories, to name the first exemplar that came to mind to represent the categories, and then to name other exemplars that they thought might occur to them first on other occasions. Approximately one month later, the students completed exactly the same questionnaire. At each of the two sessions, a randomly selected subset of students reported their attitudes before their exemplars, and the others reported their exemplars before their attitudes. Figure 4 shows that students who named the same first exemplar at Time 2 as they had at Time 1 (for whom the assumed target identity at Time 1 matched the assumed target identity at Time 2) had greater attitude stability than did students who named different first exemplars (for whom the assumed target identity at Time 1 mismatched the assumed target identity at Time 2). The figure shows attitude-attitude correlation coefficients for each of four social categories. (The other three categories showed the same pattern.) Across the seven categories, students who named the same exemplar displayed attitude stability of .74, whereas students who named different exemplars showed attitude stability of .53.Also, when each student’s seven attitudes were divided into those for which the student had named the same exemplar and those for which the student had named a different exemplar, attitude stability was greater for the former categories than for the latter, regardless of the order in which students had named attitudes and exemplars. Furthermore, naming the same rather than a differ-
ATTITUDE REPRESENTATION THEORY
0.9] 1
283
0 Named different exemplar
Named same exemplar
0.8
Politicians
Talk show hosts
Televangelists
Homosexuals
Social category
Fig. 4. Attitude stability coefficients for participants who named the same versus a different exemplar, one month apart, for each of four social categories. (Data from Sia. Lord, Blessum, Ratcliff, & Lepper, 1997, Experiment 1.)
ent exemplar was category-specific, not an individual difference that applied across categories. Other analyses suggested that participants were likely to change their attitudes only when they named a different exemplar that they reported liking more, or less, than the one they had named first during the initial attitude assessment, not when they merely switched to naming a different exemplar that they liked or disliked equally with the first one. Furthermore, the students who named a different first exemplar tended to switch to one they had said might occur to them on a different occasion or to an exemplar that they had not named but presumably knew (e.g., Newt Gingrich), as suggested by the fact that many other participants had named that person. These results supported A R T by showing that response stability decreases when the elements activated from a representation of the attitude object are different on separate occasions. For ART, moreover, it is reassur-
284
CHARLES G. LORD AND MARK R. LEPPER
ing that the order of naming attitudes versus exemplars did not affect the relationship between naming the same exemplar and retaining one’s initial attitude. The results, of course, were only correlational, and did not, by themselves, rule out the possibility that students assessed their attitudes first-perhaps relying only on a spontaneous affective response to the category’s name-and then named an exemplar that they thought would justify such an attitude. This “attitude first” claim, however, is called into doubt by the results of the activation study described earlier (Sia, et al., in press), in which students were merely asked to assess and report their attitudes, and yet they still selectively activated exemplars. Automatic activation of an affective response takes less than 500 ms (Logan, 1989; Neely, 1977), after which students could have pressed a key to go on to the next question. Would they have waited 1500 additional milliseconds before going on (see response times in Figure 3) if they were only activating exemplars after-the-fact to justify their attitude rather than activating exemplars because they were relevant to assessing an attitude? Also, as we shall see, empirical evidence supports a causal relationship such that manipulating exemplar accessibility affects the stability of evaluative responses to what is objectively the same stimulus. C. CONSISTENT ACTIVATION AND RESISTANCE TO ATTITUDE CHANGE If consistent exemplar activation promotes the stability of evaluative responses, then it might also produce resistance to attitude change. Many previous studies have addressed the factors that impart resistance to attitude change, for instance, in the face of disconfirming evidence (e.g., Lord, Ross, & Lepper, 1979; Lydon & Zanna, 1990; McGuire, 1985; Petty & Cacioppo, 1986; Watts & Holt, 1979). The gist of the empirical results is that attitudes resist change better when they consist of strong, durable, or interconnected beliefs about the attitude object. People who always activate the same exemplar when they assess a particular attitude are likely to have, according to ART, a consistent and interrelated set of assumptions regarding the attitude object. People routinely attribute cross-temporally and cross-situationally consistent characteristics to others (Ross, 1977). Someone who always activates the same exemplar for a social category (the same assumed target identity), therefore, is very likely to activate the same set of assumed characteristics to represent the category. Also, someone who activates the same exemplar for a social category might have a history of previous responses associated with the exemplar (the same associated responses). An individual who always acti-
ATTITUDE REPRESENTATION THEORY
285
vates Bill Clinton to represent “politicians,” for instance, is very likely to activate the same set of Clinton’s characteristics every time, and the same associated responses, such as voting for or against him. If the exemplar remains unchanged from one attitude-relevant response to the next, interventions that do not change the activated exemplar might not have much effect on the underlying attitude. A person who activates Clinton at one time, Dole the next, and Gingrich the time after that, in contrast, is likely to activate different sets of individuating characteristics and a different set of previous evaluative responses (perhaps knowing that Gingrich used to be a professor, or being bored by one of Dole’s televised speeches) on each occasion. If the exemplar changes from one attitude-relevant response to the next, even interventions that do not address the activated exemplar might have a good chance of altering the underlying attitude, because that attitude has a shaky foundation. A persuasive message that said nothing about any exemplars, for instance, might introduce a new set of principles or associations that-without an “anchoring” exemplar to counteract them-might more easily undermine an existing attitude. From just such reasoning about the implications of ART, Sia, Thomas, Lord, and Lepper (1998) devised the first longitudinal study of exemplar activation. At Time 1, they asked college students to assess their attitudes toward several social categories, one of which was politicians, and to name the first exemplar that came to mind to represent each category. Once a week for the next five weeks (Times 2-5), the students completed the same questionnaire. Finally, at Time 6, all students participated in an “unrelated” study, conducted by a different experimenter from the one who had been giving them attitude questionnaires for the previous five weeks, in which they were asked to “help us prepare some stimulus materials.” To avoid possible demand characteristics had participants thought the persuasive messages were aimed at them, the experimenter asked the students to read several persuasive essays and, for each message, to summarize the topic, the author’s position, and the main arguments, as well as to rate the essay’s clarity and objectivity. Unknown to the students, the essay about politicians had been prepared to argue a position opposite to their own initial attitude at Time 1. Each of the two essays (one in favor of politicians and the other against) was also prepared to take a strong position without ever mentioning a specific politician exemplar. Instead, each essay concentrated on logical, principled arguments. The “pro” essay, for instance, argued that politicians served a valuable purpose in our society and that elected office was a more difficult task than in previous years. The “con” essay argued that the campaign
286
CHARLES G. LORD AND MARK R. LEPPER
finance laws allowed large contributors to exert undue influence on the political process and that reform of the political system is badly needed. After they had summarized and critiqued these persuasive essays, the students were thanked for “helping us prepare these stimulus materials,” and dismissed. Immediately thereafter, the experimenter who had been giving them attitude questionnaires for five consecutive weeks entered the room and distributed the attitude questionnaires for the sixth time (Time 6). Figure 5 shows how much student’s attitudes toward politicians changed from Time 5 (before the persuasive message) to Time 6 (after the persuasive message). As the figure shows, students who retained the same politician exemplar every week for five consecutive weeks changed their attitudes only slightly, and nonsignificantly, in the direction of the persuasive message (i.e., adopted slightly less extreme attitudes at Time 6 than at Time 5). Students who had changed exemplars at least once during the initial five weeks. in
0 Pre-message
Post-message
Retained Changed Exemplar generated prior to message
Fig. 5. Mean attitude extremity before and after reading a persuasive message, by participants who named the same versus different politician exemplars across five weekly attitude assessments. (Data from Sia, Thomas, Lord, & Lepper, 1998.)
ATTITUDE REPRESENTATION THEORY
287
contrast, changed their attitudes significantly in the direction of the persuasive m e ~ s a g eThis . ~ study thus established that, even beyond the previously demonstrated impact of exemplar stability on attitude stability, exemplar stability lends resistance to persuasive messages that are intended to change attitudes.
D. MAKING ASPECTS OF A REPRESENTATION ACCESSIBLE As described, people who name the same first exemplar of a social category twice, one month apart, are more likely to report an unchanged attitude than are people who name different exemplars (Sia et al., 1997). One interesting implication of these findings is that the latter group might have changed their attitudes, even without the introduction of new information about the attitude object, merely because they activated different aspects of the attitude representation at Time 2 than at Time 1.This finding thus suggests that temporarily accessible constructs can alter attituderelevant responses from one time to the next, even in the absence of new information about the stimulus, as Eagly and Chaiken (1998, p. 217) have suggested. Reminding people of a very likable social category exemplar, for instance, might cause them to favor the category and its interests more than they had in the past. In one particularly relevant study, students first estimated the heights of several celebrities (Bodenhausen, Schwarz, Bless, & Wanke, 1995). For some of the students, the last name on the list was a well-liked successful African-American (either Michael Jordan or Oprah Winfrey). For others it was just another White person, like all the others whose heights they had estimated. Immediately after these ratings, in an “unrelated” experiment, the same students indicated how much they believed that “racial discrimination against Blacks is no longer a significant problem in the U.S.” (Bodenhausen et al., 1995, p. 53). As the authors predicted, students who had been reminded of Michael or Oprah saw present-day racial discrimination as more of a problem than did students who had not been so reminded, presumably because the height-estimation task made a very likable exemplar temporarily more accessible than it would otherwise have been. The accessible exemplar presumably ameliorated attitudes toward AfricanAmericans, which in turn altered beliefs about the problem of discrimination.
’
Although students who retained the same politician exemplar had more stable attitudes across the first five weeks than did students who changed exemplars, the relationship between exemplar stability and attitude change was independent of attitude variance during the first five weeks.
288
CHARLES G. LORD AND MARK R. LEPPER
In a related study that measured attitude change more directly, Sia and her colleagues (1997, Experiment 3) had students report their attitudes toward, and exemplars for, several social categories, one of which was politicians. From a separate experimental session, the researchers had already learned how much each student like each of the exemplars that he or she had generated. Approximately one month after the initial attitude questionnaire, the same students rated the heights of several celebrities, the last of whom was either the same politician exemplar they had named earlier, a politician the individual liked more, or a politician the individual liked less. Immediately after the height-estimation task, in a supposedly unrelated additional experiment, the students completed the same attitude questionnaire they had completed one month earlier. Even when queried directly, no student realized the connection between the height-estimation task and the attitude questionnaires. Figure 6 shows the Time 1 and Time 2 attitudes of students in each condition. As can be seen, the manipulation of exemplar accessibility had a significant impact. Students who estimated the height of the same politician they had named earlier tended to retain their initial attitudes. Students who estimated the height of an exemplar they liked more than their earlier first politician exemplar changed their attitudes toward politicians in a positive direction, whereas students who estimated the height of an exemplar they liked less changed their attitudes in a negative direction. Consistent with many other studies of temporarily accessible constructs (Schwarz & Bless, 1992; Smith, 1992; Tesser, 1978; Wilson & Hodges, 1992), manipulations that alter the probability of activating different aspects of a representation can change attitudes without the presentation of additional information.
E. ART AND ALLPORT’S “CONTACT HYPOTHESIS” The ART model shown in Figure 1 also has implications for attitude change that occurs in response to new information. According to the ART perspective, such change can occur either when new information alters the weights attached to various aspects of a representation or when it changes the underlying representation itself. One instance of the latter case that is of particular social and theoretical interest occurs when a new, previously unknown exemplar becomes part of the way an individual cognitively represents an attitude object. Just as a temporarily accessible but previously known exemplar such as Bill Clinton might suggest different characteristics than would be suggested by a different politician who was more accessible the last time an attitude was measured, so a new, previously unknown
ATTITUDE REPRESENTATION THEORY
0.5
289
0 Pre-height estimation 0 Post-height estimation
0 cn C
3..-
-0.5
U
Ti E
!
-1
Y
9)
U
.-3 - 1 . 5
s
-2
-2.5
liked
Same exemplar
More liked
Politican whose height was estimated
Fig. 6. Mean pre- and postmanipulation attitudes toward politicians, for students who had recently estimated the height of a politician they liked less, the same politician, or a politician they liked more than the politician they had named when their attitudes were assessed 1 month earlier. (Data from Sia, Lord, Blessurn, Ratcliff, & Lepper, 1997. Experiment 3.)
exemplar might become more accessible than those previously associated with the attitude, changing the default assumptions associated with that category. Attitudes might change because the assumed exemplar at Time 2 does not match the assumed exemplar from Time 1. This reasoning about the potential impact of new exemplars is part of the rationale behind Allport’s (1954) famous contact hypothesis. People who avoid interacting with members of a negatively stigmatized group will never get to know any of them as potential exemplars of the category, so they will have little reason to change their negative attitudes. When we arrange situations so that people meet and interact positively with one or more members of a previously despised group, we increase the probability that a new, more positive exemplar will supplant whatever negative exemplar had previously symbolized or represented the category (Hewstone & Brown, 1986; Rothbart & John, 1985; Sherif, Harvey, White, Hood, &
290
CHARLES G. LORD AND MARK R. LEPPER
Sherif, 1961). When the well-liked basketball player Magic Johnson announced that he had AIDS, for instance, surveys showed widespread amelioration of negative attitudes toward people with AIDS (Graham, Weiner, Giuliano, & Williams, 1993).
1. Abstract versus Specific Representations Such an analysis of the contact hypothesis suggests an additional aspect of ART that we have not yet discussed. As Smith (1998) demonstrates, an exemplar representation can range from a specific, personally encountered instance to a fuzzy abstraction garnered from media and other secondhand accounts. One of these representations, however, ought to be easier for a new, concrete, and immediate exemplar to supplant. One goal of the contact hypothesis is to present the individual with a specific exemplar who undeniably disconfirms their assumptions (Blaney, Stephan, Rosenfield, Aronson, & Sikes, 1977; Brewer & Miller, 1984; Brown & Wade, 1987; Cook, 1984; Norvell & Worchel, 1981; Slavin, 1979). Contact thus pits a new, more positive exemplar against the old exemplar to determine which will represent the stimulus in the future. The new exemplar presumably has a better chance of winning such a contest if the old exemplar is a vague abstraction than if the old exemplar is a specific, concrete individual such as a person with whom one has had personal experience. Werth and Lord (1992), therefore, asked students to assess their attitudes toward persons with AIDS and to report whether, when they thought of a typical person with AIDS, they had in mind a specific person (whose initials they were asked to provide) or an abstraction. Subsequently, as part of a regularly scheduled small-enrollment class, some of these students had an actual AIDS victim-a very personable man who was an excellent public speaker-visit their classroom, give a brief presentation, and then answer students’ questions for the rest of the class period. During the next class period, all students completed the attitude questionnaire again. Figure 7 shows the mean attitudes of students who had the two types of pre-contact representations. Students who initially represented the category with an abstract exemplar changed their attitudes toward people with AIDS more following contact with an actual AIDS sufferer than did students who initially represented the category with a specific, namable exemplar. At Time 1,before the classroom visit, approximately 45% of students said they represented the category with a specific exemplar. At Time 2, approximately the same percentage of students in classes that had not been visited said they had in mind a specific, namable person. In the classes he visited, the percentage of students who said they had a specific exemplar rose to 68%
291
ATTITUDE REPRESENTATION THEORY
4.2
0 Pre-contact 0 Post-contact
-
4 -
rn
9
3
3.0
-
3.6
-
u)
C 0
r8
;
-g
’
3.4
-
3.2
-
0)
TI
-
3
U
3
1
1
Specific
Abstract
Pre-contact exemplar representation Fig. 7. Mean attitudes toward persons with AIDS by students whose exemplar for this category was a specific person or an abstraction. before and after they had positive contact with an actual AIDS victim. (Data from Werth & Lord, 1992.)
after the visit.6 One of the ways in which the contact hypothesis works, then, is by altering the underlying attitude representation and supplanting old exemplars with new, more positive ones-a process that is easier when the old exemplars are nothing more than vague abstractions.
2. Representativeness and the Problem of Generalization A R T also offers some insight into the vexing problem of generalization in Allport’s (1954) contact hypothesis. Allport believed that contact would, when it entailed equal status and mutual goals, change attitudes not only toward the specific group member or members with whom the contact Naturally, students who had a specific exemplar in mind at Time 1 reported having met a larger number of persons with AIDS in person than did students who had an abstraction in mind. This effect, however. was statistically independent of, and did not qualify, the effect of exemplar specificity on attitude change.
292
CHARLES G . LORD AND MARK R. LEPPER
occurred, but also toward the larger group as a whole-that the beneficial effectsof contact would generalize. Instead, most of the empirical evidence suggests that even when contact is very successful in changing specific attitudes, it often leaves negative attitudes toward the group unchanged (Hewstone & Brown, 1986). The ART explanation is that contact with a positive exemplar will be unlikely to change general attitudes unless it alters the underlying representation of the category by adding one or more new, more positive exemplars or by altering the default assumptions that go with existing exemplars. The important construct, as recognized by many theorists, is representativeness (Hewstone & Lord, 1998; Pettigrew, 1979; Rothbart & John, 1985). The crucial element is “ensuring that in some way the participants in the contact encounters see each other as representatives of their groups and not merely as ‘exceptions to the rule’ ” (Hewstone & Brown, 1986, p. 18). An individual who is an exception, no matter how positively he or she may be regarded after contact, cannot possibly alter the underlying attitude representation. That is why positive contact with a typically dressed member of a disliked group changes general attitudes toward the group more than does positive contact with an atypically dressed member (Wilder, 1984) and why positive contact with a self-disclosing group member is less effective in producing generalization than positive contact with someone who reveals very little individuating information (Scarberry, Ratcliff, Lord, Lanicek & Desforges, 1997). In a relevant study by Desforges et al. (1991), for instance, students assessed their attitudes toward several social groups and categories, including former mental patients. The students also provided a personality trait profile of “the typical former mental patient.” Approximately one month later, the same students participated in an “unrelated” study of cooperative learning. They studied cooperatively for 50 minutes with another student, about whom they had seen background information that led them to believe she had been successfully treated for mental illness. Actually, the other student was a confederate who was blind to the hypothesis, but had been trained to be pleasant and friendly. Before contact, when all they knew was that their partner-to-be was a former mental patient, the students described the personality traits that they expected her to have. This trait profile matched very well the trait profile the students had earlier generated as their cognitive representation of the typical former mental patient, as it would if they were reporting assumed characteristics from an underlying representation. After contact, the students described their partner again, now that they had met her. These after-contact descriptions were uniformly more positive than the expectancies that had been based only on categorical assumptions. Subsequently, in a “separate” experiment conducted by a different experimenter,
ATTITUDE REPRESENTATION THEORY
293
the students completed the initial survey of global attitudes yet again. Some students described the typical former mental patient more positively than they had before contact, but others did not. As predicted, changes in the profile of the typical former mental patient significantly predicted changes in general attitudes. Changes in attitudes toward the group as a whole-the generalization part of Allport’s (1954) contact hypothesis-were significantly related to changes in the underlying representation. In a follow-up study, Desforges, Lord, Pugh, Scarberry, and Ratcliff (1997, Experiment 1) manipulated perceived representativeness by telling students that their interaction partner belonged to not one, but two, groups that they disliked equally (for instance, a former mental patient who was also homosexual or a drug-abuser). Some students were told that the other student had received an affirmative action preference, being selected over other applicants to participate in this study for money, because he belonged to one (or the other) of the two groups and was representing that group. The group that he represented had, naturally, been selected at random. Students in the two conditions liked their partner equally, but, as Figure 8 shows, the proportion of that specific liking that generalized to more
1.2
Nontarget category E Target category
1
S
g
0.8
.-g 0.6
1
0
U
a
U
0.4
0.2
0 Neither
Target
group
group
Partner was chosen to represent
Fig. 8. Mean attitude change toward the nontarget and target social categories following positive contact with a partner who was said to have been chosen as a representative of the target category. (Data from Desforges et al., 1997, Experiment 1.)
294
CHARLES G . LORD AND MARK R. LEPPER
positive attitudes toward each group depended on his being thought of as a group representative. When the confederate was merely said to belong to two groups (one of which was randomly selected for analysis as the “target group”) but represented neither, students translated a statistically equal proportion of their liking for him to each of the groups. When he represented one of those groups, even though membership in that group got him what some might regard as an unfair advantage, students transferred a significantly higher proportion of their liking for him to the group that he represented than the other group to which he belonged. The A R T interpretation is that newly encountered group members, no matter how much they are liked, are not likely to be adopted as exemplars, and their characteristics are not likely to become part of the underlying representation that informs general attitudes, unless they are viewed as “legitimate” representatives of that group.’
F. SUMMARY OF RESEARCH RELEVANT TO THE REPRESENTATION POSTULATE To review, the representation postulate of Attitude Representation Theory holds that a person’s response to any attitude-relevant stimulus depends not only on the perceived properties of that stimulus and the situation surrounding it, but also on the subjective representation of that stimulus by the person. Supporting the A R T model shown in Figure 1, studies have shown that people do often activate such subjective representations when they assess their attitudes and that they base their responses at least in part on the evaluative implications of whichever aspects of the subjective representation were activated (Sia et al., in press). When they activate the same representation at two separate times, they respond consistently (Sia et al., 1997) and are less likely to alter their attitudes in response to persuasive arguments (Sia, Thomas, & Lord, 1998). When they activate a different representation that is more likable or less likable than the first, they change their attitudes in a predictable direction (Sia et al., 1997). These studies support the contention that attitude-relevant responses involve activating aspects of the attitude objects’s subjective representation.
’
Related studies (e.g., Smith, Fazio, & Cejka, 1996) showed that people who belong to two social categories (e.g., Mike Tyson as a boxer or as a rapist) are identified more closely with the category toward which the individual has more frequently assessed his or her attitude. One category or the other might also be rendered more important for a particular judgment by the norms that exist within a particular culture (Smith & Zarate, 1992).
ATTITUDE REPRESENTATION THEORY
295
If attitude-relevant responses involve activating aspects of the attitude object’s representation, then it follows that one might change attitudes by changing aspects of the representation. Consistent with this reasoning, other studies have shown that positive contact with one member of a social category is more likely to ameliorate negative attitudes toward the group as a whole when the person’s initial attitude involves an abstract exemplar rather than a specific, namable exemplar (Werth & Lord, 1992). Likewise, positive contact with one group member ameliorates negative attitudes toward the group to the extent that such contact alters the way the group is represented. Furthermore, positive contact with a person who belongs to two equally disliked groups is also more likely to ameliorate negative attitudes toward whichever group the person more clearly represents (Desforges et al., 1991). All of these studies-of exemplar activation, response stability, resistance to persuasion, exemplar accessibility, exemplar specificity, and attitude change via equal-status contact-thus provide results consistent with derivations of the representation postulate. We turn next to studies that have addressed derivations of our second, matching postulate.
IV. Research Relevant to the Matching Postulate The matching postulate of A R T holds that the closer the match between the representations and perceptions to which a person is responding in different situations, the more consistency there will be in that person’s responses to attitude-relevant stimuli. The A R T model in Figure 1 shows four types of representation-based assumptions: target identity, target characteristics, associated responses, and surrounding context. Relevant studies have provided evidence concerning the matching postulate for each of these four types of assumptions. In one set of studies we tested whether consistent exemplar activation promotes consistent attitude-relevant responding (often labeled “attitude-behavior consistency”). In a second set of studies we tested whether a match between assumptions and perceptions of an attitude target’s characteristics promotes attitude-behavior consistency. Other studies tested whether the matching principle can also be applied to matches between assumptions and perceptions with regard to actions and to contexts. Finally, additional studies investigated the impact on the matching principle of factors that increase or decrease the relevance and importance of representation-based assumptions.
296
CHARLES G. LORD AND MARK R. LEPPER
A. CONSISTENT EXEMPLAR ACTIVATION AND RESPONSE GENERALITY Technically, studies that are described in the literature as concerning “attitude-behavior consistency” do not measure the relationship between “a behavior’’ and “an attitude.” According to one widely accepted definition, an attitude is “a psychological tendency that is expressed by evaluating a specific entity with favor or disfavor” (Eagly & Chaiken, 1993, p. 1). A tendency is a bias, inclination, or propensity to do something. It is not a sure thing. When we say that “Susan tends to vote for conservative political candidates,” we do not mean that she alwaysvotes for the more conservative candidate in every election. Any single sample of Susan’s voting behavior, therefore, is just that-only a sample. Similarly,any single attitude-relevant response is merely a sample. An individual’sunderlying attitude is a hypothetical construct-a hypothesized thread of consistency that weaves together many of a person’s thoughts, feelings, and actions. The attitude is something that we infer from one or more attitude-relevant responses. When LaPiere’s (1934) proprietors answered “no” on the questionnaire about serving Chinese people, the answer was not an “attitude.” It was an attitude-relevant evaluative response. When they served the actual Chinese couple, that was also an attitude-relevant evaluative response. One could infer an underlying attitude from either of these responses. It is customary in the literature to call a written or oral response to only the name of the attitude object an “attitude” or “attitude measure” and to call a response to any actual or hypothetical behavioral situation a “behavior” or “behavioral intention,” but technically either could be considered an expression of the underlying attitude. Studies of “attitudebehavior consistency,” then, are actually studies of consistency between two or more evaluative responses to the same stimulus, or studies of response generality. From a practical standpoint, however, a response to the mere name of the attitude object, as in “What is your attitude toward Chinese people” or “What is your attitude toward capital punishment?” ought usually to be the most direct and comprehensive expression of the underlying attitude, because it theoretically includes all possible instantiations of the target, all possible evaluative responses, and all possible contexts. It is also the sort of question that has been most influential in the great success of the attitude concept. Hence, although individual construals of targets, responses, and contexts lie at the heart of ART, in describing studies of “attitude-behavior consistency,” we will follow the tradition and label one of the measures an “attitude” and the other a “behavior” or “behavioral intention.” The
ATTITUDE REPRESENTATION THEORY
297
reader should remember, however, that an answer to a single attitude question is merely an expression of an underlying tendency and should not be confused with the attitude itself.
B. MATCHING AND ATTITUDE-BEHAVIOR CONSISTENCY That said, let us examine the implications of A R T for attitude-behavior consistency. Consider one common case: Suppose that the attitude-relevant response required at Time 2 in Figure 1 is not a second attitude assessment, but is instead an assessment of behavior or behavioral intentions toward the category as a whole, rather than toward specific category members. ART would argue that when people make an attitude-relevant response to a general stimulus rather than to a specific one, they will use the default values that they assume from their underlying representation of the stimulus. People who activate Bill Clinton as their exemplar of politicians when they assess their attitudes, for instance, might also activate Bill Clinton as their politician exemplar when they decide whether to support an otherwise unspecified politician or politicians (e.g., to sign a petition to encourage university administrators to bring more “politicians” t o campus as guest lecturers). They have no information, that is, that would allow them to replace Bill Clinton with a different exemplar. With no behavioral target other than the group as a whole, attitudebehavior consistency will be greater when the assumed target identity at Time 1 matches rather than mismatches the assumed target identity at Time 2. Put another way, people who are thinking about Bill Clinton at both times-when they assess their attitudes and when they choose a behavior or behavioral intention toward otherwise unspecified “politicians”-will display greater consistency or generality across evaluative responses than will people who are thinking about Bill Clinton on only one of the two occasions and about a different politician on the other occasion. Furthermore, it should be possible to predict not only the occurrence of attitudebehavior inconsistency, but also the direction of attitude-behavior inconsistency. To test these hypotheses, Sia and her colleagues (1997, Experiment 2) obtained students’ attitudes and first exemplars for politicians and several other social categories in an initial questionnaire. Two weeks later, the same students participated in an “unrelated” study in which they reported their liking for 150 well-known objects and people (e.g., Rice Krispies or Madonna), a few of which were the politician exemplars the individual student had named earlier. Two weeks after that, the students again completed the initial questionnaire, in which they provided politician and other
298
CHARLES G. LORD AND MARK R. LEPPER
category exemplars for a second time. This questionnaire session, however, was “unexpectedly” interrupted by a representative of the student council who had supposedly received university permission to distribute petitions for bringing an increased number of homosexuals, environmentalists, and politicians to speak on campus. For each of these three (otherwise unspecified) groups, the student was asked to sign the petition, and to agree to write letters, make phone calls, work in groups, and commit to other supportive activities. Figure 9 shows how well students’ initial attitudes toward politicians predicted subsequent behavior (signing the petition) and behavioral intentions (expressing a willingness to engage in the other supportive behaviors). For both measures, attitude-behavior consistency was higher for students who named the same politician to represent the category both times (matching exemplars) than for students who named different politicians (mismatching exemplars). Further analyses revealed a strong correlation, for
o.6
1
0 Mismatching exemplar 0 Matching exemplar
2 0.4 zz -Em 0.3 88 0.2 ‘E B O*’ 0.5
0
I
al
2 CI
0-
E
a
-0.1
-
-0.3 --
-0.2
I
Signing the petition
1
Other supportive activities Behavioral response
Fig. 9. Attitude-behavior correlations for behavior (signing a petition) and behavioral intentions (willingness to engage in other supportive activities) toward otherwise unspecified “politicians,” by participants who named the same or a different politician exemplar 1 month apart. (Data from Sia, Lord, Blessum, Ratcliff, & Lepper, 1997, Experiment 2.)
ATTITUDE REPRESENTATION THEORY
299
those who named a different exemplar, between the difference in liking for the two exemplars and the extent to which students behaved either more positively or more negatively than would have been expected from their initial attitudes. Those who named a politician exemplar they liked better than the one they had named a month earlier made more positive behavioral decisions than their earlier attitudes had suggested. Those who named a politician exemplar they liked less made more negative behavioral decisions than their earlier attitudes had suggested. ART thus allowed researchers to predict not only the occurrence of attitude-behavior inconsistency, as had been predicted in dozens of earlier studies, but also its specific direction. C. MATCHING TARGET IDENTITIES The study just described was designed to test the impact of consistent exemplars on attitude-behavior consistency when the behavior is directed toward the group as a whole. Another frequent scenario occurs when behavior is directed toward one or more individuals who are identified as a subtype within the larger group. Many categories have a hierarchical structure, with distinctions at various levels of the hierarchy that play an important role in perceptions and judgments concerning both natural (Rips, Shoben, & Smith, 1973; Rosch, 1978) and social objects (Brewer, Dull, & Lui, 1981; Brewer, Ho, Lee, & Miller, 1987; Hewstone & Brown, 1986; Stangor, Lynch, Duan & Glas, 1992; Weber & Crocker, 1983). Politicians, for instance, include subtypes such as Democrats and Republicans, plus finer distinctions such as southern Democrats and northern Democrats. From this perspective, an individual exemplar, such as Bill Clinton or the Chinese man who accompanied LaPiere (1934), represents the most specific level in the representational hierarchy for social categories. ART, then, makes the same prediction for matching subtypes and attitude-behavior consistency as it does for matching exemplars and attitudebehavior consistency. Attitudes will predict behavior and behavioral intentions better when assumed and perceived target subtypes match rather than mismatch. If LaPiere‘s (1934) proprietors had in mind the subtype “Chinese laborers” and not “Chinese academics” when they completed the attitude questionnaire, then their attitudes toward members of the Chinese race might have predicted their behavior had a coolie laborer requested service, but not when an academic couple requested service. To test this derivation, Ramsey, Lord, Wallace, and Pugh (1994, Experiment 2) asked undergraduate students to report their attitudes toward former substance abusers and to write a paragraph describing the typical
300
CHARLES G . LORD AND MARK R. LEPPER
former substance abuser. The researchers classified these descriptions according to the subtype of substance abuser the student had in mind. Most of the descriptions involved one and only one type of substance abuser, such as an alcohol, cocaine, heroin, or marijuana abuser. Approximately three weeks later, the 80% of the students who had expressed negative attitudes toward “substance abusers” participated in an “unrelated” study in which they were told that students from their own university were working in pairs with partners from another university to conduct a joint “survey of campus life.” To facilitate the survey, the students were allowed to read descriptions of two potential partners and to indicate, for each of 23 possible activities that varied in psychological closeness, from “just be introduced” to “host the student for the weekend,” how willing they would be to interact in that way with each of the potential partners. Unknown to the students, the two descriptions had been carefully prepared so that pretest participants had rated them as equally likable, but one of these potential partners was said to have been treated successfully for abusing the substance the student had described, whereas the other was said to have been treated successfully for abusing one of the other substances (and was thus identified as belonging to a different or mismatching subtype). As predicted by ART, students displayed greater attitude-behavior consistency toward a substance abuser who was identified as belonging to the subtype they had described when they had earlier assessed their attitudes than toward a substance abuser who was identified as belonging to a different subtype. Figure 10 shows attitude-behavior correlations, both for the number of activities the students were willing to undertake with each of the potential partners and for the psychological closeness of the average activity to which they agreed. Clearly, attitudes predicted these supportive activities and their psychological closeness better when the target was identified as from a matching than from a mismatching subtype. The subtypes study also yielded two other results that supported the matching postulate of ART. First, one week later, all students returned, expecting to meet their partner. The experimenter randomly told students that their partners were from either the matching or a mismatching subtype. The partner had apparently left the room momentarily, but had left some books where he or she was sitting. The students, all of whom had negative attitudes, chose to sit farther from a substance abuser of the subtype that matched rather than mismatched the subtype they had earlier described as representing the typical category member. Second, an additional group of students were told only that their potential partner was a former substance abuser, but got no descriptions that would help them to identify the subtype. According to ART, these additional
ATTITUDE REPRESENTATION THEORY
301
Mismatching subtype Matching subtype
0.6
0.5 16 C 0 P
-m
0.4
V
5
0.3
z
0.2
.-6
4
s.-
0.1
c
*: 4
0 -0.1
Supportive activities
Psychological closeness
Behavioral response
Fig. 10. Attitude-behavior correlations for supportive activities and psychological closeness of those activities toward a substance abuser who was said to belong to a subtype that matched or mismatched the substance abuser subtype participants described when they assessed their attitudes. (Data from Ramsey, Lord, Wallace, & Pugh. 1994, Experiment 2.)
students, with the subtype unspecified, should have retained their assumptions about the target’s identity and assumed that the partner belonged to the subtype they had in mind when they assessed their attitudes. Their behavioral responses should have been equivalent to those of students who got exactly the subtype they expected. Consistent with this reasoning, the behavioral responses of the additional students were almost identical to those of students who got the “confirming” description of someone from the subtype they associated with their attitude, and significantly different from those of students who got a “disconfirming” description of someone from a different subtype.
D. MATCHING TARGET CHARACTERISTICS
Sometimes a behavioral target is unspecified, as it was in the study of bringing “politicians” to campus, or only partly specified, as in the subtype
302
CHARLES G. LORD A N D MARK R. LEPPER
study. More frequently, the behavioral response is directed toward a specific target with known characteristics. LaPiere’s (1934) proprietors, for instance, did not have to decide whether to serve a tour bus full of unspecified “Chinese people,” but instead whether to serve a specific Chinese couple, whose characteristics may have been different from those that the proprietors had in mind when they had assessed and reported their attitudes. At such times, the A R T prediction is clear: Response consistency depends on the match between the assumed and perceived target characteristics. Attitudes will predict behavior better when assumed and perceived target characteristics match rather than mismatch. To test this hypothesis, which concerns target characteristics rather than the identity or subtype of exemplars, Lord, Lepper, and Mackie (1984, Experiment 1) asked Princeton undergraduates to express their attitudes toward various campus eating clubs (similar to fraternities and sororities) and to list six trait characteristics that they attributed to “the typical member” of each club. One student, for instance, described the typical member of Cap & Gown as ambitious, aggressive, playful, approval-seeking, domineering, and persistent. Approximately three months later, in an “unrelated” study, the same students learned that they would be working closely on a task with one of two potential partners. To help the students express a preference for working with one or the other of these two potential partners, the experimenter let them read two “thumbnail sketches” that had supposedly been written by an experienced clinical psychologist as part of an earlier interview study. According to the thumbnail sketches, both potential partners belonged to the same eating club-one of those on the earlier attitude questionnaire. Although the student did not realize it, the two sketches had also been meticulously prepared and pretested so that, although the two potential partners were equally likable, one had 100% of the trait characteristics the individual student had earlier attributed to the typical member of the club, whereas the other potential partner had only 50% of those characteristics and was portrayed as exactly opposite to the student’s reported attributions on the other 50%. Attitudes toward the eating club predicted commitment to work with the 100% typical club member ( r = .69) significantly better than they predicted commitment to work with the 50% typical club member ( r = .32). An ART analysis is that for six of the 100%typical member’s characteristics, assumed and perceived target characteristics were identical-a match found on only three of the 50% typical member’s characteristics. For the other three characteristics, perceived characteristics were opposite to assumed characteristics.
ATTITUDE REPRESENTATION THEORY
303
The experimenters also asked participants, at the end of the experiment, to describe the “typical” member of each eating club again. LaPiere’s (1934) proprietors reported no less negative attitudes when they had previously served an obviously cultured Chinese couple than when they had not, so it seemed unlikely that a one-paragraph written description, no matter that it was 50% disconfirming, would have much impact on the underlying representation itself (see Lord, Ross, & Lepper, 1979). Consistent with this reasoning, the students changed their portraits of the “typical” club member very little, and changed their portraits no more on the trait dimensions that the potential partner’s description had explicitly disconfirmed than on the trait dimensions that a potential partner’s description had explicitly confirmed. In short, the underlying attitude representation had escaped unscathed, ready to be activated the next time an evaluative response was required. The A R T principle of matching target characteristics applies to all types of representations.* In associated systems theory (Carlston, 1994), for instance, cognitive representations can include not only verbal propositions such as personality traits, but also visual images. In addition, cognitive representations need not correspond with reality. An individual might get a visual image of the typical gay male as dressed in drag, for instance, or the typical mental patient as having very wide, frantic eyes and wild, bushy hair. It does not matter whether these mental images are fanciful or realistic. According to ART, fanciful mental images should affect attitude-behavior consistency just as much as realistic ones, as long as they are part of the attitude representation. To explore the role of mental images in attitude representations and attitude-behavior consistency, Lord, Desforges, Ramsey, Trezza, and Lepper (1991, Study 1) created attitudes toward a social group from scratch, using a traditional concept formation task. Students viewed yearbook photographs of women with different hairstyles. For each photograph, they guessed whether the woman belonged to an unnamed “organization” (presumably a sorority) at another university, after which the experimenter told them the “correct” answer. In reality, the experimenter told students that the woman belonged to the organization each time she had any two of three key physical characteristics-short hair, dark hair, and bangs-and that she did not belong when she had only one or none of the three. Although students had no idea how they formed the concept, all of them could guess at better than chance, by the end of the training session, which women belonged to the group and which did not. To manipulate attitudes, As Zajonc (1998) suggests, a “representation” need not be purely cognitive. but can include affective and behavioral aspects as well.
304
CHARLES G. LORD AND MARK R. LEPPER
the experimenter also told students each photographed woman’s opinions on controversial issues. As expected (Byrne, 1971), students who were told that most group members agreed with them reported a positive attitude toward the fictitious group, whereas students who were told that most group members disagreed with them reported negative attitudes toward the group. Subsequently, all students attended a class taught by a woman who was said to have been a group member at the other university. The woman wore either her normal hair style, which was medium length, dirty blond, half off the forehead, or she wore a short, dark wig with bangs. The students were asked to evaluate her lecture for a professor in her graduate speech communications course, believing that their evaluations would influence the woman’s grade in the course. Although the students had never seen a member of the organization who had all three of these physicai characteristics during the training phase, they were expected to have formed a visual image of the prototypic group member and to use that image as part of their cognitive representation for the group. Match to the cognitive representation, in turn, was predicted to influence attitude-behavior consistency, even though the visual image had not been the basis for adopting a positive or negative attitude. Attitudes predicted class evaluations better for students who saw the woman lecture in the wig than with her normal hair, presumably because when she wore the wig, she matched a visual image the students had constructed of the typical group member. Similarly, in a follow-up study (Lord et al., 1991, Study 2), students responded more in line with their attitudes toward former mental patients when they had an opportunity to interact socially with a former mental patient who had wide eyes and bushy hair than with a former mental patient who had normal eyes and hair, even though the two men’s looks had been rated as equally likable by pretest students, none of whom thought that the man with wide eyes looked anything other than normal. Whatever characteristics comprise an individual’s attitude representation, whether they be personality traits or physical features, can affect attitude-behavior consistency when perceived target characteristics do not match assumed target characteristics. General attitude will better predict responses to category members who fit one’s stereotype and “look the part” than to category members who do not.
E. MATCHING ASSOCIATED RESPONSES As shown in Figure 1, ART holds that an attitude representation can involve not only default assumptions concerning a target’s identity and
ATTITUDE REPRESENTATION THEORY
305
characteristics, but also default assumptions concerning the types of responses typically associated with that target. Ajzen and Fishbein (1977) were the first to observe that attitudes predict behavior better when the attitude and the behavior correspond in terms of both target and action. Lapiere’s (1934) restaurant and hotel proprietors, for instance, might have responded negatively to Chinese people in many ways, including voting for stricter immigration laws, refusing to donate money to help the Chinese poor, and avoiding public transportation because it might put them in close contact. LaPiere (1934) might have happened to measure the one type of evaluative response (accepting Chinese people as paying customers) in which the proprietors’ deeds would not match their words-even though both responses in LaPiere’s (1934) study did concern the same action. We know from previous research that people’s actions toward an attitude object can alter their attitudes (Bem, 1972; Festinger, 1957). Indeed, actions toward the object have long been regarded as one of the three components that form the basis for attitudes (Breckler, 1984; Esses, Haddock, & Zanna, 1994), and actions are widely regarded as one of three ways in which attitudes are expressed (Eagly & Chaiken, 1993). It seems likely, then, that one aspect of the representation for a stimulus might be concerned with the types of responses that the stimulus typically either has elicited or might elicit from the individual. In assessing their attitudes, that is, people might activate both assumptions about the target and assumptions about attituderelevant actions. If they do, then the consistency or generality of evaluative responses toward that stimulus might be influenced as much by a match between the assumed and perceived actions as by a match between the assumed and perceived target identity or characteristics. Women who say they have a positive attitude toward Bill Clinton might mean they would vote for him to continue as president, not necessarily that they would trust him in a long-term romantic relationship. Thomas, Lord, and Lepper (1998) noted that studies of stereotyping had established the types of personality traits, physical features, and exemplars that people associate with attitudes toward negatively stigmatized or unfairly treated minority groups (Stangor & Lange, 1994), but that no previous studies had identified the types of actions that might be associated with such attitudes. They sought to create an action typology that might be used to determine whether a match between assumed and perceived actions would promote response consistency. In terms of the ART model shown in Figure 1, Thomas and her colleagues (1998) wanted a typology they could use to test whether attitudes predict behavior better when assumed and perceived behavioral responses match than when they do not match. In an initial study, the researchers asked 100 college students to assess their attitudes toward gay males, politicians, and persons with AIDS. For
306
CHARLES G. LORD AND MARK R. LEPPER
each social category, the researchers also asked the students to list up to five actions that they either had performed or might perform toward one or more members of the category, plus up to five actions that people with the opposite attitude to theirs had performed or might perform. After eliminating nonactions such as characteristics and cfeelings, the researchers were left with 1703 different actions. Many of these specific actions were redundant, such as “hug them” and “give them a hug.” Eliminating all such redundancies left 301 specific actions. The researchers classified these 301 specific actions into eight positive and eight negative action types. Naive raters validated the researchers’ placement of the 301 specific actions into the 16 action types. The researchers then asked additional students to rate the similarity of each pair of the eight positive action types and the similarity of each pair of the eight negative action types. Separate ADDTREE clustering solutions produced the tree diagrams shown in Figure 11, from which the researchers concluded that the 16 action types formed eight bipolar action dimensions. Thomas and her colleagues (1998, Experiment 1) further discovered that
’t
r
PURSUE
IL
SEEK CONTACT
IN-rERAcr
46
HELP
SUPPORT GROUP
AGREE W T H
OPPOSE GRO
DISAGREE WITH
I-,RESPECT
Fig. 11. ADDTREE solution derived from similarity ratings of eight types of positive and eight types of negative actions found in the specific actions that students associate with their attitudes toward various social categories. (Data from Thomas. Lord, & Lepper, 1998, Experiment 1.)
ATTITUDE REPRESENTATION THEORY
307
the eight bipolar action dimensions differed in how frequently the initial participants had said they associated each dimension with their attitudes. Of the 1703 specific actions that students had listed, 369 were on the respect versus disrespect dimension; 255 were on the support versus oppose dimension; 229 were on the help versus harm dimension; 202 were on the seek information about versus ignore dimension; 197 were on the pursue versus be aloof dimension; 175 were on the interact versus shun dimension; 161 were on the agree with versus disagree with dimension; and 115 were on the contact versus no contact dimension. These differing frequencies suggested that, at least for these three social categories, some action dimensions are more likely to be part of people’s representations than others. Just as participants in the exemplar study described earlier were more likely to associate Bill Clinton with their attitudes toward “politicians” than other exemplars, so participants in the action types study were more likely to associate actions such as being courteous, rude, accepting or scornful (respect versus disrespect) with their attitudes toward the three categories than actions such as accepting or rejecting their ideas (agree with versus disagree with), or seeking physical closeness or distance (contact versus no contact). Thomas and her associates (1998) believed that these action dimensions and their relative frequencies were not specific to the three social categories, but rather represented general action tendencies that people cognitively associate with social category attitudes. To test this possibility, they reviewed the published literature on attitude-behavior consistency toward social groups or categories. Examining only studies published between 1934 (the publication date of Lapiere’s study) and 1994 that measured attitudes as an individual difference (not as a stereotype), the researchers identified 30 publications that had reported 42 effect sizes for predicting from attitudes to a wide range of behaviors and behavioral intentions. Then they classified each behavioral measure into one of the eight bipolar action dimensions and validated with naive raters that their classifications were reliable. Using meta-analytic techniques (Cooper & Hedges, 1994), the researchers found a significant relationship between the frequency with which the 100 students had spontaneously associated an action dimension with their attitudes and the magnitude of the attitude-behavior correlation for published studies that used such a dimension. Figure 12 shows the mean effect sizes for studies within each of three sets of action dimensions, arranged according to how frequently the 100 students generated each attitude action dimension (Thomas et al., 1998, Experiment 2). The linear arrangement of these mean effect sizes is especially compelling because most of the published studies investigated attitudes toward social categories different from the ones used to develop the attitude action typology. Approximately
308
CHARLES G . LORD AND MARK R. LEPPER 0.6
-
0.5
-
0.4
-
u)
-
.-3
u)
8 a=
0.3
w
0.2
0.1
0 Least frequent actions
Medium frequent actions
Most frequent actions
Studies that used each type of action as “behavior“
Fig. 12. Mean effect sizes in published studies of attitude-behavior consistency that used as the “behavior” action dimensions that varied in how frequently students associated those actions with their attitudes toward social categories. (Data from Thomas, Lord, & Lepper, 1998, Experiment 2.)
half of the published studies, for instance, concerned attitudes toward Blacks, which was not one of the three categories used in developing the typology.
F. MATCHING CONTEXTS
Finally, according to ART, an attitude representation can contain not just assumed target characteristics and types of evaluative response, but also assumptions about the social context that might accompany an evaluative response. When LaPiere’s proprietors assessed their attitudes toward “accepting members of the Chinese race,” for instance, they might have meant not only a particular type of Chinese person and a particular way of serving that person, but also a particular type of context. Ross and his colleagues, for example, have demonstrated the ubiquity with which people answering questions about their own or others’ behaviors, or making a decision about
ATTITUDE REPRESENTATION THEORY
309
how to behave, typically go far beyond the information immediately available t o make inferences about “indiscernible or unspecified” details of the situations in which those behaviors will take, or have taken, place (Griffin & Ross, 1991; Griffin, Dunning, & Ross, 1990; Ross, 1977; Ross & Ward, 1995). This process of construal, which appears to operate in a relatively automatic and unrecognized fashion, may in turn determine the responses or actions of the individual, just as Asch (1940) might have suggested. LaPiere’s (1934) proprietors might have imagined, for instance, that other customers would become irate and openly voice displeasure had they tried to serve a Chinese person-an imaginary circumstance that, at least according to LaPiere’s (1934) account, never arose. The proprietors might also have imagined that it would be very easy for them to eject Chinese customers, whereas the real situation imposed formidable obstacles to doing so. The studies described earlier suggest the benefits for predicting behavior if we can learn more than just an individual’s attitude. In the study of subtypes of former substance abusers (Ramsey et al., 1994), for instance, knowing the assumed target subtype allowed the researchers to predict more precisely than did knowing only the attitude. Of two participants who had equally negative attitudes toward former substance abusers, that is, the researchers could predict that the one who meant by substance abusers “people who used marijuana” would sit farther from an alleged former marijuana abuser. In general, the more we know about the individual’s assumptions regarding target identities, target characteristics, and the type of evaluative response involved, the better we can predict what the individual will do in a specific behavioral situation. Two well-established social psychology theories hold that the predictive power of attitudes is also enhanced by knowing about an individual’s subjective norms and perceived behavioral control. According to the theory of reasoned action (Ajzen & Fishbein, 1980), attitudes guide behavior through a reasoning process that takes into account the subjective norms surrounding a behavior. In one relevant study, Manstead, Proffitt, and Smart (1983) tried to predict from expectant first-time mothers’ attitudes toward breastfeeding their babies to their actual behavior after the baby was born. The researchers realized, however, that attitudes by themselves might not be sufficient, because once they became mothers, many of the women would be subjected to intense social pressure. Before the birth, many of the women might not have discussed their private decision about breast-feeding versus bottle-feeding with others, assuming that everyone else would just quietly go along with their own personal preference. Once the baby was born, in contrast, their mothers, spouses, friends, and relatives might have exerted subtle and not-so-subtle influence on the behavioral situation.
310
CHARLES G. LORD AND MARK R. LEPPER
Because Manstead and his colleagues (1983) realized that people do not act in a social vacuum, they did not merely measure the women’s attitudes toward breast-feeding. Instead, they also asked the women about the subjective norms that they associated with breast-feeding, posing questions such as “How much does your husband want you to breast-feed and how much do you care what he wants?” and “How much does your mother want you to breast feed and how much do you care what she wants?” They then calculated the positive or negative impact of subjective norms by multiplying each person’s assumed desires by how important they were to the expectant mother, and then summing across persons. Attitudes by themselves predicted whether the women would breast-feed, but knowing about the women’s subjective norms allowed the researchers to predict subsequent breast-feeding significantly better than merely knowing about the attitudes by themselves. Information about subjective norms has also added to the predictive power of attitudes in studies of behaviors as diverse as getting exercise, recycling, dating, the use of contraceptives, and AIDS risk prevention (Bentler & Speckart, 1979; Cochran, Mays, Ciarletta, Caruso, & Mallon, 1992; Doll & Orth, 1993; Gallois et al., 1992; Page1 & Davidson, 1984; Vining & Ebreo, 1992;Winslow, Franzini, & Hwang, 1992). The ART interpretation is that, if we are trying to predict from attitudes to behavior, it helps to know what type of facilitating or inhibiting context people assume at the time they assess and report their attitudes and to know whether those assumptions match perceptions of a specific behavioral situation. Subjective norms might be one part of the assumed context; perceived control might be another. In the theory of planned behavior (Ajzen, 1985), people differ in how much control they think they have over whether to behave in a particular way. Perceived behavioral control is sometimes unnecessarily limited by a misperceived lack of time, money, skills, and other resources. People sometimes do not realize how much they can accomplish if they would only try (Bandura, 1977). Assumptions about perceived control may be more salient in the representation of some attitude objects than others. Especially for actions that involve self-discipline such as dieting, however, people often overestimate how much control they can exert. According to the theory of planned behavior, then, we can increase the predictive power of attitudes by knowing how capable people think they are of engaging in a particular behavior. The theory has been well supported by studies of attitude objects as diverse as eating less, getting better grades, and engaging in health-protective behaviors such as breast self-exams (Ajzen & Madden, 1986; Madden, Ellen, & Ajzen, 1992: McCaul, Sandgren, O’Neill, & Hinsz, 1993; Schifter & Ajzen, 1985). In each case, merely knowing that someone has a positive attitude toward engaging in the behav-
AlTITUDE REPRESENTATION THEORY
311
ior only moderately predicts whether he or she will actually d o it. Information about the individual’s perceptions of control allows researchers to predict behavior in the real situation significantly better than merely knowing about attitudes, or even knowing about both attitudes and subjective norms (Ajzen, 1987). According to ART, both subjective norms and perceived control can constitute assumptions within an individual’s representation of a stimulus when assessing his or her attitude. By learning these assumptions about contextual factors, researchers can better predict when behavior in attitude-relevant situations will match versus mismatch general attitudes.
G. WHEN MATCHING REPRESENTATIONS BECOME IRRELEVANT One of the key concepts in the A R T model, as suggested by Figure 1, is that the assumptions that inform responses to a general attitude question can be contradicted by the more direct perceptions that define a specific behavioral situation. Assumptions about barefoot coolies in LaPiere’s (1934) study, for instance, might have been rendered relatively unimportant when a smiling, cultured, well-dressed couple requested service, because perceptions of concrete, specific target characteristics were more immediate and salient. More generally, ART predicts that discrepancies between assumptions and perceptions will have less impact on attitude-behavior consistency when either the assumptions or the perceptions seem vague or unimportant, and thus receive small or zero weights. One such instance arises when particular assumptions play a negligible role in an individual’s attitude representation. Lord, Desforges, Fein, Pugh, and Lepper (1994) applied this reasoning about the importance or relevance of target characteristics to attitudes concerning social policy issues such as capital punishment and welfare. In one of their studies, Lord and his colleagues (1994, Study 4) had students draw “attitude concept maps” about welfare, by starting from a central node labeled “welfare” and drawing whatever nodes and pathways came to mind-a procedure used in previous studies to investigate people’s associative networks (Armbruster & Anderson, 1984; Holley & Dansereau, 1984; Reder & Anderson, 1982). In this study, these graphic representations contained three reliably recognizable types of nodes: person nodes (e.g., minorities, promiscuous women, people lacking in ambition), principle nodes (e.g., social safety net, wasteful use of tax dollars, fosters dependency), and factual nodes (e.g., begun in the 1930s).
312
CHARLES G. LORD AND MARK R. LEPPER
Two weeks after drawing the concept maps, the same students participated in an “unrelated” study in which they pretended to be members of a welfare review board charged with deciding the fate of welfare recipients who had been accused of cheating by having hidden sources of income. In each instance, the caseworker had discovered unusual assets, such as a new projection television set, which the welfare recipient explained by bargain shopping or claimed were merely on loan from someone else. The facts of the case were deliberately ambiguous, so that positive or negative attitudes might bias the student’s decisions about whether to cut off each recipient’s funds. Each student read and made decisions about two case histories. The recipient in one of the case histories matched perfectly the prevailing stereotype of people on welfare. She was said to be intellectually limited and somewhat overwhelmed by her responsibilities, lacking in self-confidence, lacking in drive, and resigned to her impoverished lifestyle. The recipient in the other case history mismatched the prevailing stereotype. She was said to be shrewd and exceedingly ambitious, proud, self-assured, competitive, and wasting an excellent education. These two sets of target characteristics, of course, had previously been rated by other students as equally (un)likable. The researchers counted the number of person nodes and the number of principle nodes in each student’s attitude concept map. Some students included more person nodes than principle nodes. Others included more principle nodes than person nodes. The ART prediction was that students who included relatively many person nodes in their attitude representation of welfare would show less attitude-behavior consistency when faced with mismatching target characteristics than would students who included relatively few person nodes, because target characteristics would matter more to the former than to the latter. Students who included few person nodes in their attitude concept maps would presumably use lower weights for assumed target characteristics, so a mismatch between assumptions and perceptions would be less important for them than for others. Figure 13 shows that this prediction was confirmed. Students who included many person nodes behaved in a more attitude-consistent way (those with positive attitudes deciding to keep this person on the welfare roles and those with negative attitudes deciding to cut her off) toward the matching or stereotypical than the mismatching or nonstereotypical recipient. Students who included few person nodes in their welfare concept maps, however, made no such distinction. When they were asked, after making the decision, how typical each case history seemed of welfare recipients, both “personoriented students” and “principle-oriented students” equally recognized that one recipient was far more typical than the other. Indeed, when asked
ATTITUDE REPRESENTATION THEORY 0.6 0.5
1
313
0 Mismatching recipient
Matching recipient C
.-o
0.4
s .s 5
0.3
-a C
L
0
-g
s.-
0.2 0.1
o -0.1 -0.2
Few person nodes
Many person nodes
Nodes in welfare attitude concept map
Fig. 13. Attitude-behavior correlations for a recipient who matched versus mismatched stereotypical characteristics, for students who included either few or many person nodes in their welfare attitude concept maps. (Data from Lord. Desforges, Fein, Pugh, & Lepper, 1994.)
directly which of the two was more typical, every student answered correctly. To some of them, however, the information about target characteristics was less relevant for making a behavioral decision. The study of attitude concept maps and welfare thus supported the ART prediction that the more heavily weighted one particular feature is within an individual’s attitude representation, the more important matches or mismatches on that feature will be to consistency between two attituderelevant responses. The other side of the coin-the lessened impact of information in the immediate stimulus when that information is made less salient or important-was studied by Blessum, Lord, and Sia (1998), in an investigation of the effects of cognitive load and positive mood on attitude-behavior consistency. According to ART, individuals who assess their general attitudes rely on default assumptions from an attitude representation (e.g., that gay men are sensitive, nonathletic, have feminine characteristics, and display a good sense of humor). When they encounter a man they know to be gay, these representation-based assumptions are integrated with perceived individuat-
314
CHARLES G. LORD AND MARK R. LEPPER
ing characteristics (e.g., that this specific gay man lifts weights and is very serious). To the extent that the representation-based assumptions dominate the impression that the individual forms of a specific gay man, attitudes will predict behavior, because the representation that informs the attitude and the representation that informs the behavior will both be based on assumed default values. When immediate perceptions of individuating characteristics dominate the impression, in contrast, attitudes might not predict behavior, because the perception that informs the behavior might be different from the representation that informs the attitude. Assumptions about target characteristics, as we have seen in the previously described studies, might not match perceptions of a specific category member. Many studies of stereotype impact have identified two factors that reliably increase reliance on representation-based assumptions over more directly perceived characteristics in forming an impression: cognitive load and positive mood (Forgas, 1995; Gilbert, 1989; Mackie & Worth, 1989; Schwarz, Bless, & Bohner, 1991). In one illustrative study of cognitive load, for instance, students who performed a complicated letter-matching task while listening to a tape of a woman they thought was Asian concluded that she was more timid, intelligent, and calm (the stereotypic representation) than did students who listened to the same tape without performing a concurrent task (Gilbert & Hixon, 1991,Experiment 2). Similarly, in one representative study of positive mood, students who read the case history of a man named Juan Garcia found him more guilty of assault (thus relying more on the stereotypic representation of Hispanic men as violent) than did students who read the same case in a neutral mood (Bodenhausen, Kramer, & Susser, 1994). Cognitive load increases reliance on representation-based assumptions over individuating characteristics because it decreases the cognitive capacity available to integrate the assumptions with perceptions of the immediate stimulus. Positive mood, in contrast, increases reliance on representationbased assumptions by reassuring people that all is well, so that in-depth thinking about the target is not required (Bless, et al. 1996; Macrae, Milne, & Bodenhausen, 1994; Schwarz & Clore, 1996). In both circumstances, however, people rely more on their default assumptions and less on perceptions of a person’s individuating characteristics than they might otherwise. Blessum and his colleagues (1998) asked students to report their attitudes toward several social categories, one of which was gay men. Two to three weeks later, in an “unrelated” experiment, the same students were asked to form impressions of two men from written personality descriptions. Before they read the descriptions, some of the students were placed under cognitive load by being given an eight-digit number to rehearse-a manipu-
ATTITUDE REPRESENTATION THEORY
315
lation used previously by Gilbert and Osborne (1989). Other students were placed in an especially positive mood by writing vivid accounts of recent events that made them very happy-a manipulation used previously by Strack, Schwarz, and Gschneidinger (1985). A third group of students merely read the personality descriptions without any preamble. Both of the men were said to be gay, but one had characteristics that fit the student’s stereotype perfectly, whereas the other gay man had several opposite characteristics. The two descriptions, though, had previously been deemed equally likable by another set of students. After the students had read both descriptions, the experimenter revealed that both men would be visiting the campus soon, as new transfer students from other universities. The student participants were asked to volunteer for as many or as few as they wanted of 15 activities that would help a potential transfer student get to know the new university in advance, such as showing him around campus, introducing him to friends, and hosting him for a weekend visit. Figure 14 shows how well attitudes toward gays predicted students’ willingness to help each of the gay transfer students. In the control condition, where neither cognitive load nor positive mood were involved, attitudes predicted willingness to interact with and help the transfer student who had stereotypical characteristics better than they predicted willingness to interact with and help the transfer student with nonstereotypical characteristics, presumably because his assumed and perceived characteristics matched when he was typical, but not when he was atypical. No such difference occurred in the other two conditions. Although students in the three conditions equally realized that one of the gay men was more typical than the other, students who were under cognitive load when they formed an impression of the target behaved as much in line with their general attitudes whether he had typical or atypical characteristics, presumably because their limited cognitive resources made them rely more on assumed characteristics. Similarly, students who were in a positive mood when they formed an impression of the target behaved as much in line with their attitudes whether he had typical or atypical characteristics, presumably because their positive mood reassured them that they did not need to think deeply and instead could rely on what has been termed “heuristic processing” (Chaiken, 1987; Schwarz & Clore, 1996). With either cognitive load or a positive mood, students seemed to be saying “he’s one of them, and that’s good enough for me to know what he’s like and how I should behave toward him.” The message from this study and from the welfare study is that the match between assumptions and perceptions matters only to the extent that the perceptions are salient, important, and thus heavily weighted in the overall evaluation.
316
CHARLES G. LORD AND MARK R. LEPPER 0.6
-
0.55
-
0.5
-
0.45
-
0.4
-
0.35
-
0.3
-
0.25
-
0 Mismatching characteristics
Ci Matching characteristics
llD
1
C
0
8 L
.p
2
s
? s sa 0
0.2
1
-
1
Cognitive load
Positive mood No manipulation
Experimental condition
Fig. 14. Attitude-behavior correlations for participants who formed an impression of a gay male, who matched versus mismatched participants’ stereotypes, while participants were under cognitive load, in a positive mood, or received no experimental manipulation. (Data from Blessum, Lord, & Sia, 1998.)
H. SUMMARY OF RESEARCH RELEVANT TO THE MATCHING POSTULATE
To review, the matching postulate of Attitude Representation Theory holds that the closer the match between the representations and perceptions to which a person is responding in different situations, the more consistency there will be in that person’s responses to various attitude-relevant stimuli. Supporting the ART model in Figure 1, several studies have shown that people who consistently identify the same exemplar to represent a social category also display greater response generalization (attitude-behavior consistency) than do people who identify different exemplars (Sia et al., 1997). Also people who identify a social category with a specific category subtype behave more predictably toward category members who belong
ATTITUDE REPRESENTATION THEORY
317
to that subtype than to a different subtype (Ramsey et al., 1994). These studies support the contention that attitude-relevant responses are more consistent when they involve matching rather than mismatching target identities. Other studies have shown that attitude-relevant responses are more consistent when they involve matching than mismatching target characteristics. Thus people behave more predictably toward social group members whose characteristics match than mismatch what they assumed the typical group member to be like, whether those characteristics were personality traits or physical features (Lord et al., 1984, 1991). Still other studies have shown that attitude-relevant responses are more consistent when they involve matching rather than mismatching types of associated responses. Specifically, attitude-relevant actions toward members of a social category may be divided into eight bipolar attitude dimensions, some of which are more likely to come to mind as ways to express social attitudes (Thomas et al., 1998, Experiment 1). Furthermore, published studies of attitude-behavior consistency, collapsed across target categories, have reported larger effect sizes when they used behavioral measures that matched rather than mismatched the action dimensions that naive participants most frequently mentioned when they assessed their social category attitudes (Thomas e t al., 1998, Experiment 2). Other studies have shown that attitude-relevant responses are more consistent when they involve matching than mismatching contexts. Thus one can predict better from people’s general attitudes to their behavioral responses when the social norms that surround those behavioral responses occur in contexts that match rather than mismatch participants’ assumptions regarding subjective norms (Ajzen & Fishbein, 1980; Manstead et al., 1983). One can also predict better from people’s general attitudes to their behavioral responses when the difficulty of enacting the behavioral response matches rather than mismatches participants’ assumptions about the extent to which they can exert control (Ajzen, 1987; Ajzen & Madden, 1986). Finally, studies derived from ART correctly predicted times when matching might not matter. One such case occurs when an individual’s representation for an attitude object contains few assumptions regarding a social policy’s targets (Lord et al., 1994). Another such time occurs when the individual is under cognitive load or in an especially good mood, because at such times people rely almost exclusively on representation-based assumptions at the expense of immediate perceptions (Blessum et al., 1998). Like any effective process model, then, A R T predicts times when the postulated processes will and will not occur. In the final section of the chapter, we will discuss not only limitations, but also several other issues
318
CHARLES G. LORD AND MARK R. LEPPER
that are raised by examining the similarities and differences between A R T and other attitude theories.
V. Issues Related to the Attitude Representation Theory Model Numerous studies, then, have supported both the representation postulate and the matching postulate of ART. The model built around these studies, though, raises several related issues.
A. A CONTINUUM OF ATTITUDE-RELEVANT SITUATIONS The first step in the ART model shown in Figure 1 is that an individual identifies what we call an “attitude-relevant situation.” We define an attitude-relevant situation as any situation in which the individual has an opportunity to respond evaluatively to the attitude object (Eagly & Chaiken, 1993). Attitude-relevant situations, according to ART, differ in how much immediate perceptual information they afford. At one end of the continuum is a situation in which the individual responds evaluatively to the mere name of the attitude object, with no other information available. In such a relatively impoverished situation the individual is being asked, in theory, to respond to some “average” of all possible identities, characteristics, associated responses, and contexts that involve the attitude object. Thus, this particular evaluative response is often equated with, or at least taken as the most general and direct indicator of, the individual’s underlying attitude-i.e., the underlying psychological tendency to evaluate the attitude object with favor or disfavor (Eagly & Chaiken, 1993). A t the other end of the continuum is first-hand contact with some actual, specific attitude object. The word snake, for instance does not afford the same direct perceptions as an actual snake (Breckler, 1984). The general attitude question presumably involves many assumptions that are derived from a representation of the attitude object, whereas first-hand contact might involve fewer assumptions about the attitude object and more immediate perceptions of a specific stimulus, response type, and context. Most attitude-relevant situations probably fall somewhere on the continuum between a general attitude question and a first-hand encounter. Near the general attitude side of the continuum are situations in which the individual has a little more information available than merely the name of the attitude object. Instead of assessing their attitudes toward “Chinese people,” for instance, people might assess their attitudes toward “Chinese
ATTITUDE REPRESENTATION THEORY
319
professors” (target identity specified), “polite Chinese” (target characteristic specified), “serving Chinese people” (type of response specified), or “serving Chinese people when other customers do not object” (type of response and context specified). Thus the valence of the evaluative response might be altered by the addition of even a small amount of extra information-by knowing, for instance, that the attitude object is Chinese academics rather than Chinese people in general, or that the evaluative response in question involves serving them rather than other ways (cognitive, affective, and/or behavioral) of responding evaluatively. Similarly, as we approach the other, first-hand-encounter end of the continuum, the importance of assumptions about the general class might decrease and the importance of immediate perceptions might increase. Thus the proprietor who is asked by a Caucasian professor about “serving my polite Chinese friends” has available additional information that is almost, but not quite, as compelling as would be available from a firsthand encounter. Depending on the source’s perceived reliability, individuals might regard second-hand information of this type as almost as important as first-hand perceptions, thus decreasing the importance of representationbased assumptions. Such second-hand information about the target’s identity and characteristics, the type of response involved, and the context is often the basis for behavioral intentions rather than for actual behaviors (Ajzen & Fishbein, 1980). Even when the attitude-relevant situation conveys considerable perceptual information about the stimulus, however, it is still possible for representation-based assumptions to influence evaluative responses. “Educated” Chinese people could be interpreted as “having attended elementary school before dropping out.” “Accepting” Chinese people as guests in a restaurant might be interpreted to mean “giving them take-out food from a window at the back of the restaurant.” We know from numerous studies of stereotyping that reality itself can be misinterpreted when assumptions and expectations bias our perceptions (Bruner, 1957; Olson, Roese, & Zanna, 1996; Trope, 1986). People who read the same evidence from a court case, for instance, perceive exactly the same defendant, who engaged in exactly the same criminal activities, as more guilty if his surname triggers assumptions about the propensity of his race or nationality to commit certain types of crimes (Bodenhausen, 1988; Gordon, 1993). Similarly, personality traitrelevant behavior that appears to be inconsistent often varies because individuals differ in how they construe situations that might appear to be objectively similar (Bern & Funder, 1979; Lord, 1982; Mischel, 1968). Finally, when concrete situations appear to afford all the information one would need to respond evaluatively, seemingly obviating the need for representation-based assumptions, many studies have shown that people
320
CHARLES G. LORD AND MARK R. LEPPER
may still differ in their construals of the situation (Ross, 1987,1989). People who are distracted or not motivated to think in depth about the “real” situation, in particular, may often fall back on default assumptions from their representation of a stimulus and rely on heuristic as opposed to systematic processing (Chaiken, 1987; Schwarz & Clore, 1996). It is all too easy for assumptions to influence evaluative responses at all points along the continuum of attitude-relevant situations (Ajzen, 1996).
B. NUMBER AND TYPE OF REPRESENTATIONBASED ASSUMPTIONS The classical explanation of attitude change had been that people altered their evaluations of the same attitude object-for instance, that they changed their opinions of politicians. When it became apparent that such changes were too frequent and too facile to fit what we knew about ourselves and other people, Asch (1940) turned the classical explanation on its head, by speculating that such changes could also occur because people had altered the specific instance that they had in mind. Asch’s (1940) ideas applied only to changes in the specific category exemplar that the individual assumed would represent or symbolize a category. Attitude Representation Theory, in contrast, holds that people assume not only exemplars, but also characteristics of the attitude object, likely responses associated with that object, and the context in which such responses might occur. We chose to include these four aspects of an attitude representation because we believe that they can all play a part in determining an individual’s evaluative response to a stimulus. As we explain in the next section, we do not believe that all four aspects of the object’s representation are always activated every time an individual makes an evaluative response (a positive or negative thought, feeling, or action) toward an attitude object. We do, however, believe that at least one of them is activated for every evaluative response, whether the individual is aware of it or not (Tesser & Martin, 1996). We would rather err on the side of including too many possible aspects of an attitude representation than too few. Some authors, in fact, have suggested an even larger number factors that determine evaluative responses than in ART, including, for instance, the impact of a particular action on the individual’s self-identity (Eagly & Chaiken, 1993). We included the target’s identity among representation-based assumptions for obvious reasons. Asch’s (1940) initial insight about changes in the object of judgment involved the assumed identity of politicians. Smith and Zarate’s (1992) extensive program of research established that exemplars are activated for and influence many types of social judgments, and that
ATTITUDE REPRESENTATION THEORY
321
exemplars are merely instances and need not involve people who are members of a social group or category. “Electrocution” might be an exemplar of “capital punishment.” “Recycling” might be an exemplar of “helping the environment.” Exemplars can also be based on direct experience or only imagined (Smith, 1998). An exemplar of “Chinese people” might be a coolie that a 1930s proprietor actually saw working in the fields or a fictional hired assassin from a lurid novel. In fact, it is sometimes difficult to separate in memory the real exemplars from the imagined ones (Johnson, Hastroudi, & Lindsay, 1993). Exemplars can also occur in subsets (Zarate & Smith, 1990), as has been shown by numerous studies of subtyping (e.g., Hewstone, Hopkins, & Routh, 1992) including the study of substance abuser representations that we described earlier (Ramsey et al., 1994). Finally, A R T argues that an attitude object can be identified not merely as a name (e.g., former mental patients), but also as a mental image that has physical characteristics, just as people can have both verbal propositions and mental images about objects as seemingly abstract as “nuclear war” (Fiske, Pratto, & Pavelchak, 1983). Considerable modern research suggests that the target’s identity constitutes an important part of how we represent not only social categories, but also attitude objects of all types. Particular characteristics or attributes of the attitude object have likewise played an important role in an array of attitude theories. The most frequently included characteristic in early theories of evaluative responses was the object’s instrumental value. In these theories, one could predict an individual’s overall evaluation of an attitude object by asking about the object’s desirable or undesirable attributes, and the probability that the object would have or produce those attributes (Fishbein, 1963, 1967; Peak, 1955; Rosenberg, 1956). Thus people’s attitudes toward engaging in a health-protective behavior could be predicted from their assumptions about whether the procedure would do any good (Rogers, 1983), and people’s attitudes toward racial integration could be predicted by knowing what the individual assumed would be its consequences (Carlson, 1956). A R T acknowledges that instrumental beliefs are important assumptions about the characteristics of an attitude object, but also includes characteristics that are usually referred t o as the elements within a stereotype (i.e., rude and uneducated for Chinese people in the 1930s, or shy and polite for Asian people today). Some of these characteristics may be personality traits and some may be demographic, as in the erroneous assumption that the typical murderer has a dark skin (Lord et al., 1994). Some may also be attached to specific exemplars. “Jogs,” for instance, could be activated as an assumed characteristic of politicians, but only for individuals who have Bill Clinton as their primary politician exemplar.
322
CHARLES G. LORD AND MARK R. LEPPER
A potentially controversial facet of ART is the inclusion of “associated responses” among representation-based assumptions about an attitude object. “It seems redundant,” one might argue, “to say that people base their evaluative responses on other, associated responses to the same attitude object. It is like saying that we base responses on responses.” We argue, however, that modern research suggests just such a possibility. Recall that evaluative responses include thoughts, feelings,and actions-the three ways of expressing an underlying attitude (Eagly & Chaiken, 1993). The first time an individual evaluates an attitude object, the individual necessarily has to take into account whatever information seems relevant. For subsequent evaluations, however, many modern theories hold that an additional piece of information has now been added-namely, the outcome of the previous evaluative response (Eagly & Chaiken, 1993;Judd & Brauer, 1995).Considerable evidence suggests, in fact, that people sometimes remember a previous evaluative response without being able to recall the specific pieces of information on which the response was based, for instance, when we say that “I remember that I loved that book, but I can’t for the life of me remember what it was about” (Chaiken & Stangor, 1987; Dreben, Fiske, & Hastie, 1979; Hastie & Park, 1986; Lingle & Ostrom, 1981; Zajonc, 1980). Self-generated thoughts and feelings from a previous real or imagined encounter with the attitude object clearly influence future evaluative responses to the object (Greenwald, 1968; Petty & Cacioppo, 1986). Another type of evaluative response that deserves to be included in the attitude representation is a real or imagined action toward the attitude object. According to self-perception theory (Bern, 1972), people often base their evaluative responses to an attitude object on their own previous actions. Many models of attitude processes have included past behaviors as influences on current evaluative responses (Bentler & Speckart, 1979; Liska, 1984;Triandis, 1977,1980). Moreover, as part of their mental representations, people can have scripts about their own actions or people’s usual actions toward an attitude object (Abelson, 1976, 1981; Schank & Abelson, 1977). In addition, our own study of effect sizes in published studies of attitude-behavior consistency indicated that the actions laypeople frequently associate with their attitudes have produced higher attitude-behavior correlations than the actions that laypeople infrequently associate with their attitudes toward social categories (Thomas et al., 1998). It would be interesting in this regard to investigate the details of an individual’s assumptions about an attitude object, including the level of specificity at which the individual identifies an attitude-relevant action (Vallacher & Wegner, 1987). Finally, by including previous evaluative responses as part of an attitude representation, ART allows for the influence of numerous factors of which the individual is not aware, such as classical conditioning
ATTITUDE REPRESENTATION THEORY
323
(Cacioppo, Marshall-Goodell, Tassinary, & Petty, 1992; Krosnick, Betz, Jussim, & Lynn, 1992; Staats, 1983: Zanna, Kiesler, & Pilkonis, 1970), mere exposure (Zajonc, 1968), and physical movements (Cacioppo, Priester, & Berntson, 1993; Strack, Martin, & Stepper, 1988). If we agree that attitude assessment can involve assumptions about associated evaluative responses, then it is obvious from theories of attribution that the meaning of such responses can be dramatically altered by the surrounding context (Gilbert, 1998: Jones & Davis, 1965; Kelley, 1967). An action performed in a facilitative context has a different meaning than the same action performed in an inhibiting context. Thus assumptions about contextual factors such as subjective norms and perceived control are likely aspects of an attitude object’s representation (Miniard & Cohen, 1981; Smetana & Adler, 1980). Like the other assumptions about the target’s identity and characteristics, assumptions about contextual factors can be at odds with perceptions of an actual situation. People often fail to realize, for example, the extent to which social norms of tolerance and politeness constrain actual behaviors (Campbell, 1963). When LaPiere’s (1934) proprietors projected that they would refuse service to members of the Chinese race, they might well have assumed that they could rely on the staff and patrons to cooperate rather than protest. Similarly, people often underestimate the skills and resources required to perform an action (Fishbein & Ajzen, 1975). Perceived control over a behavior often falls short of actual control (Buehler, Griffin, & Ross, 1994; Kahneman & Tversky, 1979; Langer, 1975), which is one reason why students’ perceived control over getting an A near the end of a course corresponds better with their actual grades than their perceived control earlier in the course (Ajzen & Madden, 1986). Including contextual factors such as subjective norms and perceived control in the A R T model signals yet another way in which the matching principle can be violated-another way in which assumptions can mismatch perceptions.
C. COMBINING ASSUMPTIONS WITH PERCEPTIONS
In the ART model shown in Figure 1, representation-based assumptions and more direct perceptions are combined to form an overall evaluation of the attitude object. Although individuals might not be aware of the combination process that they use to form an overall evaluation (Tesser & Martin, 1996), many attitude theories and studies have provided important insights. Previous research suggests, for instance, that people base their evaluative responses on a very small subset of all the information they have
324
CHARLES G. LORD AND MARK R. LEPPER
available about a stimulus (Ajzen, 1996; Ajzen & Fishbein, 1980; Eagly & Chaiken, 1998; Higgins, 1996; Wilson & Hodges, 1992). In some instances, people may consider just one salient dimension of the attitude representation or perceived situation to evaluate an attitude object ( Jaccard, 1981; McGuire, 1985; Tversky, 1972), especially if the evaluative task seems simple or trivial (Guerin & Innes, 1989;Zajonc, 1960). When more than one dimension matters, though, the weights assigned to specific assumptions and perceptions during the combination process exert considerable impact on an overall evaluation (Abelson & Levi, 1985;Anderson, 1991;Atkinson, 1958;Feather, 1982; Kaplan & Anderson, 1973;Lewin, 1938). Political conservatives and political liberals, for instance, often agree on the characteristics that they attribute to a social policy, but differ in the weights that they assign to each characteristic, thus arriving at different overall evaluations (Kerlinger, 1984; Tourangeau, Rasinski, & D’Andrade, 1991).Smokers and nonsmokers can agree on the characteristics of smoking, but differ in how heavily they weight those characteristics (van der Pligt & Eiser, 1984), just as Blacks and Whites can agree on the Characteristics of affirmative action, but differ in the weight they assign to each characteristic (Kluegel & Smith, 1986). Especially if an individual knows of both positive and negative characteristics of an attitude object, the individual’s overall attitude can change from one time to the next, merely because of assigning different weights at different times to the same representation-derived assumptions (Bargh et al., 1992; Hass, Katz, Rizzo, Bailey, & Einstadt, 1991; Katz, 1981). Previous research also suggests some factors that can increase or decrease the weights assigned to specific aspects of an attitude representation and/ or to the corresponding attitude-relevant perceptions. First, the scope of the attitude object logically limits the weights assigned to specific assumptions. Frequently, as described earlier, people are asked to assess their attitudes toward a stimulus that is identified only by its name, as in “Chinese people” or “capital punishment.” At such times, according to ART, they are free to assume “default values” for the target’s identity and characteristics, the responses it has elicited or might elicit, and the context surrounding those responses. At other times, people may be asked to report their attitudes toward a stimulus that specifies a particular type of evaluative response, as in “serving Chinese people.” In such cases, they can be more confident that they know the nature of the evaluative response involved, and assumptions about the type of response (e.g., its likely consequences) will merit greater importance than would have been the case had a specific type of evaluative response not been specified. Second, some factors are likely to increase the weights given to immediate perceptions rather than to representation-based assumptions. We have al-
ATTITUDE REPRESENTATION THEORY
325
ready discussed the simple availability of more specific information. Another factor might be a reluctance to use stereotypical assumptions, o r at least to weight them heavily, in arriving at an overall evaluation. Research suggests that prejudiced and nonprejudiced people are equally aware of the characteristics that comprise the stereotype of various racial and ethnic minority groups (Devine, 1989). They differ, however, in their willingness to use those assumptions when forming an overall evaluation (Devine & Monteith, 1993; Yzerbyt, Schadron, Leyens, & Rocher, 1994). In terms of the ART model, prejudiced and nonprejudiced people might activate the same representation of the social group or category, but nonprejudiced people might assign much lower weights to the assumed target characteristics than do prejudiced people, because they feel guilty about relying on representation-based assumptions that might be politically incorrect, or because the social structure discourages stereotyping and prejudice (Devine & Monteith, 1993; Fiske, 1998; Monteith, Devine, & Zuwerink, 1993). Low weights on assumptions might reflect inhibition of a stereotype’s initial activation, inhibition of its use, or both (Bodenhausen & Macrae, 1998; Dijksterhuis & van Knippenberg, 1998; Stangor, Thompson, & Ford, 1998; Stroessner, 1998). Other factors may work in the opposite direction and increase the likely weights accorded to representation-based assumptions and decrease the weights accorded to immediate perceptions. An overall evaluative response to the mere name of an attitude object, which involves no immediate stimulus, for instance, may predict evaluative responses in subsequent behavioral situations when people are not willing or able to pay much attention to the perceptions that the concrete behavioral situation affords (Fazio, 1990; Schuette & Fazio, 1995). At such times, people become “cognitive misers” who fall back on (i.e., give greater weight to) their representationbased assumptions than to specific situational perceptions (Chaiken, 1987; Fiske & Neuberg, 1990; Taylor & Fiske, 1978). Experts on a topic might also be more likely than novices to attach large weights to their representation-based assumptions. In the study we described earlier of responses to the women whose hairstyles matched versus mismatched the mental image we had created for members of a fictitious organization, the average participant displayed greater attitudebehavior consistency toward a matching than a mismatching exemplar (Lord et al., 1991, Experiment 1). Some participants, however, were relatively more expert (i.e., more accurate) at discriminating members from nonmembers, whereas others could tell the difference at just barely better than chance. The novices’ attitudes toward the group predicted behavior toward a woman with the hairstyle that matched their representation, but not their behavior toward a woman with a mismatching hairstyle-as
326
CHARLES G. LORD AND MARK R. LEPPER
though novices took into account both their representation-based assumptions and their more direct perceptions. Experts’ attitudes, in contrast, predicted behavior toward both women, regardless of their hairstyle, as though the experts relied almost entirely on their representation-based assumptions about the group and gave very little weight to their immediate perceptions. In a follow-up study, participants were divided into “experts,” who knew many, and “novices,” who knew few, gay men (Lord et al., 1991, Experiment 2). Again, novices’ attitudes predicted their behavior toward a typical but not an atypical gay man, whereas experts’ attitudes predicted their behavior toward both targets, as though experts relied on heuristic or assumption-dominated processing in the behavioral situation. Expertise, then, may alter the relative weights that people assign to representation-based assumptions versus more direct perceptions in determining their overall evaluation and evaluative response. D. EFFECTS OF EXPERIENCE “Experience” can occur either in reality or in the imagination. A number of studies have shown that direct experience with a real attitude object increases the consistency of evaluative responses toward that object (Fazio & Zanna, 1981). Attitudes toward getting flu shots predicted actually getting the shots better, for instance, for people who had had flu shots previously than for people who had not (Davidson, Yantis, Norwood, & Montano, 1985). Similarly, attitudes toward smoking predicted behavior better for people who had previously smoked than for people who had never smoked (Sherman et al., 1982). These effects of direct personal experience fit well with the ART model shown in Figure 1,if we assume that direct experience will, on average, increase the correspondence between representations and perceptions of attitude objects. Direct experience may also make the representation of an attitude object more accessible (Fazio, Chen, McDonel, & Sherman, 1982). People base their evaluative reactions on whatever information is most readily accessible at the time (Bodenhausen & Wyer, 1985; Fishbein & Ajzen, 1975; Fishbein, Ajzen, & MacArdle, 1980; Mitchell & Olson, 1981; Olson, Toy, & Dover, 1982;Zanna & Rempel, 1988).Thus people are more apt to base evaluations of their lives on marital satisfaction if they have recently thought about their marriages (Schwarz, Strack, & Mai, 1991), and people are more apt to evaluate the president’s performance on the state of the economy if they have recently thought about their incomes (Iyengar & Ottati, 1994). Also, people who have recently recalled their own actions toward an atti-
ATTITUDE REPRESENTATION THEORY
327
tude object resist attitude change better than do people who have not recently recalled their actions (Ross, McFarland, Conway, & Zanna, 1983). In weighting assumptions and perceptions to form an overall evaluation, then, direct experience with the attitude object might lend weight to representation-based assumptions by making them more readily accessible for future evaluative responses (Doll & Ajzen, 1992). Even imagined experience, involving nothing more than thinking about the attitude object in depth, can increase the accessibility of representationbased assumptions. Repeated expression of an attitude using the same type of evaluative response over and over may make the process of combining assumptions with perceptions more efficient ( Judd & Brauer, 1995). With each response, the process becomes more proceduralized (Anderson, 1983). All other aspects of the representation except for the one most recent evaluative response may come to have near-zero weights (Judd & Brauer, 1995). Thus people who are asked about their attitude toward the same stimulus six times in a row can answer very rapidly, merely by repeating the previous overall evaluation (Fazio, 1990). They will also display considerable attitude-behavior consistency (Houston & Fazio, 1989). Repeated evaluative responses to an attitude object might not only make representation-based assumptions more accessible, but might also create a sense of fluency that imparts confidence in those assumptions (Mandler, Nakamura, & Van Zandt, 1987). We get two types of information when representations are activated: the content of the representation itself, and the subjective ease of bringing that content to mind (Schwarz, 1998). Participants in one relevant study reported more favorable attitudes toward BMWs after generating one reason for buying a BMW than after generating ten reasons for buying one, because the ones who had to generate only one reason found the task so easy that such a purchase must be wise, whereas the ones who had to generate ten reasons found the task so difficult that they drew the opposite conclusion (Wanke, Bohner, & Jurkowitz, 1997). Repeated evaluative responses to the same stimulus might thus increase the importance assigned to representation-based assumptions by making them seem to come to mind very easily (Downing, Judd, & Brauer, 1992; Roese & Olson, 1994). Finally, thinking in depth about an attitude object could also decrease the consistency of evaluative responses, according to the matching principle-if it makes accessible a different aspect of the representation than is activated as the basis for a subsequent behavioral response. In several studies, for instance, Wilson and his colleagues have shown that people who think in depth about the characteristics of an attitude object (specifically those characteristics that might justify the attitude) sometimes report more extreme attitudes and sometimes report more moderate attitudes
328
CHARLES G. LORD AND MARK R. LEPPER
(Wilson, Dunn, Bybee, Hyman, & Rotondo, 1984; Wilson, Dunn, Kraft, & Lisle, 1989; Wilson & Kraft, 1993). The reason is that such thinking increases the weights given to the attitude object’s characteristics at the expense of other aspects of the representation such as previous evaluative reactions, often altering the overall evaluation (Wilson et al., 1989). When the individual subsequently enters an attituderelevant situation that involves more direct perceptions, in contrast, target characteristics might not carry as much weight as previous evaluative responses, so the attitude will not predict the subsequent behavior as well. Having people think about their previous affective responses to an attitude object, on the other hand, may increase attitude-behavior consistency for subsequent evaluative responses that are primarily concerned with feelings, rather than primarily concerned with instrumental calculations (Millar & Tesser, 1992). Thus the type of direct experience that we get from actual encounters with an attitude object is likely to make evaluative responses more consistent, but imagined encounters with the attitude object can make subsequent evaluative responses either more or less consistent, depending on the types of perceptions that are most useful for the specific response. We will close with one final speculation.’ Throughout this chapter we have argued, and presented evidence, that people will often think of broad categories and even abstract concepts in terms of specificexemplars. Indeed, several theorists have suggested that people seldom reason with abstractions (Griggs & Cox, 1982; Kolodner, 1992; Reich & Ruth, 1982). Instead, their reasoning is much more concrete, based on knowledge about specific situations. They make judgments about abstract concepts by using specific remembered instances (Brooks, 1987; Estes, 1993;Hintzman, 1986;Nosofsky, 1986; Reed, 1972; Shin & Nosofsky, 1992). They find it easier to reason with concrete examples than with abstractions (Bracewell & Hidi, 1974; Gilhooly & Falconer, 1974; Johnson-Laird, Legrenzi, & Sonino-Legrenzi, 1972), especially when they are familiar with a particular judgment task or decision (Griggs & Cox, 1982), as people presumably are with assessing their own attitudes. Beyond that, however, why might people prefer to think of abstract attitude objects in concrete terms? One possible perspective on this question is an evolutionary one. Perhaps, early in our ancestral history, before the ability for abstract, conceptual thinking had become highly developed, there may have been adaptive advantages to the development of a process that automatically evaluated the concrete stimuli that the organism encountered in terms of associated positive or negative affect (e.g., Zajonc, 1980, 1998).
’
We owe this idea to Daniel Gilbert, who could no doubt have expressed it much more elegantly.
ATTITUDE REPRESENTATION THEORY
329
If such affective reactions were indeed an evolutionarily older development than the higher cognitive processes involved in conceptual abstraction, our tendency to think of specific exemplars even when assessing our attitudes to complex categories or abstract social policies may be, in part, the residual of earlier adaptations. We may be “programmed,” as it were, to reduce the general to the particular in making affective evaluations.
VI. Concluding Remarks The concept of general social attitudes became ubiquitous in social psychology because of their implicit promise of stability across time and consistency across situations in people’s responses to various attitude-relevant stimuli. Empirical and theoretical challenges to these assumptions offered by Asch’s (1940) analysis of the “attitude-object problem” and LaPiere’s analysis of the “attitude-behavior problem,” however, strongly questioned the utility of both of these premises. Like many previous efforts, A R T involves an attempt to specify at least some of the sources of these problems and to identify at least some of the conditions under which attitudes will show cross-temporal stability versus instability and cross-situational consistency versus inconsistency. In particular, the ART framework, as outlined in Figure 1, involves two basic postulates. The first, representation postulate holds that a person’s responses to any attitude-relevant object will be a joint product of that person’s current subjective representation of that object or category and that person’s more direct perceptions of any specific attitude object that is immediately present. The second, matching postulate then holds that the consistency or stability of different attitude-relevant responses will depend critically on the degree of match between both the subjective representations and the immediate perceptions of the attitude object at those different times or in those different situations. If the central “message,” in a sentence, of the attitude concept could perhaps be summarized as that to know how a person will act, one must first know the person’s attitudes, then the central “message” of Asch’s and LaPiere’s criticisms of the attitude concept could be summarized as that to know a person’s attitudes, one must first know how that person thinks. Together, as described by the ART model, they point to the critical role of recollection, construal, and other subjective cognitive processes in determining how people will behave in various attitude-relevant situations. As Lord Halifax put it, quite succinctly, over three hundred years ago:
330
CHARLES G. LORD AND MARK R. LEPPER
Could we know what men are most apt to remember, we might know what they are most apt to do.
Acknowledgments We thank Alice Eagly, Daniel Gilbert, Eliot Smith, and Mark Zanna for their helpful comments on an earlier draft. Preparation of this chapter was supported, in part, by National Institute of Mental Health Grant MH-44321 to the second author.
References Abelson, R. P. (1976). Script processing in attitude formation and decision making. In J. S. Carroll & J. W. Payne (Eds.), Cognition and social behavior (pp. 33-46). Hillsdale, NJ: Erlbaum. Abelson, R. P. (1981). The psychological status of the script concept. American Psychologist, 36, 715-729. Abelson, R. P., & Levi, A. (1985). Decision making and decision theory. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (3rd ed., Vol. 1, pp. 231-309). New York: Random House. Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). Theauthoritarian personality. New York: Harper & Row. Ajzen, 1. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckman (Eds.), Action-control: From cognition to behavior (pp. 11-39). Heidelberg: Springer. Ajzen, I. (1987). Attitudes, traits, and actions: Dispositional prediction of behavior in personality and social psychology. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 20, pp. 1-63). San Diego, CA: Academic Press. Ajzen, I. (1996). The social psychology of decision making. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 297-325). New York: Guilford Press. Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84, 888-918. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall. Ajzen, I., & Madden, T. J. (1986). Prediction of goal-directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experimental Social Psychology, 22,453-474. Allport, G. W. (1935). Attitudes. In C. Murchison (Ed.), A handbook of social psychology (pp. 798-844). Worcester, MA: Clark University Press. Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press. Anderson, N. H. (Ed.) (1991). Contributions to information integration theory (Vols. 1,2 and 3). Hillsdale, NJ: Erlbaum.
ATTITUDE REPRESENTATION THEORY
33 1
Anderson, R. C., & Pichert, J. W. (1978). Recall of previously unrecallable information following a shift in perspective. Jourtial of Verbal Learning and Verbal Behavior, 17, 1-12. Armbruster, B. B., & Anderson, T. H. (1984). Mapping: Representing informative text diagrammatically. In C. H. Holley & D. F. Dansereau (Eds.). Spatial learning straregies: Techniques, applications, and related issues (pp. 189-209). San Diego, CA: Academic Press. Asch, S. E. (1940). Studies in the principles of judgments and attitudes: 11. Determination of judgments by group and by ego standards. The Journal of Social Psychology, S.P.S.S.1. Bulletin, 12, 433-465. Atkinson. J. W. (Ed.) (1958). Motives in fantasy, action, and society: A method of assessment and siudy. Princeton, NJ: Van Nostrand. Bandura, A. (1977). Self-efficacy:Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215. Bargh, J. A. (1992). The ecology of automaticity: Toward establishing the conditions needed to produce automatic processing effects. American Journal of Psychology, 105, 181-199. Bargh, J. A. (1996). Automaticity in social psychology. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 169-183). New York: Guilford Press. Bargh, J. A,, Chaiken, S., Govender, R., & Pratto, F. (1992). The generality of the automatic attitude activation effect. Journal of Personaliry and Social Psychology, 62, 893-912. Bellezza, F. S. V. (1984). Reliability of retrieval from semantic memory: Common categories. Bulletin of the Psychonomic Society, 22, 324-326. Bern, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experirnentrd social psychology (Vol. 6, pp. 1-62). San Diego. CA: Academic Press Bem, D. J., & Funder, D. C. (1979). Predicting more of the people more of the time: Assessing the personality of situations. Psychological Review, 85, 485-501. Bentler, P. M., & Speckart, G. (1979). Models of attitude-behavior relations. Psychological Bulletin, 86, 452-464. Blaney, N. T., Stephan. S., Rosenfield, D., Aronson, E., & Sikes, J. (1977). Interdependence in the classroom: A field study. Journal of Educational Psychology, 69, 121-128. Bless, H.. Clore, G. L., Schwarz, N., Golisano, V., Rabe, C., & Wolk. M. (1996). Mood and the use of scripts: Does a happy mood really lead to mindlessness? Journal of Personality and Social Psychology, 71, 665-679. Blessum, K. A,, Lord, C. G., & Sia, T. L. (1998). Cognitive load and positive mood reduce typicality effects in attitude-behavior consistency. Personality and Social Psychology Bulerin, 24, 496-504. Bodenhausen, G. V. (1988). Stereotyopic hiases in social decision making and memory: Testing process models for stereotype use. Journal of Personality and Social Psychology, 55, 726-737. Bodenhausen, G. V., Kramer. G. P., & Siisser K. (1994). Happiness and stereotypic thinking in social judgment. Journal of Personality and Social Psychology, 66, 621-632. Bodenhausen. G. V., & Macrae, C. N. (1998). Stereotype activation and inhibition. In R. Wyer (Ed.), Advances in social cugnifion (Vol. 1 1 , pp. 1-52). Mahwah, NJ: Erlbaum. Bodenhausen, G. V., Schwarz, N., Bless, H., & Wanke, M. (1995). Effects of atypical exemplars on racial beliefs: Enlightened racism or generalized appraisals? Journal of Experitnentul Sociul Psychology, 31, 48-63. Bodenhausen, G. V., & Wyer. R. S. (1985). Effects of stereotypes on decision making and information-processing strategies. Journal of Personality and Social Psychology, 48, 267-282. Bracewell, R. J., & Hidi, S. E. (1974). The solution of an inferential problem as a function of stimulus materials. Quarterly Journal of Experimenial Psychology, 26, 480-488.
332
CHARLES G. LORD AND MARK R. LEPPER
Breckler, S. J. (1984). Empirical validation of affect, behavior, and cognition as distinct components of attitude. Journal of Personality and Social Psychology, 47, 1191-1205. Brewer, M. B., Dull, V., & Lui, L. (1981). Perceptions of the elderly: Stereotypes as prototypes. Journal of Personality and Social Psychology, 41, 656-670. Brewer, M. B., Ho, H. K., Lee, J. Y., & Miller, N. (1987). Social identity and social distance among Hong Kong school children. Personality and Social Psychology Bulletin, 13, 156-165. Brewer, M. B., & Miller, N. (1984). Beyond the contact hypothesis: Theoretical perspectives on desegregation. In N. Miller & M. B. Brewer (Eds.), Groups in contact: The psychology of desegregation (pp. 281-302). San Diego, CA: Academic Press. Brooks, L. (1987). Decentralized control of categorization: The role of prior processing episodes. In U. Neisser (Ed.), Concepts and conceptual development (pp. 141-174). Cambridge: Cambridge University Press. Brown, R., & Wade, G. (1987). Superordinate goals and intergroup behavior: The effect of role ambiguity and status on intergroup attitudes and task performance. European Journal of Social Psychology, 17, 131-142. Bruner, J. S. (1957). On perceptual readiness. Psychological Review, 64, 123-152. Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the “planning fallacy”: Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67, 366-381. Byrne, D. (1971). The attraction paradigm. New York: Academic Press. Cacioppo, J. T., Marshall-Goodell, B. S., Tassinary, L. G., & Petty, R. E. (1992). Rudimentary determinants of attitudes: Classical conditioning is more effective when prior knowledge about the attitude stimulus is low than high. Journal of Experimental Social Psychology, 28, 207-233. Cacioppo, J. T., Pnester, J. R., & Berntson, G. G. (1993). Rudimentary determinants of attitudes: 11. Arm flexion and extension have differential effects on attitudes. Journal of Personality and Social Psychology, 65, 5117. Campbell, D. T. (1963). Social attitudes and other acquired behavioral dispositions. In S . Koch (Ed.), Psychology: A study of a science (Vol. 6, pp. 94-172). New York: Mcgraw-Hill. Carlson, E. R. (1956). Attitude change through modification of attitude structure. Journal of Abnormal and Social Psychology, 52, 256-261. Carlston, D. E. (1994). Associated systems theory: A systematic approach to cognitive representations of persons. In R. S . Wyer (Ed.), Advances in social cognition (Vol. 7, pp. 1-78). Hillsdale, NJ: Erlbaum. Carlston, D. E., & Skowronski, J. J. (1994). Savings in the relearning of trait information as evidence for spontaneous inference generation. Journal of Personality and Social PSyCholOgy, 66, 840-856. Chaiken, S. (1987). The heuristic model of persuasion. In M. P. Zanna, J. M. Olson, & C. P. Herman (Eds.), Social influence: The Ontario Symposium (Vol. 5, pp. 3-34). Hillsdale, NJ: Erlbaum. Chaiken, S., Pomerantz, E. M., & Giner-Sorolla, R. (1995). Structural consistency and attitude strength. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 387-412). Mahwah, NJ: Erlbaum. Chaiken, S.,& Stangor, C. (1987). Attitudes and attitude change. Annual Review of Psychology, 38, 575-630. Cochran, S. D., Mays, V. M., Ciarletta, J., Caruso, C., & Mallon, D. (1992). Efficacy of the theory of reasoned action in predicting AIDS-related sexual risk reduction among gay men. Journal of Applied Social Psychology, 22, 1481-1501.
ATTITUDE REPRESENTATION THEORY
333
Collins, A,, & Quillian, M. R. (1969). Retrieval time from semantic memory. Journal of Verbal Learning and Verbal Behavior, 8, 240-247. Converse, P. E. (1970). Attitudes and non-attitudes: Continuation of a dialogue. In E. R. Tufte (Ed.), The quantitative analysis of social problems (pp. 168-189). Reading, MA: Addison-Wesley. Cook, S. W. (1984). Cooperative interaction in multiethnic contexts. In N. Miller & M. B. Brewer (Eds.), Groups in contact: Thepsychology ofdesegregation (pp. 155-1 85). Orlando, FL: Academic Press. Cooper, H., & Hedges. L. V. (Eds.) (1994). The handbook of research synthesis. New York: Russell Sage Foundation. Davidson, A. R., Yantis. S.. Norwood, M., & Montano, D. E. (1985). Amount of information about the attitude object and attitude-behavior consistency. Journal of Personality and Social Psychology, 49, 1184-1198. Desforges. D. M., Lord, C. G., Pugh, M. A,, Scarberry, N. C., & Ratcliff, C. D. (1997). Role of group representativeness in the generalization part of the contact hypothesis. Basic and Applied Social Psychology, 19, 183-204. Desforges, D. M.. Lord, C. G.. Ramsey, S. L., Mason. J. A., Van Leeuwen, M. D., West, S. C., & Lepper, M. R. (1991). Effects of structured cooperative contact on changing negative attitudes toward stigmatized social groups. Journal of Personality and Social Psychology, 60, 531-544. Deutscher, I. (1973). What we say/whaf we do: Sentiments and acts. Glenview, IL: Scott, Foreman. Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5-1 8. Devine, P. G., & Monteith, M. J. (1993). The role of discrepancy-associated affect in prejudice reduction. In D. M. Mackie & D. L. Hamilton (Eds.). Affect, cognition, and stereotyping: fnteractive processes in intergroup perception (pp. 317-344). San Diego, CA: Academic Press. Dijksterhuis, A., & van Knippenberg, A. (1998). Inhibition, Aberdeen, and other cloudy subjects. In R. Wyer (Ed.), Advances in social cognition (Vol. 11, pp. 83-96). Mahwah, NJ: Erlbaum. Doll, J., & Ajzen, I. (1992). Accessibility and stability of predictors in the theory of planned behavior. Journal of Personalily and Social Psychology, 63, 754-765. Doll, J., & Orth. B. (1993). The Fishbein and Ajzen theory of reasoned action applied to contraceptive behavior: Model variants and meaningfulness. Journal of Applied Social Psychology, 23, 39.5-415. Dovidio, J. F., Evans, N., & Tyler, R. B. (1986). Racial stereotypes: The contents of their cognitive representations. Journal of Experimental Social Psychology, 22, 22-37. Downing, J. W., Judd, C. M., & Brauer, M. (1992). Effects of repeated expressions on attitude extremity. Journal of Personality and Social Psychology. 63, 17-29. Dreben, E, K., Fiske, S. T., & Hastie, R. (1979). The independence of evaluative and item information: Impression and recall order effects in behavior-based impression formation. Journal of Personality and Social Psychology, 37, 1758-1768. Eagly, A. H., & Chaiken. S. (1993). The psychology of anirudes. New York: Harcourt Brace Jovanovich. Eagly, A. H., & Chaiken, S. (1998). Attitude structure and function. In D. T. Gilbert, S. T. Fiske. & G. Lindzey (Eds.), Handbook of socialpsychology, (4th ed., Vol. 1, pp. 269-322). New York: McGraw-Hill. Edwards, K. (1990). The interplay of affect and cognition in attitude formation and change. Journal of Personality and Social Psychology, 59, 212-216.
334
CHARLES G. LORD AND MARK R. LEPPER
Esses, V. M., Haddock, G., & Zanna, M. P. (1994). The role of mood in the expression of intergroup stereotypes. In M. P. Zanna & J. M. Olson (Eds.), The psychology ofprejudice: The Ontario symposium (Vol. 7, pp. 77-101). Hillsdale, NJ: Erlbaum. Estes, W. (1993). Concepts, categories, and psychological science. Psychological Science, 4, 143-153. Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 23, pp. 75-109). New York: Academic Press. Fazio, R. H., Chen, J., McDonel, E. C., & Sherman, S. J. (1982). Attitude accessibility, attitudebehavior consistency, and the strength of the attitude-object association. Journal of Enperimental Social Psychology, IS, 339-357. Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986). O n the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229-238. Fazio, R. H., & Zanna, M. P. (1981). Direct experience and attitude-behavior consistency. In L. Berkowitz (Ed.), Advances in experimental socialpsychology (Vol. 14, pp. 161-202). New York: Academic Press. Feather, N. T. (Ed.) (1982). Expectations and actions: Expectancy-value models in psychology. Hillsdale, NJ: Erlbaum. Festinger, L. (1957). A theory ofcognitive dissonance. Stanford, CA: Stanford University Press. Festinger, L. (Ed.) (1964). Conflict, decision, and dissonance. Stanford, CA: Stanford University Press. Fishbein, M. (1963). An investigation of the relationships between beliefs about an object and the attitude toward that object. Human Relations, 16, 233-240. Fishbein, M. (1967). Attitude and the prediction of behavior. In M. Fishbein (Ed.), Readings in attitude theory and measurement (pp. 477-492). New York: Wiley. Fishbein, M., & Ajzen, I. (1975). Belief; altitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley. Fishbein, M., Ajzen, I., & McArdle, J. (1980). Changing the behavior of alcoholics: Effects of persuasive communication. In I. Ajzen & M. Fishbein (Eds.), Understanding attitudes and predicting social behavior (pp. 217-242). Englewood Cliffs, NJ: Prentice-Hall. Fiske, S. T. (1998). Stereotyping, prejudice, and discrimination. In D. T. Gilbert, S. J. Fiske, & G. Lindzey (Eds.), Handbook of social psychology (4th ed., Vol. 2, pp. 357-411). New York: McGraw-Hill Fiske, S. T., & Neuberg, S. L. (1990). A continuum of impression-formation, from categorybased to individuating processes: Influences of information and motivation on attention and interpretation. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 23, pp. 1-74). New York: Academic Press. Fiske, S. T., Pratto, F., & Pavelchak, M. A. (1983). Citizens’ images of nuclear war: Content and consequences. Journal of Social Issues, 39, 41-65. Forgas, J. P. (1995). The affect infusion model (AIM): Review and an integrative theory of mood effects on judgments. Psychological Bulletin, I 1 7, 39-66. Gallois, C., Kashima, Y., Terry, D., McCamish, M., Timmins, P., & Chauvin, A. (1992). Safe and unsaf.: sexual intentions and behavior: The effects of norms and attitudes. Journal of Applied Social Psychology, 19, 1521-1545. Gilbert, D. T. (1989). Thinking lightly about others: Automatic components of the social inference process. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 189-21 I). New York, NY: Guilford Press. Gilbert, D. T. (1998). Ordinary personology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook of Social Psychology (4th Ed., Vol. 2, pp. 89-150). New York: McGraw-Hill.
ATTITUDE REPRESENTATION THEORY
335
Gilbert, D. T.. & Hixon, J. G. (1991). The trouble of thinking: Activation and application of stereotypic beliefs. Journal of Personality and Social Psychology, 60, 509-51 7. Gilbert, D. T., & Osborne, R. E. (1989). Thinking backward: Some curable and incurable consequences of cognitive busyness. Journal of Personality and Social Psychology, 57, 940-949. Gilhooly, K. J., & Falconer, W. (1974). Concrete and abstract terms and relations in testing a rule. Quarterly Journal of Experimental Psychology, 26, 355-359. Gilovich. T. (1981). Seeing the past in the present: The effect of associations to familiar events on judgments and decisions. Joirrnal of Personalicy and Social Psychology, 40, 797-807. Gordon, R. A. (1993). The effect of strong versus weak evidence on the assessment of race stereotypic and race nonstereotypic crimes. Journal of Applied Social Psychology, 23, 734-749. Graham, S., Weiner, B., Giuliano, T., & Williams. E. (1993). An attributional analysis of reactions to Magic Johnson. Journal of Applied Social Psychology, 23, 996- 1010. Greenwald, A. G. (1968). Cognitive learning, cognitive response to persuasion, and attitude change. In A. G. Greenwald, T. C. Brock, & T. M. Ostrom (Eds.), Psychologicalfoundations of attitudes (pp. 147-170). New York Academic Press. Greenwald, A. G. (1990). What cognitive representations underlie social attitudes? Bulletin of the Psychononiic Society, 28, 254-260. Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, self-esteem, and stereotypes. Psychological Review, 102, 4-27. Griffin, D. W.. Dunning, D., & Ross, L. (1990). The role of construal processes in overconfident predictions about the self and others. Journal of Personality and Social Psychology, 59, 1128-1 139. Griffin, D., & Ross, L. (1991). Subjective construal, social inference, and human understanding. In M. P. Zanna (Ed.), Advances in experimental socialpsychology (Vol. 24, pp. 319-359). New York: Academic Press. Griggs, R., & Cox. J. R. (1982). The elusive thematic-materials effect in Wason’s selection task. British Journal of Psychology, 73, 407-420. Guerin, B., & Innes, J. M. (1989). Cognitive tuning sets: Anticipating the consequences of communication. Current Psychology: Research & Reviews, 8, 234-239. Hass, R. G., Katz, I., Rizzo, N., Bailey, J., & Einstadt, D. (1991). Cross-racial appraisal as related to attitude ambivalence and cognitive complexity. Personality and Social Psychology Bulletin, 17, 83-92. Hastie, R., & Park, B. (1986). The relationship between memory and judgment depends on whether the judgment task is memory-based or on-line. Psychological Review, 93,258-268. Hewstone, M., & Brown, R. (1986). Contact is not enough: An intergroup perspective on the ‘contact hypothesis.’ In M. Hewstone & R. Brown (Eds.). Contact and conflict in infergroup encounters (pp. 1-44). New York: Basil Blackwell Ltd. Hewstone, M., Hopkins, N., & Routh, D. A. (1992). Cognitive models of stereotype change: Generalization and subtyping in young people’s views of the police. European Journal of Social Psychology, 22, 219-234. Hewstone, M., & Lord, C. G. (1998). Changing intergroup cognitions and intergroup behavior: The rolc of typicality. In C. Sedikides, J. Schopler. & C. Insko (Eds.), lnrergroup cognition and intergroup behavior (pp. 367-392). Hillsdale, NJ: Erlbaum. Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Socialpsychobgyc Handbook ofbasic principles (pp. 131-168). New York: Guilford Press. Higgins, E. T.. & King, G. (1981). Accessibility of social constructs: Information processing consequences of individual and contextual variability. In N. Cantor & J. F. Kihlstrom (Eds.), Personality, cognition, and social interaction (pp. 69-121). Hillsdale. NJ: Eribaum.
336
CHARLES G. LORD AND MARK R. LEPPER
Higgins, E. T., Rholes, W. S., & Jones, C. R. (1977). Category accessibility and impression formation. Journal of Experimental Social Psychology, 13, 141-154. Hintzman, D. L. (1986). “Schema abstraction” in a multiple-trace memory model. Psychological Review, 93, 411-428. Holley, C. D., & Dansereau, D. F. (Eds.) (1984). Spatial learning strategies: Techniques, applications, and related issues. San Diego, CA: Academic Press. Houston, D. A,, & Fazio, R. H. (1989). Biased processing as a function of attitude accessibility: Making objective judgments subjectively. Social Cognition, 7, 51-56. Hovland, C. L., Janis, I. L., & Kelley, H. H. (1953). Communication andpersuasion: Psychological studies of opinion change. New Haven, CT Yale University Press. Iyengar, S., & Ottati, V. (1994). Cognitive perspect ve in political psychology. In R. S. Wyer, Jr. & T. K. Srull (Eds.), Handbook of social cognition (Vol. 2, pp. 143-187). Hillsdale, NJ: Erlbaum. Jaccard, J. J. (1981). Attitudes and behavior: Implications of attitudes toward behavioral alternatives. Journal of Experimental Social Psychology, 17, 286-307. Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological Bulletin, 114, 3-28. Johnson-Laird, P., Legrenzi, P., & Sonino-Legrenzi, M. (1972). Reasoning and a sense of reality. British Journal of Psychology, 63, 395-400. Jones, E. E., & Davis, K. E. (1965). From acts to dispositions. The attribution process in person perception. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 220-266). New York: Academic Press. Judd, C. M., & Brauer, M. (1995). Repetition and evaluative extremity. In R. E. Petty & J. A. Krosnick (Ed.), Attitude strength: Antecedents and consequences (pp. 43-72). Mahwah, NJ: Erlbaum. Kahneman, D., & Tversky, A. (1979). Intuitive prediction: Biases and corrective procedures. TIMS Studies in Management Science, 12, 313-327. Kaplan, M. F., & Anderson, N. H. (1973). Information integration theory and reinforcement theory as approaches to interpersonal attraction. Journal of Personality and Social Psychology, 28, 301-312. Katz, I . (1981). Stigmas: A social psychological analysis. Hillsdale, NJ: Erlbaum. Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium on Motivation, 14, 192-241. Kerlinger, F. N. (1984). Liberalism and conservatism: The nature and structure of social attitudes. Hillsdale, NJ: Erlbaum. Kiesler, C. A., Collins, B. E., & Miller, N. (1969). Attitude change: A critical analysis of theoretical approaches. New York: Wiley. Klein, S. B., & Loftus, J. (1993). The mental representation of trait and autobiographical knowledge about the self. In T. S. Srull & R. S. Wyer, Jr. (Eds.), Advances in social cognition (Vol. 5, pp. 1-49). Hillsdale, NJ: Erlbaum. Klein, S. B., Loftus, J., & Burton, H. A. (1989). Two self-reference effects: The importance of distinguishing between self-descriptiveness judgments and autobiographical retrieval in self-referent encoding. Journal of Personality and Social Psychology, 56, 853-865. Kluegel, J. R., & Smith, E. R. (1986). Beliefs about inequality: Americans’ views of what is and what ought to be. New York: Aldine de Gruyter. Kolodner, J. (1992). An introduction to case-based reasoning. Artificial Intelligence Review, 6, 3-34. Krosnick, J. A., Betz, A. L., Jussim, L. J., & Lynn, A. R. (1992). Subliminal conditioning of attitudes. Personality and Social Psychology Bulletin, 18, 152-162.
ATTITUDE REPRESENTATION THEORY
337
Kutner, H., Wilkins, C., & Yarrow, P. R. (1952). Verbal attitudes and overt behavior involving racial prejudice. Journal of Abnormal and Social Psychology, 47, 649-652. Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology, 32, 311-328. LaPiere, R. T. (1934). Attitudes versus actions. Social Forces, 13, 230-237. Lewicki, P. (1986). Nonconscious social information processing. San Diego, CA: Academic Press. Lewin. K. (1938). The conceptual representation and the measurement of psychological forces. Durham, N C Duke University Press. Lingle, J. H., & Ostrom, T. M. (1981). Principles of memory and cognition in attitude formation. In R. E. Petty, T. M. Ostrom, & T. C. Brock (Eds.), Cognitive responses in persuasion (pp. 399-420). Hillsdale, NJ: Erlbaum. Liska, A. E. (1984). A critical examination of the causal structure of the FishbeinIAjzen attitude-behavior model. Social Psychology Quarterly, 47, 61-74. Logan, G. D. (1989). Automaticity and cognitive control. In J. S. Uleman & J. A. Bargh (Eds.). Unintended thought (pp. 52-74). New York: Guilford Press. Lord, C. G . (1982). Predicting behavioral consistency from an individual’s perception of situational similarities. Journal of Personality and Social Psychology, 42, 1076-1088. Lord, C. G., Desforges, D. M., Chacon. S., Pere, G . , & Clubb, R. (1992). Reflections on reputation in the process of self-evaluation. Social Cognition, 10, 2-29. Lord. C. G., Desforges, D. M., Fein, S.. Pugh. M. A,, & Lepper, M. R. (1994). Typicality effects in attitudes toward social policies: A concept-mapping approach. Journalof Personality and Social Psychology, 66, 658-673. Lord, C. G., Desforges, D. M., Ramsey, S. L.. Trezza, G . R., & Lepper, M. R. (1991). Typicality effects in attitude-behavior consistency: Effects of category discrimination and category knowledge. Journal of Experimental Social Psychology, 27, 550-575. Lord, C. G . , Lepper, M. R., & Mackie, D. (1984). Attitude prototypes as determinants of attitude-behavior consistency. Journal of Personality and Social Psychology, 46, 12541266. Lord, C. G., Ross, L., & Lepper, M. R. (1979). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personaliry and Social Psychology, 37, 2098-2109. Lydon, J. E., & Zanna, M. P. (1990). Commitment in the face of adversity: A value-affirmation approach. Journal of Personality and Social Psychology, 58, 1040-1057. Macht, M. L.. & O’Brien. E. J. (1980). Familiarity-based responding in item recognition: Evidence for the role of spreading activation. Journal of Experimental Psychology: Human Memory and Learning, 6, 301-318. Macht, M. L.. & Spear, N. E. (1977). Priming effects in episodic memory. Journalof Experimental Psychology: Human Memory and Learning, 3, 733-741. Mackie, D. M., & Worth, L. T. (1989). Cognitive deficits and the mediation of positive affect in persuasion. Journal of Personality and Social Psychology, 57, 27-40. Macrae, C. N., Milne, A. B., & Bodenhausen, G. V. (1994). Stereotypes as energy-saving devices: A peek inside the cognitive toolbox. Journal of Personality and Social Psychology, 66, 37-47. Madden, T. J., Ellen, P. S., & Ajzen, I. (1992). A comparison of the theory of planned behavior and the theory of reasoned action. Personality and Social Psychology Bulletin, 18, 3-9. Mandler, G . , Nakamura, Y., & Van Zandy, B. J. S. (1987). Nonspecific effects of exposure on stimuli that cannot be recognized. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 646-648.
338
CHARLES G. LORD AND MARK R. LEPPER
Manstead, A. S. R., Proffitt, C., & Smart, J. L. (1983). Predicting and understanding mothers’ infant-feeding intentions and behavior: Testing the theory of reasoned action. Journal of Personality and Social Psychology, 44, 657-671. McArthur, L. Z., & Baron, R. M. (1983). Toward an ecological theory of social perception. Psychological Review, 90, 215-238. McCaul, K. D., Sandgren, A. K., O’Neill, H., & Hinsz, V. B. (1993). The value of the theory of planned behavior, perceived control, and self-efficacy for predicting health-protective behaviors. Basic and Applied Social Psychology, 14, 231-252. McGuire, W. J. (1969). The nature of attitudes and attitude change. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology (2nd ed., Vol. 3, pp. 136-315). Reading, MA: Addison-Wesley. McGuire, W. J. (1976). Some internal psychological factors influencing consumer choice. Journal of Consumer Research, 2, 302-319. McGuire, W. J. (1985). Attitudes and attitude change. In G. Lindzey & E. Aronson (Eds.), Handbook ofsocialpsychology (3rd ed., Vol. 2, pp. 233-346). New York: Random House. Millar, M. G., & Tesser, A. (1992). The role of beliefs and feelings in guiding behavior: The mismatch model. In L. L. Martin & A. Tesser (Eds.), The construction of social judgments (pp. 277-300). Hillsdale, NJ: Erlbaum. Miniard, P. W., & Cohen, J. B. (1981). An examination of the Fishbein-Ajzen behavioralintentions model’s concepts and measures. Journal of Experimental Social Psychology, 17, 309-339. Mischel, W. (1968). Personality and assessment. New York: Wiley. Mitchell, A. A,, & Olson, J. C. (1981). Are product attribute beliefs the only mediator of advertising effects on brand attitude? Journal of Marketing Research, 18, 318-332. Monteith, M. J., Devine, P. G.. & Zuwerink, J. R. (1993). Self-directed versus other-directed affect as a consequence of prejudice-related discrepancies. Journal of Personality and Social Psychology, 64, 198-210. Neely, J. H. (1977). Semantic priming and retrieval from lexical memory: Roles of inhibitionless spreading activation and limited-capacity attention. Journal of Experimental Psychology: General, 106, 225-254. Neisser, U. (1967). Cognitive Psychology. EnglewoodCliffs, NJ: Prentice Hall. Newell, A., Rosenbloom, P. S., & Laird, J. E. (1989). Symbolic architectures for cognition. In M. Posner (Ed.), Foundations in cognitive science (pp. 93-131). Cambridge, MA: MIT Press. Niedenthal, P. M. (1992). Implicit perception for affective information. Journal of Experimental Social Psychology, 26, 505-527. Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231-259. Norvell, N., & Worchell, S. A. (1981). A reexamination of the relation between equal status contact and intergroup attraction. Journal of Personality and Social Psychology, 41, 902-908. Nosofsky, R. (1986). Attention, similarity, and the identification-categorization relationship. Journal of Experimental Psychology: General, 115, 39-57. Olson, J. M., Roese, N. J., & Zanna, M. P. (1996). Expectancies. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 211-238). New York: Guilford. Olson, J. C., Toy, D. R., & Dover, P. A. (1982). Do cognitive responses mediate the effects of advertising content on cognitive structure? Journal of Consumer Research, 9, 345-362. Olson, J. M., & Zanna, M. P. (1993). Attitudes and attitude change. Annual Review of Psychology, 44, 117-154.
ATTITUDE REPRESENTATION THEORY
339
Ottati, V. C. (1997). When the survey question directs retrieval: Implications for assessing the cognitive and affective predictors of global evaluation. European Journal of Social P.yychology, 27, 1-21. Pagel, M. D., & Davidson, A. R. (1984). A comparison of three social-psychological models of attitude and behavior plan: Prediction of contraceptive behavior. Journal of Personality and Social Psychology, 47, 517-533. Peak, H. (1955). Attitude and motivation. In M. R. Jones (Ed.), Nebraska Symposium on Motivation (Vol. 3, pp. 149-188). Lincoln: University of Nebraska Press. Perdue, C. W.. & Gurtman, M. B. (1990). Evidence for the automaticity of ageism. Journal of Experimental Social Psychology, 26, 199-216, Pettigrew, T. F. (1979). The ultimate attribution error: Extending Allport’s cognitive analysis of prejudice. Per.sonality and Social Psychology Bulletin, 5, 461-476. Petty, R. E., & Cacioppo. J. T. (1981). Attitudes and persuasion: Classic and conretnporary approaches. Dubuque, IA: Brown. Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 19. pp. 124-205). New York: Academic Press. Rajaram, S., & Roediger. H. L. (1993). Direct comparison of four implicit memory tests. Journal of Experimental Psychology: Learning, Memory and Cognition, 19, 765-776. Ramsey, S. L., Lord, C. G., Wallace, D. S., & Pugh. M. A. (1994). The role of subtypes in attitudes towards superordinate social categories. British Journal of Social Psychology, 33, 387-403. Reder, L. M., & Anderson, J. R. (1982). Effects of spacing and embellishment on memory for the main points of a text. Memory and Cognition, 10, 97-102. Reed. S. (1972). Pattern recognition and categorization. Cognitive Psychology, 3, 383-407. Reich, S. S., & Ruth, P. (1982). Wason’s selection task: Verification. falsification, and matching. Brirish Journal of Psychology, 73, 395-405. Rips, L. J.. Shoben, E. J., & Smith, E. E. (1973). Semantic distance and the verification of semantic relations. Journal of Verbal Learning and Verbal Behavior, 12, 1-20. Rock, 1. (1984). Perceptioti. New York: W. H. Freeman. Roediger, H. L. (1990). Implicit memory: Retention without remembering. American Psychologist, 45, 1043-1056. Roese, N. J., & Olson, J. M. (1994). Attitude importance as a function of repeated attitude expression. Joitrnal of Experimental Social Psychology, 30, 39-51. Rogers, R. W. (1983). Cognitive and physiological processes in fear appeals and attitude change: A revised theory of protection motivation. In J. T. Cacioppo & R. E. Petty (Eds.), Social psychophysiology: A sourcebook (pp. 153-176). New York: Guilford Press. Rosch. E. (1978). Principles of categorization. In E. Rosch & B. Lloyd (Eds.), Cognition and categorization (pp. 28-48). Hillsdale, NJ: Erlbaum. Rosenherg. M. J. (1956). Cognitive structure and attitudinal affect. Journal of Abnormal and Social P,yychology, 53, 367-372. Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 10. pp. 173-220). San Diego. CA: Academic Press. Ross. L. (1987). The problem of construal in social inference and social psychology. In N. E. Grunberg, R. E. Nisbett, J. Rodin, & J. E. Singer (Eds.), A distinctive approach to psychological research: The influence of Stanley Schachter (pp. 77-96). Hillsdale, NJ: Erlbaum. Ross, L. (1989). Recognizing construal processes. I n I. Rock (Ed.), The legacy of Solomon Asch: Essays in cognition and social psychology. Hillsdale. NJ: Erlbaum.
340
CHARLES G. LORD AND MARK R. LEPPER
Ross, L., & Ward, A. (1995). Psychological barriers to dispute resolution. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 21, pp. 255-304). San Diego, CA: Academic Press. Ross, M., McFarland, C., Conway, M., & Zanna, M. P. (1983). Reciprocal relation between attitudes and behavior recall: Committing people to newly formed attitudes. Journal of Personality and Social Psychology, 45, 257-267. Rothbart, M., & John, 0.P. (1985). Social categorization and behavioral episodes: A cognitive analysis of the effects of intergroup contact. Journal of Social Issues, 41, 81-104. Scarberry, N. C., Ratcliff, C. D., Lord, C. G., Lanicek, D. L., & Desforges, D. M. (1997). Effects of individuating information on the generalization part of Allport’s contact hypothesis. Personality and Social Psychology Bulletin, 23, 1291-1299. Schank, R. C., & Abelson, R. P. (1971). Scripts, plans, goals, and understanding: An inquiry into human knowledge structures. Hillsdale, NJ: Erlbaum. Schifter, D. B., & Ajzen, I. (1985). Intention, perceived control, and weight loss:An application of the theory of planned behavior. Journal of Personality and Social Psychology, 49, 843-851. Schuette, R. A., & Fazio, R. H. (1995). Attitude accessibility and motivation as determinants of biased processing: A test of the MODE model. Personality and Social Psychology Bulletin, 21, 104-110. Schwarz, N. (1994). Judgments in a social context: Biases, shortcomings, and the logic of conversation. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 26, pp. 123-162). San Diego, CA: Academic Press. Schwarz, N. (1998). Accessible content and accessibility experiences: The interplay of declarative and experiential information in judgment. Personality and Social Psychology Review, 2, 87-99. Schwarz, N., & Bless, H. (1992). Scandals and the public’s trust in politicians: Assimilation and contrast effects. Personality and Social Psychology Bulletin, 18, 514-579. Schwarz, N., Bless, H., & Bohner, G. (1991). Mood and persuasion: Affective states influence the processing of persuasive communications. In M. P. Zanna (Ed.) Advances in experimental social psychology (Vol. 24, pp. 161-199). San Diego, CA: Academic Press. Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45, 513-523. Schwarz, N., & Clore, G. L. (1996). Feelings and phenomenal experiences. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 433-465). New York: Guilford Press. Schwarz, N., Strack, F., & Mai, H. P. (1991). Assimilation and contrast effects in part-whole question sequences: A conversational logic analysis. Public Opinion Quarterly, 55, 3-23. Sedikides, C. (1990). Effects of fortuitously activated constructs versus activated communication goals on person impressions. Journal of Personality and Social Psychology, 58, 397-408. Sherif, M., Harvey, 0. J., White, B. J,, Hood, W. R., & Sherif, C. W. (1961). Intergroup conflict and cooperation: The Robber’s Cave experiment. Norman, OK. University of Oklahoma Press. Sherman, S. J., Presson, C. C., Chassin, L., Bensenberg, M., Corty, E., & Olshavsky, R. W. (1982). Smoking intentions in adolescents: Direct experience and predictability. Personality and Social Psychology Bulletin, 8, 376-383. Shin, H., & Nosofsky, R. (1992). Similarity-scaling studies of dot-pattern classification and recognition. Journal of Experimental Psychology: General, 121, 278-304.
ATTITUDE REPRESENTATION THEORY
34 1
Sia, T. L., Lord, C. G., Blessum, K. A., Thomas, J. C., & Lepper. M. R. (in press). Activation of exemplars in the process of assessing social-category attitudes. Journal of Personality and Social Psychology. Sia, T. L., Lord, C. G., Blessum, K., Ratcliff. C. D., & Lepper, M. R. (1997). Is a rose always a rose? The role of social category exemplar-change in attitude stability and attitudebehavior consistency. Journal of Personality and Social Psychology, 72, 501-514. Sia, T. L., Thomas J. C., Lord, C. G., & Lepper, M. R. (1998). A longitudinal investigation of social category exemplar activation: Consistency imparts resistance to attitude change. Unpublished manuscript. Slavin, R. E. (1979). Effects of biracial learning teams on cross-racial friendships. Journal of Educational Psychology, 71, 38 1-387. Smetana, J. G., & Adler, N. E. (1980). Fishbein’s Value X Expectancy Model: An examination of some assumptions. Personality and Social Psychology Bulletin, 6 , 89-96. Smith, E. R. (1992). The role of exemplars in social judgment. In L. L. Martin & A. Tesser (Eds.), The construction of social judgments (pp. 37-65). Hillsdale, NJ: Erlbaum. Smith, E. R. (1998). Mental representation and memory. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook ofsocialpsychology (4th ed., Vol. 1, pp. 391-445). New York: McGraw-Hill. Smith. E. R., Fazio, R. H., & Cejka, M. A. (1996). Accessible attitudes influence categorization of multiply categorizable objects. Journal of Personality and Social Psychology, 71, 888-898. Smith, E. R., Stewart, T. L., & Buttram, R. T. (1992). Inferring a trait from a behavior has longterm, highly specific effects. Journal of Personality arid Social Psychology, 62, 753-759. Smith. E. R.. & Zarate, M. A. (1992). Exemplar-based model of social judgment. Psychological Review, 99, 3-21. Srull. T. K., & Wyer, R. S. (1979). The role of category accessibility in the interpretation of information about persons: Some determinants and implications. Journal of Personality and Social Psychology, 37, 1660-1672. Staats, A. W. (1983). Paradigmatic behaviorism: Unified theory for social-personality psychology. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 16, pp. 125-179). San Diego, CA: Academic Press. Stangor, C. (1990). Arousal, accessibility of trait constructs, and person perception. Journal of Experimental Social Psychology, 26, 305-321. Stangor, C., & Lange, J. E. (1994). Mental representations of social groups: Advances in understanding stereotypes and stereotyping. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 26, pp. 357-416). San Diego, CA: Academic Press. Stangor, C., Lynch, L., Duan, C., & Glass, B. (1992). Categorization of individuals on the basis ofmultiple social features. Journal of Personality and Social Psychology, 62,207-218. Stangor, C., Thompson, E. P., & Ford, T. E. (1998). An inhibited model of stereotype inhibition. In R. Wyer (Ed.), Advances in social cognition (Vol. 11. pp. 193-210). Mahwah, NJ: Erlbaum. Strack, F. (1992). The different routes to social judgments: Experimental versus informational strategies. In L. L. Martin & A. Tesser (Eds.), The construction of social judgments (pp. 37-65). Hillsdale, NJ: Erlbaum. Strack, F., Martin, L. L.. & Stepper, S. (1988). Inhibiting and facilitating conditions of the human smile: A non-obtrusive test of the facial feedback hypothesis. Journalof Personality and Social Psychology, 53, 168-777. Strack, F., Schwarz, N., & Gschneidinger, E. (1985). Happiness and reminiscing: The role of time perspective, mood, and mode of thinking. Journal of Personality and Social Psychology, 49, 1460-1469.
342
CHARLES G. LORD AND MARK R. LEPPER
Stroessner, S. J. (1998). Varieties of inhibition in social stereotyping. In R. Wyer (Ed.), Advances in social cognition (Vol. 11, pp. 211-226). Mahwah, NJ: Erlbaum. Taylor, S. E., & Fiske, S. T. (1978). Salience, attention, and attribution: Top of the head phenomena. In L. Berkowitz (Ed.), Advances in experimentalsocial psychology (Vol. 11, pp. 249-288). San Diego, CA: Academic Press. Tesser, A. (1978). Self-generated attitude change. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 11, pp. 289-338). San Diego, CA: Academic Press. Tesser, A,, & Martin, L. (1996). The psychology of evaluation. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of bnsic principles (pp. 400-432). New York: Guilford. Thomas, J. C., Lord, C . G . , & Lepper, M. R. (1998). Which behaviors do attitudes predict? A typology of attitude actions. Manuscript in preparation. Thorndike, E. L. (1920). A constant error in psychological ratings. Journal of Applied Psychology, 4, 25-29. Tourangeau, R., & Rasinski, K. A. (1988). Cognitive processes underlying context effects in attitude measurement. Psychological Bulletin, 103, 299-314. Tourangeau, R., Rasinski, K. A,, & D’Andrade, R. (1991). Attitude structure and belief accessibility. Journal of Experimental Social Psychology, 27, 48-75. Triandis, H. C. (1977). Interpersonal behavior. Monterey, CA: Brooks/Cole. Triandis, H. C . (1980). Values, attitudes, and interpersonal behavior. In M. E. Howe, Jr. & M. M. Page (Eds.), Nebrasku Symposium on Motivation, 1979 (Vol. 27, pp. 195259). Lincoln: University of Nebraska Press. Trope, Y. (1986). Identification and inferential processes in dispositional attribution. Psychological Review, 9.7, 239-257. Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science, 247, 301-305. Tulving, E., Schacter, D. L., & Stark, H. (1982). Priming effects in word-fragment completion are independent of recognition memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 8, 336-342. Tversky, A. (1972). Elimination by aspects: A theory of choice. Psychological Review, 7Y, 281-299. Vallacher, R. R., & Wegner, D. M. (1987). What do people think they’re doing? Action identification and human behavior. Psychological Review, 94, 3-15. van der Pligt, J., & Eiser, J. R. (1984). Dimensional salience, judgment, and attitudes. In J. R. Eiser (Ed.), Attitudinal judgment (pp. 161-177). New York: Springer-Verlag. Vining, J., & Ebreo, A. (1992). Predicting recycling behavior from global and specific environmental attitudes and changes in recycling opportunities. Journal of Applied Social Psychology, 22, 1580-1607. Wanke, M.. Bohner, G., & Jurkowitz, A. (1997). There are many reasons to drive a BMW: Does imagined ease of argument generation influence attitudes? Journal of Consumer Research. Watts, W. A., & Holt, L. E. (1979). Persistence of opinion change induced under conditions of forewarning and distraction. Journal of Personality and Social Psychology, 37,778-789. Weber, R., & Crocker, J. (1983). Cognitive processes in the revision of stereotypic beliefs. Journal of Personality and Social Psychology, 45, 961-977. Werth, J. L., & Lord, C. G. (1992). Previous conceptions of the typical group member and the contact hypothesis. Basic and Applied Social Psychology, 13, 351-369. Wicker, A. W. (1969). Attitude versus actions: The relationship of verbal and overt behavioral responses to attitude objects. Journal of Social Issues, 25, 41-78. Wilder, D. A. (1984). Intergroup contact: The typical member and the exception to the rule. Journal of Experimental Social Psychology, 20, 177-194.
ATTITUDE REPRESENTATION THEORY
343
Wilson, T. D., Dunn, D. S., Bybee, J. A., Hyman, D. B.. & Rotondo, J. A. (1984). Effects of analyzing reasons on attitude-behavior consistency. Journal of Personality and Social Psychology, 47, 5-16. Wilson, T. D., Dunn, D. S.. Kraft, D., & Lisle. D. J. (1989). Introspection, attitude change, and attitude-behavior consistency: The disruptive effects of explaining why we feel the way we do. In L. Berkowitz (Ed.), Advances in e.wperimental social psychology (Vol. 22, pp. 287-338). New York: Academic Press. Wilson, T. D.. & Hodges. S. D. (1992). Attitudes as temporary constructions. In L. L. Martin & A. Tesser (Eds.), The construction ofsocialjudgmenfs (pp. 37-65). Hillsdale, NJ: Erlbaum. Wilson, T. D., & Kraft, D. (1993). “Why do I love thee? Effects of repeated introspections about a dating relationship on attitudes toward the relationship. Personality and Social Psychology Bulletin, 19, 409-418. Winslow, R. W., Franzini, L. R., & Hwang. J. (1992). Perceived peer norms. casual sex, and AIDS risk reduction. Journal of Applied Social Psychology, 22, 1809-1827. Wyer, R. S., & Srull, T. K. (1986). Human cognition in its social context. Psychological Review. 93, 322-359. Yzerbyt, V. Y., Schadron, G., Leyens, J., & Rocher, S. (1994). Social judgeability: The impact of meta-informational cues on the use of stereotypes. Journal of Personality and Social Psychology, 66, 48-55. Zajonc, R. B. (1960). The process of cognitive tuning in communication. Journalof Abnormal arid Social Psychology, 61, 159-167. Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personaliry and Social Psychology [Monograph]. 9, 1-27. Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35, 151-175. Zajonc, R. B. (1998). Emotions. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook of social psychology, (4th ed., Vol. 1, pp. 591-632). New York: McGraw-Hill. Zanna. M. P., Kiesler, C. A.. & Pilkonis. P. A. (1970). Positive and negative attitudinal affect established by classical conditioning. Journal of Personality and Social Psychology, 14, 321-328. Zanna, M. P.. & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In D. BarTal & A. W. Kruglanski (Eds.), The social psychology of knowledge (pp. 315-334). Cambridge, England: Cambridge University Press. Zirate, M. A,, &Smith, E. R. (1990). Person categorization and stereotyping. Social Cognition, 8. 161-185.
This Page Intentionally Left Blank
DISCONTINUITY THEORY: COGNITIVE AND SOCIAL SEARCHES FOR RATIONALITY AND NORMALITY-MAY LEAD TO MADNESS Philip G. Zimbardo
I. Overview A theoretical framework is presented to understand the fundamental process of adapting to experiential discontinuities that are of personal significance to one’s self-concept. The central dynamic of this approach is the motivation triggered by perception of a discontinuity, which is defined as a violation of expectation in any domain of functioning highly valued by the self. Such discontinuities may arise from the experience of failures, deficits, or transformations in intellectual, social, sexual, sensory, physical, or career domains. The motivation to understand the cause of a discontinuity, and to appear rational to one’s self and others, generates a cognitive search process, while the motivation to appear normal, to be like others, generates a social search process. A behaviorally oriented search may be generated to take actions to suppress the arousal and deflect or ignore the distress associated with discontinuities. Although these three search processes are variants of what makes humans smart, connected, and functional, they may become biased in their operation in ways that distort their typically adaptive use. When that happens, various forms of social pathology and psychopathology may emerge in people without evident premorbid dispositions. These pathologies are referred to as aspects of “madness.” This broad term is used to cover a continuum from ordinary irrationality, fears, anxieties, obsessions, compulsions, addictive behaviors, prejudice, and mass hysteria, to more extreme forms of mental disorders such as phobia, paranoia, and depression. Direct links are posited between the kinds of search process biases activated in some people and the specific forms of pathology that appear as people try to make sense of, or cope with, their discontinuity. ADVANCES IN EXPERIMENTAL SOCIAL PSYCHOLOGY VOL 31
345
Copyright Q 1Y9Y hy Academi‘ Prerr All rights of reproduction in m y form rewved 0065 2601/99 63000
346
PHILIP G. ZIMBARDO
This chapter first details the assumptions and principles of Discontinuity Theory, elaborates on the search processes and their biases, and relates their rigid use to predictable symptom development in the first stages of the transformation of ordinary people becoming “mad.” A variety of case studies then illustrate many of the interesting features and wide-ranging implications of this approach in everyday life. Next, a series of experimental investigations are presented that were designed to test predictions about the operation of several kinds of discontinuities and associated search biases. The chapter concludes by considering the ethical issues raised by this research and outlines procedures used to reconcile them. The model I’ve developed blends constructs from cognitive and social psychology with those from clinical psychology, and its research foundation revives a dormant paradigm of experimental psychopathology (Kietzman, Sutton, & Zubin, 1975; Maser & Seligman, 1977). The reason for this integrated approach is that my starting point is the ordinary person faced with a serious challenge to his or her basic sense of self-identity and selfworth (Markus, Cross, & Wurf, 1990; Steele, 1988). That Actor resorts to a host of strategies and tactics to achieve the goal of re-establishing internal and external equanimity-actions that cut across any of our traditional disciplinary boundaries. A scientific understanding of them demands a holistic orientation based on a wide-ranging psychological foundation. A paradox that emerges from this approach is that while innovation and scientific curiosity are induced by discontinuities that lead to seeking new meanings and understandings, some personal searches triggered by discontinuities may follow a twisted path that ends in madness of various kinds. It will be instructive to compare the objective search of the empirical scientist guided by the canons of the scientific method with the egocentrically biased search of a naive theorist seeking to confirm hypotheses with selected evidence and distorted validity checks (Ross & Sicoly, 1979; Skov & Sherman, 1986). Before turning to the detailed exposition of this approach, let’s consider its relevance in making sense of a recent report from the U.S. Secret Service about the profiles of political assassins (Dedman, 1998). This new study in “Preventing Assassinations” concludes that the 83 people who attacked or tried to attack an American political figure or celebrity in the past 50 years fit no common physical or psychological profile (Fein & Vossekuil, 1997). Rather, the strongest pattern that emerged from an analysis of their recent experiences is that nearly all had suffered from a recent trauma, such as a marital breakup or job loss. While these “discontinuities” are experienced daily by many ordinary people (see Meyer, 1995), they led to a “downward spiral” in the lives of those who became assassins. The report noted that “relatively few suffered from serious mental illnesses that caused their
DISCONTINUITY THEORY
347
attack behaviors” (Dedman, 1998, p. A6). Instead of labeling such individuals as “irrational” or “crazy,” the Secret Service report argues that it is more productive and accurate “to examine a chain of thinking that leads a person to see assassination as an acceptable and necessary action, and to attend to behaviors that may precede an attack” (Dedman, 1998, p. A6). Let’s turn now to do just that, to examine chains of thinking and acting that may lead ordinary people to do extraordinary things in their lives as they grapple with significant personal discontinuities, such as loss of jobs, spouses, status, virility, and physical functioning, and of course, the divorce, illness. and death of loved ones.
11. Discontinuity Theory Man never reasons so much and becomes so introspective as when he suffers: since he is anxious to get at the cause of his sufferings. . . . When [one] is happy. he takes his happiness as it comes and doesn’t analyze it. just as if happiness were his right. -Luigi Pirandello, Six Characten in Search of an Author
Do you know what you have done? [Sherlock Holmes asked Sigmund Freud] You have succeeded in taking my methods-observation and inference-and applied them to the inside of a subject’s head. [Freud replies to Holnies]: Yet. what I have surmised might be totally erroneous, for you yourself have noted the dangers of reasoning with insufficient data at one’s disposal. -Meyer. The Seven Percent Solution
-Techniques need to be devised to discover and recognize the indelible imprints that discontinuous perturbations invariably leave behind. -Klauder. 1978
I am sure that you, like me, live your daily life relying on a set of deterministic operating assumptions of lawfulness, regularity, consistency, and continuity. It is curious that we do so despite functioning in a world filled with randomness, coincidence, chaos, and discontinuity. The human mind seems to have evolved to discover causes among correlated events, correlations among coincidentally arranged events, patterns amid flux, and structure where ambiguity resides (Chapman & Chapman, 1967,1969; Hamilton, Dugan, & Trolier, 1985). Noticing an effect, we infer its cause or we construct an attributional analysis that gives it causal significance (Jones & Davis, 1965; Hastie, 1985; Heider, 1958; Kelley, 1967,1971; Nisbett & Ross, 1980; Proudfoot & Shaver, 1975). Imposing such cognitive organization on
348
PHILIP G. ZIMBARDO
our experiences makes the new familiar, the confusing understandable, and structures the transient or scattered into neat mental categories. This process gives rise to a comforting sense of predictability and welcome to its companion, the illusion of control (Geer, Davison, & Gatchel, 1970; Langer, 1975). Yet the brain is also designed to attend to change, to detect novelty, to orient with surprise to sudden perturbations in our environment, in short, to dishabituate in a moment’s notice to anything that is new to us. Even before we are consciously aware of an anomalous stimulus event, the brain registers its presence by the P300 electroencephalogram (EEG) “surprise” wave pattern (Donchin, 1981; Donchin & Coles, 1988). All novel stimuli, change, discontinuities, and perturbations get fast-tracked to our attention center because they carry the potential for danger or opportunity, until proven otherwise, and then we can put them out of mind by silently engaging automatic habituation processes. There is obvious evolutionary significance to developing “discontinuity detectors” and to acting rapidly and wisely on the information they offer for aiding survival to environmental threats, adjusting to homoeostatic imbalances, and discovering unexpected pleasures. However, although this mind machine works wonderfully most of the time for most of its users, sometimes the wrong modules are engaged when dealing with input from its discontinuity detectors (Fodor, 1983; Ornstein, 1986). Then the mind generates premature conclusions based on minimal, readily available information, mistakes correlations for causation, or uses familiar, but inappropriate, heuristics for dealing with new experiences (Massad, Hubbard, & Newtson, 1979; Tversky & Kahneman, 1973). Discontinuity Theory is in part a formalized attempt to recognize and investigate some consequences of such curious aspects of the human experience and the fallibilities of our usually reliable and trustworthy minds (Ross, 1977, 1978). Starting with a few basic assumptions, the theory elaborates the processes by which perception of an anomalous experience triggers a motivational state that drives a host of reactions at various levels of functioning. After outlining the main features of this approach, I expand on the nature and operation of discontinuities, examine in detail the primary reactions triggered by some kinds of discontinuities, and conclude this section by considering how “madness” may become an unintended consequence, or by-product, of the very attempt by some people to appear rational and normal to themselves and others. A. BASIC ASSUMPTIONS AND PRINCIPLES The theory starts with some basic assumptions about the three fundamental human needs involved and their motivating effects. It also depicts the
DISCONTINUITY THEORY
349
nature and varieties of discontinuities that can impact on human behavior, and spells out the relationships between cognitive and social needs and processes of rationality and normality. After describing special features of reason-based choices for the explanations and misattributions generated to account for discontinuities, this section concludes by showing how “madness,” or initial forms of psychopathological symptoms, may emerge from distortions of the operation of the principles underlying rationality and sociability.
1. The Needs to Belong, to Know, and to Gain Esteem Two of the most basic needs of our species, beyond those of biological and safety concerns, are the cognitive need to know and the social need to belong. In Maslow’s (1970) hierarchy of needs they are sandwiched around esteem needs. Discontinuity theory assumes that some kinds of experienced discontinuities directly challenge personal esteem, and when they do, the needs to know and to belong are activated. The need for esteem is the individual’s generalized need for confidence, sense of personal worth and competence, self-esteem, and respect of others (Brown, 1998; Markus, 1977; Steele, 1988). The need to know, or to understand, is the desire to make sense of all human experience, to understand how the world functions, how stimuli are associated, how consequences emerge from prior events, and how the future can be predicted by knowing past and present circumstances. This cognitive need is the driving force in our constant, often automatic, processing of information bombarding us from within and without (Bargh, 1982; Cacioppo & Petty, 1982; Wegner, 1989). It is our dynamic travel guide that finds patterns in apparent randomness, extracts meaning from ambiguity, gives coherence to chaos, and discovers causal mechanisms that locate anomalous events and relationships within explanatory networks (Frankl, 1970). The need to belong is the basic human desire for attachment, connection, and affiliation to others who will recognize, accept, and love us (Baumeister & Leary, 1995). This bonding demands that people seek contact with others by going beyond the confines of self-focus to make and strengthen the human connection. It leads to behaving in ways that are socially approved by significant others in one’s life space, and for those actions to be evaluated as situationally appropriate. Much research has established that social isolation is a major risk factor in a range of pathologies, while having a sense of being enmeshed in a supportive social network contributes to one’s mental and physical well-being (Cohen & McKay, 1983; Gottleib, 1981;
350
PHILIP G . ZIMBARDO
House, Landis, & Umberson, 1988; Kulik & Mahler, 1989; Pilusuk & Parks, 1986; Taylor, 1986). In my model, the role of social isolation figures prominently in the development of delusions and other psychopathological symptoms. First, the isolated person may be unaware of the kinds of prevailing explanations that would be socially acceptable in many reference groups, and thus she or he could utilize them without prejudice. Second, once explanations are initially composed, there are not available social correctives on their reasonableness or logicality. Third, without a particular audience in mind when forming or refining explanations, they may become loosed from the moorings of social constraints and travel far from known shores of reality. Fourth, the absence of ready affiliations eliminates the option of seeking normalizing functions to cope with the threats posed by discontinuities when the rationalizing function is not working effectively. In addition to providing a sense of social security and enhancing selfworth, the need to belong provides the basis for social comparison (Festinger, 1954; Miller & Zimbardo, 1966; Schachter, 1959; Zimbardo & Formica, 1963). It is by means of social comparison processes that we assess our abilities, opinions, values, and emotions when objective standards are unavailable or vague. Less commonly recognized is the value of engaging social comparison processes as a prophylaxis against autistic thinking in which one’s personal conception of reality (including beliefs, wishes, and fantasies) can become the criterion for assessing the validity of external reality (Johnson & Raye, 1981). This is a reversal of the usual basis of reality testing in which validity checks and constraints on the appropriateness of one’s thoughts, feelings, and actions come from evidence in the external physical world and consensual validation from one’s social world. When that idiosyncratic thinking occurs chronically, it becomes the basis of the “psychotic pursuit of reality,” using internal reality as the foundation for assessing external reality (Meyer & Ekstein, 1970). Less extreme forms of this distorting, denying, or disregarding physical-social reality can be found under conditions of high ego involvement, sensory deprivation, daydreaming, taking poetic license for creativity, and other mental states that overtax or disengage ordinary consciousness (Kihlstrom, 1984). 2. The Motivating Effects of Discontinuities The grand thing about the human mind is that it can turn its own tables and see meaninglessness as its ultimate meaning. -John Cage, Silence
At the core of this model is the central role played by the perception of a discontinuity as a motivating trigger for informational, social, or behavioral
DISCONTINUITY THEORY
351
search processes. A discontinuity is here defined as a perceived change in the level or quality of one’s functioning in a domain that is important for one’s sense of self-worth or self-image. Such discontinuity involves an awareness of a noticeable deviation from an expected normative standard of how one usually feels, thinks, perceives, or acts-in those areas that figure into the calculation of one’s global sense of self. As that self-involvement increases, a threshold is reached when the discrepancy qualifies as a “significant” violation of an expectation. We know that expectancies represent beliefs about some future states of affairs, and “expectancies constitute the fundamental building blocks on which behavioral choices are built” (Olson, Roese, & Zanna, 1996, p. 211). The four dimensions on which expectancies vary, and thus their violations correspondingly vary in personal significance, are certainty, accessibility, explicitness, and importance (Olson et al., 1996). In the current theory, discontinuities assume special personal relevance when the expectancies they rest on are held with certainty and confidence, are accessible or readily activated, become explicitly generated, and have motivational importance through relevance to one’s needs and values as well as to their scope of connectivity to other expectations. A violation of expectation with those attributes is obviously a discontinuity that holds special significance for the integrity of the self-construct. Motivations arise to reduce the confusion, threat, and distress that such discontinuities engender. It is assumed that negative affect is aroused as a correlated consequence of the cognitive disruption produced by discontinuities. Attempts are made to reestablish harmony, intrapsychic and interpersonal, as well as to restore any balance between self and nature or self and spiritual forces, which existed prior to the anomalous experience. Research has found that “expectancies can play an important role in triggering individuals to undertake causal analyses” (Pysczynski & Greenberg, 1981, p. 36). In addition, more thorough attributional search processes are activated for unexpected than expected events. Expectancy violations can generate a range of reactions from humor to horror, depending on context, medium of exposure, ease of clarification, and other situational and dispositional variables (Deckers & Devine, 1981: Lefcourt & Martin, 1986: Rothbart, 1976; Suls, 1972,1983). Expectancies also color our interpretations of social issues, our framing of social problems, and interpersonal relationships (Burgoon, 1993; Jussim, 1990a,b). Violations of expectations come in many forms, each of which typically elicits characteristic affective reactions. Consider the following array of some V.O.E. types. 1. Misfortune: Distress at unexpectedly negative outcome, bad luck. 2. Good fortune: Joy at unexpectedly positive outcome, good luck.
352
PHILIP G. ZIMBARDO
3. Magic: Amazement at magician’s “trick” that defies beliefs about physical reality or human limitations. 4. Miracle: Awe (fear, reverence) at unexpected positive outcome that has no known material, physical, natural causal explanation, thus assumed to involve spiritual intervention. 5. Humor: Laughter at, enjoyment of, joke’s punch line or visual comedy routine that does not follow from initial premises; when its puzzle is resolved, we laugh, when not resolved, we are confused, “don’t get the joke.” 6. Horror: Fear revulsion, terror at sudden negative, violent, destructive outcome, which violates context expectations created in plot, or in one’s experiences, or from traditional “standards of decency.” 7. Natural disasters, cosmicperturbations: Fear, awe, confusion by experiencing earthquakes, eclipses, volcanic eruptions, and other such calamitous events when they are unpredicted. 8. Social Deviance: Rejection of, anger at, a person or group for violating social norms of situationally appropriate, relevant behavior. 9. Aesthetic Value Violation: Shock, anger, resentment, disbelief, when art challenges established way of representing, viewing, reality (such as the early labeling of Impressionist painters as crazy “Fauvists”). Informational search activities may then be initiated to understand what is happening, what is being experienced, how to explain and label the discontinuity, and to decide what personally relevant meaning it may convey. Information-gathering strategies and hypothesis-testing processes are cognitively and socially complicated since they may seek confirmation or disconformation, be influenced by action agendas, limited by cognitive skills and access to evidence (see Klayman & Ha, 1987; Skov & Sherman, 1986; Snyder, Campbell & Preston, 1982; Trope & Bassok, 1982, 1983). A memory search seeks historical comparisons, analogies, correlated similar occurrences of negative reactions and prior unusual causes, along with other reasoned analyses. There is an emotional asymmetry between positive and negative affect (Taylor, 1991). Negative outcomes elicit more extensive analytic reasoning than do positive ones (Schwarz, 1990), and are more likely to promote attributional searches (Weiner, 1985). Because negative emotions often arise under circumstances that threaten well-being and thus require immediate coping (Ellsworth & Smith, 1988;Taylor, 1991), they demand more precise identification of the cause of the current affective state. Faced with the negativity of a discontinuity, the person should engage a causal search focused on single, isolated causes for which a coherent remedial action can be taken, rather than the more diffuse, generalized search typical in identifying causes of positive emotions (Liu, Karasawa, & Weiner, 1992). We also know that having a negative affective expectation
DISCONTINUITY THEORY
353
will help to create correspondingly negative affective experiences (Wilson, Lisle, Kraft, & Wetzel, 1989). This cognitive search may be complemented by a concurrent situational analysis that checks out the social and physical environment, along with an assessment of the condition of one’s body/ healthlgeneral status for causal clues. However, when the discontinuity is perceived as having positive, favorable effects (such as an unexpected promotion or prize), there is not the same urgent need for understanding or action taking. Indeed, for many people thinking analytically about one’s good fortune seems to dilute its pleasure. So, as Pirandello (1923) reminded us in the opening quote, we accept such delightful anomalies as due us, to be accepted at face value without question. That view of the passive processing of positive discontinuities is supported also by John Stuart Mill (1873) who says: “Ask yourself whether you are happy, and you cease to be so.” Can discontinuities that bring sudden considerable benefits, such as winning millions of dollars by chance, or unexpected fame from one’s actions, ever have negative consequences and symptomatic outcomes? Although they are less likely to do so than those that threaten loss of control over important spheres of functioning, “positive discontinuities” may have similar negative consequences depending on how the discontinuity is interpreted. I am aware of one such case of a young journalist who happened to be in his home neighborhood in Watts, California, when a dramatic racial uprising erupted in 1965. Since he was African American, he was able to cover the story at a time when white reporters were banned from the area by the residents. Bob Richardson, though not actually a reporter, was given permission to send in daily reports to his boss at The Los Angeles Times. The stories won the prestigious Pulitzer prize for investigative reporting shared by The Times and Richardson, and he was awarded reporter statusthe first Black reporter on a major urban newspaper. From there it all went precipitously downhill for the neophyte reporter. He did not feel he deserved the acclaim, was lucky to happen to be on the scene at the right time, and was unable to follow up that prize-winning journalism with other important stories. He did not even attend the Pulitzer Prize ceremony, began drinking, taking drugs, missing work, not meeting deadlines, got fired. lived as a homeless person on the streets for years, hungry and penniless. His story became known only after a young screenwriter, Michael Lazarou, writing a script for a movie about the Watts riots, tracked Richardson down to invite him to be his consultant. Being able to relive his experiences at a temporal distance from them, and to share his sense of not deserving what he had earned, helped Richardson to get back on track and come up from the gutter to a more productive, if low-key, lifestyle. This moving story about the downside of a fame-based discontinuity was
354
PHILIP G. ZIMBARDO
presented in a special television show on ABC-TV’s 20/20 program (August 3, 1990). The perception of a discontinuity is usually at a level of conscious awareness, although it can also occur at preconscious or unconscious levels of awareness (Kihlstrom, 1985; Marcel, 1983). We know that actors may behave in predictable ways, yet not have access t o their underlying mental processes, for any number of reasons (Nisbett & Wilson, 1977). Some people may respond with strong, immediate emotional reactions before they know why they are “acting without thinking.” LeDoux (1989, 1995) speculates that they may be overly emotional because the direct response of their amygdala is stronger than the cortex’s slower ability to constrain it with rational interpretations. He argues that this same process may trigger “mindless acts of aggression.” The amygdala has recently been shown (on PET scans) to respond more strongly to externally stimulated emotional stimuli than to memory-generated emotions (Reiman et al., 1997). Thus, we may become aroused without knowing why for many reasons. Some of that unexplained arousal may become discontinuities of note. Freud’s (1946) repression model is consistent with aspects of Discontinuity Theory. It contains two sources of discontinuity and a uniquely psychodynamic way of coping with each of them. The initial discontinuity resides in the conflict between feeling strong biologically driven (libidinal) impulses and recognizing socially learned constraints on their expression (superego). One’s positive self-regard is threatened by experiencing these sexual and aggressive impulses, and that is the primary discontinuity. Repressing that conflict removes it from consciousness and obviates attempts to deal with it directly. The second source of discontinuity occurs later on when the person confronts a seeming innocuous situational stimulus that directly or symbolically is associated with the potential satisfaction or expression of that libidinal impulse. A state of anxiety is activated, as a signal system warning that a repressed conflict is about to emerge into consciousness. The strong autonomic response with diffuse bodily effects that occurs is clearly a negative affective state. The dominant response to that unexplained arousal, that we label as anxiety, is to socially isolate one’s self from others since the reaction seems situationally inappropriate (see Bromberg, 1968; Sarnoff, 1971). By contrast, the arousal of high levels of fear in response to an objective, external source of anticipated bodily threat activates needs for affiliation (Sarnoff & Zimbardo, 1961). In the Freudian model, this anxiety is coped with by activating any of a host of defense mechanisms that diffuse, dampen, or misdirect its impact. This second source of discontinuity is really between cognitive and affective systems that, up to this point, have been isolated by repression mechanisms. The anxious person is experiencing sudden, strong, negative affect without
DISCONTINUITY THEORY
355
cognitions about sources justifying such heightened arousal. Pathology results in this psychodynamic model only when anxiety becomes so chronic that defense mechanisms are overused or improperly developed to start with, and the individual’s coping system is overextended. a. Functions of Discontinuity Explanations. Why is it important to seek to explain discontinuities rather than to conceal them by repressing o r suppressing them? This is where the discontinuity model differs from the Freudian account in emphasizing the reason-based nature of a search for explanation rather than a reason-repressing process. Consider the following set of functions served by discontinuity explanations; it is these functions that drive the search mechanism: 1. To establish the origin of a discontinuity as either internal, in one’s mind or body, or external, “out there” 2. To establish one’s relationship to the event as either accidental, coincidental, or causally implicated 3. To reduce the mystery created by a discontinuity, by locating it in a familiar scenario with a known label 4. To minimize the potential threat discontinuity poses by understanding its origins and consequences 5. To determine whether discontinuity is transient or likely to recur 6. To distinguish between consensually validated experiences, shared by others, and idiosyncratic ones, unique to one’s self, as in the difference between illusions and hallucinations 7. To discover how to prevent, avoid, escape, minimize, or overcome the discontinuity’s impact 8. To activate self-directed interventions to cope with the anxiety the discontinuity generates 9. To respond to the problem-solving challenge that all discontinuities pose These activities overlap with the basic goals of the scientist and the clinical practitioner: to describe, explain, predict, and control behavior, as well as to take therapeutic actions to improve functioning. The key difference lies in the greater efforts scientists invest in maintaining an objective detachment, in using the scientific method as protection against uncontrolled observations, biased search for relevant data, and also premature and inflexible conclusion drawing. The ordinary person in the throes of dealing with a subjectively significant discontinuity does not have the training, the luxury of reflection time, nor the detached perspective of the scientist. Instead, she or he begins as a biased observer, selectively attending to available and especially salient evidence, making sampling errors, integrating various data sources without
356
PHILIP G . ZIMBARDO
adequately weighting the multiple factors involved (Nisbett & Ross, 1980; Ross, 1977; Ross & Sicoly, 1979). The search for explanation then comes to rely on stored schema, readily available heuristics, analogies, folk wisdom, and currently prominent “pop” psychology views, all mixed with limited and faulty reasoning strategies to generate one’s personal theory of this discontinuity (see Malle, in press). Under such circumstances, a biased theorist is born, eager to validate that theory by searching only for confirmatory data-likely of course, to be found, theory affirmed, search terminated, and true cause uncovered (sometimes) or not (perhaps more often). Regardless of the actual truth value in the outcome of this discovery process, the beliefs it generates will persevere in the mind of the theorist (Anderson, 1983; Anderson, Lepper, & Ross, 1980; Fleming & Arrowood, 1978; Jennings, Lepper, & Ross, 1981; Lord, Ross & Lepper, 1979; Ross & Lepper, 1980). When that perseverance becomes inflexible and rigidly defended despite contrary evidence, prejudice and paranoia may emerge as but two of the forms of resulting pathological thinking. Unlike Schachter and Singer’s (1962) assumption that the search to understand discontinuities in the form of “unexplained arousal” begins as a general, neutral scanning process, Discontinuity Theory assumes that for adults the experience of sudden arousal with an insufficient immediately available explanation triggers a biased search process. This assumption is predicated on the notion that such arousal signals a loss of personal control, which is a negative state of self-awareness (see Marshall & Zimbardo, 1979; Maslach, 1979). Events perceived as unpredictable and uncontrollable are threatening and anxiety arousing, and demand explanation when they are schema-incongruent (Hastie, 1985; Mineka & Kihlstrom, 1978). A later section will offer experimental data that support this assumption. b. Locating the Discontinuity withour or within. An early phase in the search for a discontinuity explanation involves locating its potential cause either in (a) some aspect of external reality-and thus perceivable, or perhaps experienced, by others in that situation; or (b) in the internal reality of one’s mental or bodily condition-and thus a private, idiosyncratic cause. Social comparison search and/or interpersonal disclosure helps to determine whether others (especially appropriate companion targets) in that situation are responding similarly to one’s self, and also differently from their usual prior state. If they are, the likely cause is having a general effect and a common search for situationally based causes can be instituted, perhaps with help from those others. However, if they are not, then the likely cause is not socially shared, and the effect may be limited, specific, and unique to one’s dispositions or state. That realization poses new concerns for our Actor since it marks the self as currently “different,” or deviant from comparable others. The discontinuity may then be interpreted as carrying
DISCONTINUITY THEORY
357
diagnostic information about something going wrong, or the start of worse things to follow. Under such circumstances, the person experiencing this profound discontinuity may choose social isolation at least temporarily, to figure out what is going on, without the intrusion and concern for others noticing his or her arousal or differentness. Disclosure, which ordinarily might help in the mutual discovery of an external cause, or get one a sympathetic hearing about an internal cause, will be avoided because to now disclose aberrant personal feelings will only heighten the interpersonal contrasts. Several experiments on disclosure of unexplained arousal will be presented that sought to examine disclosure processes among those who were in the midst of experiencing a powerful discontinuity. Discontinuities are most likely to be noticed if they are sudden, intense, disruptive of ongoing behavior, situationally inappropriate, or especially unusual for the person in a particular domain. These observations lead the Actor to make an inductive leap to a dispositional inference, one that goes beyond the observed evidence to add the baggage of surplus meaning. The behavioral state characterization, “I am acting irrationally, or anxiously, or nervously,” becomes a state description, “I am irrational, anxious, or nervous.” This progression may end up as a trait diagnosis, “I am an anxious person,” “I am a boring speaker,” “I am an incompetent student.” This attributional sequence moves the analysis of the discontinuity away from any original instigating (perhaps external) conditions toward focusing on the self as causal agent. When that happens, the likelihood of coming up with the true cause of the original discontinuity is seriously constrained. That happens because the Actor’s thinking suffers a loss of specificity in denoting any situational correlates involved, the time parameters of onset or variation of response, the response channels involved, recall of prior comparable reactions, or assessment of whether others are reacting similarly. It becomes quite reasonable for our Actor then to conclude that the problem is: “something is wrong with Me.” In addition to the original, or prime, discontinuity, there are also correlated, secondary discontinuities in physiological reactions, cognitive, emotional, behavioral, and social responding. The Actor may react with changes in one or more physiological channels, such as heart rate increase, respiration changes, tremor, muscle tension, or EEG signs of unusual brain wave activity. Or there may be evidence of distractability, distress, or nervousness which is apparent to others who want to know what is wrong. Usual social relationships may become strained as the person focuses inward, avoids self-exposure, and becomes overly sensitive to any social source of further arousal, such as criticism, “prying,” or even expectations that life should go on as if nothing had happened. The matter also gets complicated once others are involved since they will also generate explanations for the discon-
358
PHILIP G. ZIMBARDO
tinuity in behavior they are witnessing. In addition, where that behavior is extreme and atypical for the person, such as dramatic mood shifts, memory loss, or overt anger, these others will be motivated to keep up appearances by concealing it from outsiders. This was sadly true among family members of men who were unknowingly suffering from neurotoxic damage from workplace chemicals, as will be described in one of our case studies. A Machiavellian strategy might also be utilized by those without an adequate causal explanation for their experienced discontinuity. Whether enacted consciously or not, they could attempt to arouse strong, unusual reactions in others so that they are rendered more emotionally comparable. Tactically this could be accomplished in many ways: by spreading warnings about the dangers of violent crime in our midst, environmental pollution, the thinning ozone layer, conspiracies of all kinds, not to mention the imminent abduction by beings from UFOs, and the end of the world as the Millennium approaches (see Jacobs, 1993; Lattin, 1988; Mack, 1995). A word about the prevalence of conspiracy beliefs in the general population. In a field study of reactions to Oliver Stone’s provocative film, JFK, we interviewed 53 people on their way into the movie and 54 on their way out, from a middle-class, educated audience (Butler, Koopman, & Zimbardo, 1995). One of our questions asked about the likely conspiracies in the film that might have led to Kennedy’s assassination, whereas others asked about likely conspiracies in the lives of the respondents. Remarkably, 64% reported a history of one of more conspiracies in their personal lives, with the most, 42%, reporting conspiracies against them at work (50% before viewing, 35% after viewing the film). Such results add to my belief that there is much “soft-side’’ madness among otherwise normal people, whose beliefs and symptoms are not extreme enough nor dysfunctional enough to be pathological, but nevertheless, they are on the negative side of the normal-pathological continuum. There is less harm to be suffered in being mad among madmen than in being sane all by oneself. -Denis Diderot, Supplement to Bougainville’s Voyage
The cognitive and social searches may blend when the individual comes up with an explanation that seems reasonable, but is not convincing to others among her or his usual reference groups. In this case, those skeptics will be abandoned and replaced by new alliances who are more willing to accept one’s discontinuity explanation. One form of the social search seeks those who are more easily persuaded to accept, or at least tolerate, the “working discontinuity explanation” than can be expected from one’s family, friends, or co-workers. This search for approving others who will believe
DISCONTINUITY THEORY
359
the unusual discontinuity explanation may be supplemented with a second type of social search for normality by submerging one’s self in a subculture that is judged as “deviant” on some societal dimension. And to go a big step further, with sufficient social skill, personal power, or charisma, that person can create a group that endorses the discontinuity explanation as an article of faith, while also willingly acting in ways that this new group leader presents as appropriate and central to the group’s identity. This has happened in cases of cult leaders imposing their bizarre beliefs and action rituals on followers under the guise of ordinary operating procedures for that group. One of our case studies will document a recent instance of a cult leader directing his male members to castrate themselves, at a time when he was grappling with problems of strong, unacceptable homosexual feelings-his discontinuity, in my sense of the term (Miller, 1997). c. Culturally Constructed Discontinuities. Our focus thus far has been how an individual Actor attempts to cope with the challenges posed by perceiving a personally significant discontinuity. That level of analysis is appropriate in individualistic cultures, such as in the United States. However, it is possible to take the person off the hook of the “self-is-solelyresponsible” for generating acceptable discontinuity explanations. In some collectivist cultures (Marsella, 1979; Triandis, 1994), discontinuities are acknowledged as a source of potential danger to the harmonious functioning of the entire society and not just a personal problem for the individual to cope with in the solitary fashion I have outlined above. Such cultures may develop, and hand down across generations, discontinuity scenarios likely to be experienced along with societal remedies and even preventive strategies. As soon as the first signs of an individual’s discontinuity are manifested publically, societal agents take over and intervene with socially sanctioned procedures to restore the harmony disrupted by that discontinuity. Communal treatment rituals always involve the Actor along with all significant others among family and tribal members. Two examples of this social construction of discontinuity and its treatment come from the Navajo in the United States, and the Nyakusa of Tanzania, Africa. For the Navajo, well-being comes from social harmony as well as harmony between people and nature. When this balanced, ideal state of “hozho” is disrupted, illness is seen as an outcome of the resulting disharmony. The disharmony is a form of evil introduced through taboo violations, witchcraft, overindulgence, or bad dreams. The individual who experiences or reports a disharmony (a discontinuity in our terms) is neither personally responsible, nor at fault for it. Rather, the illness, or evil, is taken as a sign of broader disharmony that must be dealt with at the tribal level with communal healing ceremonies that repair that disharmony, with the everpresent goal of restoring hozho.
360
PHILIP G. ZIMBARDO
Similarly, among the Nyakusa, at the first sign of deviation from the expected norms of harmonious living, communal interventions are swiftly enacted to set the situation right. In this culture, everyone is aware of signs of discontinuities that are violations of natural laws, such as multiple births (since they are common to animals, but rare in humans), sudden death of young people, expressions of strong anger, or any form of illness. The rituals begin with social acceptance of the afflicted person(s) and a search to discover the source of the discontinuity, as it holds the potential for afflicting others and so must be counteracted immediately. Appropriate medicine is administered not only to effect biological change, but to modify the habits, dispositions, and desires of people (namely psychological cures). There is special medicine for anger in husbands, employers, and police, and other remedies to cure thieves of criminal habits. Still other tribal medicines make lonely men and women more attractive, and make those who would be leaders more persuasive. Remarkably, and perhaps as a consequence of these societal explanations and interventions, there was no evidence in this tribe of any behavior that could be classified as “pathological” over the dozen years that anthropologist Monica Wilson (1951, 1962, 1970), studied them extensively (personal communication, September 16, 1971). These examples frame in stark contrast the prevailing situation in most individualist cultures where discontinuities are experienced as personal events that must be coped with and treated at an individual level. Another current instance of utilizing a prevailing belief system to account for a widespread, shared societal discontinuity is the rise of anti-Semitism in Russia (Reynolds, 1998). Loudly and crudely blaming the country’s woes on Jews follows a “time-dishonored’’ tradition. “Anti-Semitism is as perennial in Russia as the snow. It tends to arrive in force during seasons of economic discontent, and it lurks beneath the surface the rest of the time as stubbornly as permafrost,” according to investigative reporter Maura Reynolds (p. A20). Since the Soviet collapse as a world power, the people have been suddenly thrust back into conditions more common in underdeveloped nations, of severe deprivation, crime, lawlessness, and chaos. “As a result, Russian nationalism and its anti-Semitic corollary have become a kind of default ideology” (p. A20).
3. On the Nature and Varieties of Discontinuities Discontinuities come in many shapes and forms. They may be transient events, dramatic first time events, cyclical, or cumulative over some time period. I’ll examine how these various forms of violation of expectation influence human behavior patterns.
DISCONTINUITY THEORY
361
a. Expectation Violations in Daily Functioning. The most commonly experienced discontinuities are those that violate one’s expectations about domains related to one’s daily functioning. These expectations are based on activities and observations that are routinely made about aspects of the self in relationship to others, to specific tasks, to desired goals, and to a sense of well-being and self-worth. You discover your parent does not love you, you are rejected by those you like and thought liked you, you are shown no respect when you deserve it, you fail when you should have passed or excelled. In other words, you do not perform intellectually, physically, socially, sexually, or athletically as you have come to believe (and take for granted) that you should. If the domain in which this violation of expectation occurs is one that is significant for the individual’s sense of self-worth, then it demands attention and some remedial action to cope with the challenge it poses to self-image maintenance. This type of discontinuity is the most basic foundation on which this conceptual model is constructed. 6. Dramatic First-Time Experiences. Another form that discontinuities take are first-time events in one’s life that are dramatic in their sensory, physical, or psychological impact. Consider someone’s first headache or nightmare, the death of a loved one, the sudden divorce of parents, a young girl’s first menstrual period or a boy’s first nocturnal emission, and a child’s first abuse by a parent. Assume further that neither family, friends, nor society provided preparatory information that could mitigate the confusing impact that such powerful events may have on the individual. How then would one understand the nature of these vivid discontinuities? What inferences might be made about the structure of one’s social, biological, and physical worlds? It is not unreasonable to suppose that under such dramatic and confusion-eliciting conditions, individuals might generate explanations that are mystical or spiritual, or not bound by conventional rules of reasoning. In addition, we might expect that people would also make “arrangements” with culturally dictated higher forces. They ask to be protected, to prevent recurrences, or to help conceal the fearful experiences from others. Their part of the bargain may involve promises to be good, asexual, penitent, to enact other forms of submission, or even to leave the payback open to the later demands of that higher authority. That deal will always succeed-if the event is transient and the discontinuity does not appear again with any regularity. Then even after gaining intellectual understanding about the nature of the discontinuity later on, the individual may feel secretly or nonconsciously bound for life to pay back on that higher-order contract. That arrangement could lead to a variety of dysfunctional beliefs and ritual actions. c. Cumulative Build-up of Discontinuities over Time. Other discontinuities are not noticed until their impact gradually accumulates over time to
362
PHILIP G. ZIMBARDO
create profound functional deficit or excess. This occurs, for example, with exposure over time to work situations where low levels of neurotoxins are gradually but steadily released from work materials, such as lead paints or plastics. In some instances, OSHA standards are not based on tests of longterm exposure to such chemical pollutants, thus workers, company, and union officials are unaware of this toxic consequence of their work setting. We shall see in one of our case studies how this kind of discontinuity may generate tragic outcomes, such as memory loss, dysphoric mood swings, and violent behavior-which are misattributed to psychological disorders and mistreated as such. Incipient deafness may also result in discontinuous experiences. Some forms of deafness (conductive deafness) are gradual and continuous in their progression. Consequently, they are often not noticed for a long time, if ever. The person may not realize that deafness is the problem because the auditory mechanisms have deteriorated gradually, and because he or she may have unknowingly started to use lip reading and bodily gestures to fill in what was not heard. In addition, the stigma of deafness may lead to denial among some people even after the first signs of hearing loss seem apparent. This denial may motivate misattributions to psychological causes instead of acknowledgment of their organic origin. This happened in the case of a college dean who hated to go to cocktail parties even though they were an essential part of the public relations function of his job. The background noise at such parties made it difficult to detect speech patterns, and when someone asked a question about a recent book, review, or article, he was forced to ask them to repeat the question. Believing they thought he was unsure of that information, which any scholar of his stature should know, he found such events increasingly aversive. He was about to quit his post until his wife forced him to acknowledge that his primary problem was his organic deafness (that he was denying because it symbolized old age). His was a condition that required a hearing aid, not a new job. d. Transient, Rare Phenomena That Are Shared Collectively. The most apparent discontinuities come from natural physical phenomena that are rare, irregular, and dramatic in their impact on individuals and collectives, such as comets, eclipses, hurricanes, tornadoes, earthquakes, volcanic eruptions, lightning, famine, and plague. While science now informs most societies about the dynamic properties of the physical nature and natural causes of such discontinuities, in some less scientifically informed cultures, these cosmic discontinuities assume personal relevance, demanding immediate explanation and remedial action. Some years ago it was reported (Reuters, in S.F. Chronicle, 11/23/73) that the Brazilian government was trying to give advance warning to its
DISCONTINUITY THEORY
363
Indian tribes of the impending arrival of the comet Kohoutek in order to prevent “a revival of old myths and cults” and to quell their distress. Their concerns were based on the Indians’ reactions to an earlier eclipse of the sun. In one tribe the chief demanded his members engage in self-imposed reproductive restrictions, as a pledge made in return for the normalization of the elements. Another tribe, surprised by the eclipse of the sun during their lunchtime, believed they were being punished by the gods for eating too much. They induced vomiting and the tribe began a regularly scheduled fasting ritual. The stuff of myths also comes from this same report, since the warriors in one tribe gathered in the center of the village and sent salvos of flaming arrows toward the darkened sun to light it up again. When the sun reappeared, “the warriors were hailed as heroes for having lit the sun with their arrows.” We may assume that future generations will celebrate their wisdom and bravery with a special festive ritual. There are parallels between this example and the phenomenon of “Cargo Cults.” This fascinating social movement fits the development of superstitious behavior practiced by an entire society as a collective behavioral ritual to correct a socially shared discontinuity (see Lawrence, 1964; Strelan, 1977; Worsley, 1968). Across vast distances in the Pacific islands of Micronesia, different groups practice similar rituals designed to bring back ships that once appeared filled with bounty and then suddenly vanished-a dramatic discontinuity to be explained and corrected. In earlier times, these ships were sailing vessels exploring the area for treasures; later they were Japanese and then American war ships that occupied and then abandoned these islands during World War 11. A common assumption in all these groups is that the ships were originally sent by their ancestors who filled them with goods for their tribe. But evil men stole the ships and the bounty that really belonged to them. To appease the gods and ancestors, a leader will assert that the cargo will return only if certain tribal actions are taken collectively. These may include destroying crops, burning down the village, migrating to another area, or making any dramatic change of lifestyle that can be taken as a sign of their devotion and sacrifice. Of course, when that action fails to make the cargoes materialize, the leader is deposed or killed, and a new leader with a new solution takes over. What is amazing about this superstitious collective action is that it is so widely practiced and has been going on for decades. The relevance for our theory is that the discontinuity is socially shared, an ideology constructed as an integrated set of explanations to account for an anomaly, an action sequence put into practice, and when it fails to correct the situation, the entire process is recycled. As with the celebrated case of a failed prophecy of a doomsday cult studied by Festinger, Riecken, and Schachter (1956), the greater the members’ commitment to the group’s
364
PHILIP G. ZIMBARDO
beliefs, the more postdecisional dissonance will arise when the leader or the group’s prophesy fails. The dissonance will then energize new actions to right the wrong. However, in this instance, it is remarkable that after so many replications of failed predictions, the Cargo Cults continue to operate at full strength fueled by the perception of a single discontinuity so long ago. With education, we tend not t o believe that natural disasters are causally related to the self. However, such “primitive” thinking is typical of all of us during the early childhood preoperational period. Piaget (1954, 1977) describes this stage in cognitive development as one in which children engage in egocentric thought that positions them at the center of all phenomena. However, under conditions of stress, or when personally experiencing repeated occasions of natural disasters, adults may revert to such “childish” thought patterns. Just as Job asked his god, “Why me?,” we may do likewise when faced with personal tragedy or trauma-of rape, death of our child, ruined crops, or failed ambition, to mention but a few instances. When that question is posed in a subjective form, the answer must be based on some personal qualities of the Actor, or on the operation of mystical forces conceived of as being directed against the individual-and not based on laws of nature or other external considerations. e. Categories of Discontinuities. Discontinuities can be categorized along the following dimensions: 1. Relevance to sense of survival, of self-worth: very significant to relatively insignificant 2. Duration: transient and brief to chronic and permanent 3. Scope: discrete and specific to general and pervasive 4. Onset: sudden to gradual development 5. Source: external (climactic, celestial, environmental, social-situational) to internal (biological, physiological, emotional, mental, behavioral) 6. Occurrence: rare, random, irregular; to cyclical; to recurring and common 7. Perceived hedonic valence: positive to negative 8. Degree of expectation: anticipated to some degree as a feature of a transitional period (into adolescence, for example) to totally unexpected or anomalous occurrence 9. Experiential basis: personally experienced to vicariously experienced 10. Societal preparation: none to much 11. Origins: symbolic or unconscious origins to concrete or consciously acknowledged origins
f Relating Memory Processes to Discontinuity Processes. It is also possible to relate the construct of discontinuity to three basic types of memory processes: semantic, episodic, and procedural (Tulving, 1972,1983). Seman-
DISCONTINUITY THEORY
365
tic discontinuity involves an experience in which one’s general knowledge about the world is violated when physical or social “facts” do not occur as expected from previously learned contexts. This schematic violation disrupts routine information processing, thus forcing a heightened awareness of the discrepancy. “What is wrong here?” “What doesn’t fit as it should?” are the questions that come readily to mind-to the scientist, the inventor, or the simply curious. Episodic discontinuity occurs when a semantic discontinuity becomes self-referential, seen as personally relevant, or containing information that might have predictive value for the individual’s well-being. The abstract nature of the observed informational discrepancy assumes personal significance when the individual poses the questions, Why is this happening to ME? or “What does this mean for MY survival, well-being, self-image? Finally, procedural discontinuity is the awareness that one does not know or recall the rules of logic or of inferential thinking which are essential for discovering the basis of these semantic or episodic discontinuities. The person believes he or she ought to be able to explain the discontinuity, but can’t figure out how to do it effectively. Efforts fail to clarify the situation or correct the disruption in functioning. This may occur because of extreme stress, fatigue, sensory or discontinuity overload, or pressing situational or time demands to generate an immediate and satisfactory explanation to self or others. Of course, another reason that some people are not able to figure out the nature of a given discontinuity is a basic lack of cognitive capacity to engage in such conceptual analyses.
B. KNOWING IS TO RATIONALITY AS BELONGING IS T O NORMALITY The mind is a strange machine which can combine the materials offered to it in the most astonishing ways. -Bertrand Russell. The Conquest of Happiness
In this section we will outline three search processes that individuals employ when coping with significant personal discontinuities, the cognitive and social searches and then the behavioral search. 1. The Cognitive Search Process
The desire to know-to understand the world around us and the inner workings of our minds-is a fundamental need of homo sapiens. Some would argue that the search for meaning is a powerful human drive (Frankl, 1970). Typically, it is realized through thought processes grounded in ratio-
366
PHILIP G. ZIMBARDO
nality. Under most conditions, intelligent and educated people make use of logical reasoning, rules of inference, heuristics, analogies, accepted canons of evidence, and tentative conclusion drawing based on observations that are open to correction by better data. We learn to see and represent the world conceptually according to the way “it really is” by not always accepting the phenomenal reality of perceptual experience. This change from perceptually driven views of the world to conceptually driven ones is a developmental landmark that gets elaborated as children develop “foundational theories”-frameworks for initial understanding-to explain their experiences of the world (Carey, 1985; Wellman & Gelman, 1992). Rationality differentiates analytical thinking, “I think, therefore I am,” from solipsistic thinking, “I think of it, therefore it is.” Formal education transforms magical, animistic, solipsistic thinking into reasoned analysis (cf. Bruner & Goodman, 1947). We use evidence to substantiate our beliefs, and other people to consensually validate and approve our conclusions. Moreover, we are motivated to seek reasonable causes t o account for experienced effects, and to generate causal explanations that are both analytically “correct,” and socially “right” rationalizations (Aronson, 1968). Because of the centrality of this need to know and understand, it becomes vital to our sense of self-competence that we appear to be rational to ourselves and to significant others. We put ourselves at risk for being perceived as not rational, as “ir-rational,” when we are unable to generate adequate explanations for discontinuities, or we fail to discover causes that account for powerful effects currently being experienced. As one moves closer to recognizing or experiencing that aversive state of irrationality, even greater mental energy is directed toward the internal search mission for rationality. Paradoxically, there comes a point when cognitive-emotional resources become overtaxed and the search becomes truncated, ending with an irrational conclusion that is rigidly defended. Resources become limited when the individual is dealing simultaneously with the motivations aroused by the initial discontinuity, the looming failure of the need to understand the whys of this experience, and perhaps exhaustion of one’s usual modes of rational analysis. A t the same time, there may be increasing concerns about what others will think of this peculiar state of affairs. Consider, for example, the case of an assistant professor whose whole life was entirely centered around teaching and research, but who was then told by colleagues that his chances for tenure were not good. It was suggested to him that a face-saving option to the potential failure of his tenure evaluation, and its associated stress, would be simply to decline to come up for that problematic evaluation. But by taking that option, the candidate rejected himself. Doing so denied him the possibility of expressing his rage
DISCONTINUITY THEORY
367
outward toward an unjust system or unsupportive colleagues. Who could he then blame for this discontinuity? At some level, only himself. This selfblame redirects the negative affect inward, and adds a secondary emotional load to attempts to make sense of the situation and take constructive action. (In this real-life example, such a scenario contributed to the sudden onset of this young man’s serious mental disorder.) There comes a point in the cognitive search process when a systematic, rational search is transformed into accepting any explanation which seems minimally plausible, justified by any spurious, perceptually salient, or cognitively primed data. Once such an explanation is put in place, then defenses are erected against considering alternative hypotheses. Now the Actor has switched from being a tentative hypothesis tester to a dedicated advocate for this new “theory.” Self-knowledge is a dangerous thing. tending to make man shallow or insane. -Karl Shapiro, The Bourgeois Poet
Under most circumstances the explanation advanced is accepted, or at least tolerated, by others if it fits their general understanding of the person and the situation. But sometimes it doesn’t, and instead of modifying it to make it more socially acceptable, the person may insist on telling the story without making functional adjustments. When significant others disagree with the conclusions of this “totalitarian ego” (Greenwald, 1980), and sense its rigidly defended fortress, they judge the person to be irrational. They may then begin to avoid or exclude the person, or classify her or his unacceptable thinking as “psychopathological” in clinical terms, or “crazy,” “weird,” in lay terms (Lemert, 1962). That negative social response adds to the person’s feeling of deviance and social isolation. Another aspect of the social dimension of Discontinuity Theory comes from considering the second need postulated as basic to human functioning, the need to belong and to affiliate with others.
2. The Social Search Process Through the process of belonging we are in a better position to establish the normality of our actions, feelings, and thoughts through reference to community standards of acceptability, appropriateness, and desirability. The normative basis of our behavior is usually established via social comparison processes of observing comparable others in similar behavioral settings (Festinger, 1954; Wood, 1989). In contrast to the Why? questions of the rationality-seeking internal search, the normality-seeking external search mission seeks answers to questions, such as: “Who or what is doing this to
368
PHILIP G. ZIMBARDO
me?”, “HOWshould I act and what should I feel in this setting?”, “Who else is reacting as I am?”, and “What are they thinking about me?” This search may go beyond the rather passive process of merely observing, judging, and deciding on the normativeness of one’s reaction and its situational inputs. Instead, it can be transformed into a proactive process, as when detecting discrepancy between self and others, the individual actively tries to modify others’ beliefs or actions to be more consonant with his or her own. This normative discrepancy can also be handled in other ways, by changing reference groups, by giving up on usual o r traditional ones perceived as difficult to influence or as exerting too much pressure toward unanimity of behavior. At this point, the person may seek out new groups that seem more open to such “deviant” ideas or actions. With sufficient social influence skills, one can reassert the “normality” of one’s unusual reactions by proselytizing for a new social movement, starting a cult-like group, initiating an anonymous Internet chat group around controversial issues or conspiracies and multiple chemical sensitivities, or spreading rumors that create arousal states in others which are similar to one’s own. During World War 11, research on rumor transmission revealed that there was a marked increase in the spread of fear-arousing rumors among soldiers on their way to combat zones, rather than (as might be expected) rumors that reduced the level of already high arousal. But this greater spread of fear-arousing rumors occurred only under conditions of high egoinvolvement and high ambiguity (Allport & Postman, 1947;Rosnow & Fine, 1976). By creating and transmitting such highly arousing communications among similarly situated peers, any soldier’s extreme fears were normalized by spreading a net of arousal over the entire comparison group. There are a number of measurable behavioral concomitants of each of these cognitive and social search processes. For Actors focused exclusively on the internal cognitive search for explanations of discontinuity, Discontinuity Theory predicts that they will 1. prefer a period of initial temporary social isolation rather than be in the company of others, until a suitable explanation is available to test out; 2. prefer information gathering from experts to social support from family and friends; 3. misattribute unusual psychological reactions to more ordinary biological causes (for example, anxiety may become hunger); 4. misattribute chronic, negative potential causes to reversible, modifiable ones; 5. misattribute unrecognized organic causes of discontinuity to psychological constructs that are familiar or currently fashionable;
DISCONTINUITY THEORY
369
6. use intrapsychic defense mechanisms, such as denial, compartmentalization, or intellectualization, rather than those based on interpersonal strategies, such as projection.
By contrast, the Actor enmeshed in the external situationally-centered social search for normality will: 1. be more conforming, compliant, and imitative; 2. seek out others who are reassuring and supportive, but not necessarily experts with relevant information; 3. affiliate more with others who are acting “unusual” or in ways comparable to one’s state of tension/arousal; 4. mislabel subjectively experienced states, such as anxiety, to objectively specified external dangers, such as fear, which are situationally shared; 5. use defense mechanisms that are primarily interpersonal, such as projection. When conditions make it difficult for one of these needs to be satisfied, then pressures increase for the other need to take prominence. When an adequate explanation for a discontinuity is not forthcoming, then greater effort will be expended toward normalizing activities. Similarly, if the normalizing function cannot be satisfactorily exercised in a given situation, then we should expect more sustained efforts toward generating rational or rationalizing accounts of the discontinuity. However, when it begins to appear that neither the search for rationality nor normality is functioning effectively, then arousal levels are increased because of new concerns for being perceived as both “ir-rational” and “ab-normal,” and thus not comparable to others in one’s thinking, feeling, acting. That new high level of arousal can interfere with both reasoned analyses and systematic social search processes. This dual impediment fuels a downward spiral toward more primitive and regressive ways of thinking, rigidified functional fixedness, social isolation, or even socially deviant acts. It is not clear whether these two search processes occur sequentially, simultaneously, or in an alternating pattern. It is also not clear which process takes initial precedence. That probably depends on the specific type of discontinuity encountered (to be elaborated in the next section); the person’s relative hierarchy of needs to know versus needs to belong; features of the situation in which the discontinuity emerges (alone or with others present); priming factors; and any learned tendencies or dispositions toward focusing on analytic or social modes of thinking. Discontinuity Theory posits that some social affiliations are sought by those experiencing significant discontinuities of various kinds. By doing so, the Actor “normalizes” unusual personal reactions being experienced
370
PHILIP G. ZIMBARDO
privately or enacted publically, to the extent that the group is accepting or supportive of those deviant, exaggerated, or atypical reactions. This change in reference group reduces the Actor’s risk of not being able to adequately explain the discontinuity to former peers and enhances general self-esteem through the initial acceptance and positive social comparisons with the new group members. A recent study offers some interesting support for this prediction by revealing a paradoxical relationship between self-esteem, delinquent behavior, and delinquent peer associations. Jang and Thornberry (1998) found that contrary to traditional analysis, low self-esteem does not lead either to increased association with delinquent peers or to engaging in delinquent behavior, as shown in their multiwave panel study repeated over several years. However, among their large sample of “at-risk” urban adolescents, engaging in delinquent behavior lowers self-esteem (r = -.22), whereas making delinquent associations enhances self-esteem (r = - .30). This general effect holds across gender and racial or ethnic groups, and is even stronger for females and African Americans than for Caucasian male adolescents. The elevated self-esteem that follows associating with delinquents, but not engaging in delinquent behavior, is strongly related to selfacceptance (r = .76) and negatively related to measures of self-rejection (r = -.72). Thus, global self-worth of individual “at-risk” adolescents is enriched via this social affiliation, which I would argue entails the search for normality. The authors conclude that normative support from associating with delinquent peers promotes self-esteem possibly through the intervening processes of reflected appraisals and social comparison rather than those of self attribution. 3. The Behavioral Search Process
But I will complicate the situation a bit more, by proposing that a third alternative search, the behavioral, may be instituted in place of either the cognitive or social, or at any time when those searches are not satisfactory. Although conceptually less interesting than our dynamic duo, it is likely to be more widely practiced among the less educated general population (especially by those who are not typically given to engaging intrapsychic analyses and complex reasoning) or among the introverted and already socially isolated. The behavioral search, when energized by a significant discontinuity, is concerned not with the “Why?” or the “Who?” questions of the cognitive and social searches, but instead with the “How?” question: “How can my distressing feelings be reduced?” And the answer is relatively easy. The Actor can directly or indirectly deal with the arousal or anxiety
DISCONTINUITY THEORY
371
component of the discontinuity by dulling cognitive processing through taking alcohol or drugs, by engaging in a variety of high-intensity physical or emotional activities that justify one’s strong feelings, or by immersion in distracting activities that redirect attentional focus away from even thinking about the discontinuity and its precipitates. These mental and emotional activities may create a dissociated state. But in another sense, such a person moves into a state of deindividuation where cognitive controls that usually govern behavior are dumped out of consciousness (Zimbardo, 1970). Discontinuity Theory thus combines elements of both cognitive dissonance (Festinger, 1957) and social comparison processes (Festinger, 1954) with self-perception theory (Bern, 1972) and also attributional processes (Heider, 1958; Kelley, 1967). The seeds of this approach were evident in a conclusion I advanced in my book on motivational aspects of dissonance theory (Zimbardo, 1969): A systematic integration of social-comparison theory (Festinger, 1954). Bem’s (1967) provocative self-perception model, and dissonance theory is needed to develop the social implications of this aspect of dissonance generated by inconsistency between a discrepant commitment and concern for evaluation of the self by others. (P. 280)
Counter-normative commitments elicit dissonance under high choicelow justification conditions because they are not rational actions given the subjects’ lack of awareness of the situational forces impinging on them. However, the usual “rationalizing” dissonance-reduction tactics repeatedly found in such experiments should be undercut when subjects are provided with other information that “normalizes” their discrepant decision. Craig Haney’s (1978) dissertation research supported that prediction by showing that the classic dissonance-reducing changes in attitudes (from writing essays contrary to one’s initial position) were minimized by social information that several others had also made that same commitment, but were maximized when subjects believed that others had refused to make that commitment. When they could make relevant social comparisons, the subjects’ needs to justify a discrepant attitude position, and seem more rational, were either significantly reduced, or exaggerated. C. BIASED RATIONALITY
Why is it that people do not always generate the correct causal explanation for the observed effect when they are trying to account for a significant discontinuity? Of course, many times, or even most times, we do advance
372
PHILIP G. ZIMBARDO
the correct explanation, and the search ends there. Lost a good job? Answer: company is downsizing. Failed a test? Answer: I haven’t prepared for it. But there are many answers to why that doesn’t always happen, and why we come up with false, faulty, or misleading explanations and causal misattributions. First, since the information search is personally motivated, and not a dispassionate search for truth, it is more vulnerable to distortion by the operation of various cognitive biases operating at both stages of discovery (predecision) and justification (postdecision). Prior experience, or recent priming, may make the Actor attend selectively to some probable causes more than to others. When the discontinuity involves any negative state of arousal or physical distress, the search is directed toward finding only previously established classes of negative causes, and not all possible causes. If the discontinuity occurs in an interpersonal setting where an explanation must be generated quickly to satisfy others in addition to one’s self (such as in the midst of failing to perform sexually), that additional time pressure imposes constraints on the search to overvalue situationally salient stimuli or readily available personal schema (Kihlstrom & Cantor, 1984; Natale & Hantas, 1982). More obvious figural stimuli get noticed and other background ones (that may be the causal agents) get ignored. It may also lead to reliance on the heuristic with top-of-the-memory answers seeming most relevant and true (Taylor, Kahneman, Slovic, & Tversky, 1982; Tversky & Kahneman, 1973). When others are involved, this search for reasons is also constrained from the start by awareness of what kinds of explanations they are likely to accept as reasonable (Darley & Fazio, 1984; Schlenker, 1980; Shatter, 1984; Tetlock & Manstead, 1985). Other biases may operate to influence how available data are processed, integrated, stored in memory, retrieved, and acted upon (Bower, 1981; Bern, 1972; Dodge & Frame, 1982; Goldfried & Robins, 1983; Hodgins & Zuckerman, 1993; Quattrone & Tversky, 1984). There are two other general ways in which the search for meaning can become misdirected. Accurate identification of the true source of any discontinuity may be limited by the operation of either learned or situationally induced cognitive biases (Nisbett & Ross, 1980; Wong & Weiner, 1981). While some of these biases direct attentional focus away from the true source, others misfocus search strategies toward specific types of explanation and attributions that are more appealing, though wrong. Although some misattributions may intensify emotional reactions, others may reduce them by transferring the locus of arousal from internal states (fear or anxiety) to neutral, objective external conditions, such as noise, or crowds, for example (see Brodt & Zimbardo, 1981; Davison & Valins, 1969; Nis-
DISCONTINUITY THEORY
373
bett & Schachter, 1966; Ross, Rodin & Zimbardo, 1969; Schachter & Wheeler, 1962). Let’s outline how each of these opposite forces work to produce inaccurate causal explanations-despite a lot of reasoning. I . Misattributions away from the Primary Discontinuity Source
Why does reasoning move away from the correct causal explanation of a discontinuity? Some of the relevant conditions are as follows:
1. Refusal to acknowledge causes that seem chronic, irreversible, negative and trait-like (you are ugly, stupid, incompetent, unloved, undesirable, old, etc.). 2. Refusal to accept the accountability of the true casual agent because of hidher power, credibility, authority, or dependent relationship, and social support for doing so (your father abused you, your mate is cheating on you, your leader betrayed you, etc.). 3. The cause is not salient because it is embedded in a complex stimulus context. 4. The onset of the cause is gradual, delayed, or requires an additional catalyst, or interacting second variable to become active. 5. The cause is the absence of something critical, which is harder to detect than its presence (as in the slow development of hearing loss). 6. The causal agent is not visible and needs special detection methods (as in hormonal imbalances or selective cerebral damage). 7. The causal agent is not yet known to medical science, or is not yet part of the current knowledge base. 8. The active causal agent is seemingly neutral because it is a conditioned, or symbolic, representation of the original powerful unconditioned stimulus. 9. The true cause is not accessible for conscious processing (as with emotion-arousing information “heard” under surgical anesthesia, or amnesia for childhood abuse). 10. Others purposely conceal the true cause from the Actor “for the person’s own good” (as happens when parents do not prepare a child for their imminent divorce, or the expected demise of a loved one, despite the obvious conflict or stress in the environment, or when a child is given a falsely elevated sense of her or his ability or competence). 11. The cognitive state of the Actor at the time of the search may be suboptimal for careful, reasoned analyses (fatigued, stressed, depressed, excited, confused, drunk, etc.). 12. Correlated patterns are mistaken for causally related events. 13. Chance or coincidental events are assumed to be causal agents.
374
PHILIP G. ZIMBARDO
14. The true situational determinants are cleverly concealed by experimental social psychologists in deception research paradigms. 2. Misattributions Toward Alternative but False Sources In addition to these reasons for not noticing or acknowledging the true cause, the Actor may be drawn towards certain classes of explanation that are wrong, doing so because of the operating force of other competing biases, such as the following: 1. The presence of a salient, dramatic event that dominates possible causes. 2. The desire to find and endorse explanations that enable behavior or bodily conditions to seem modifiable or improvable, or to pursue goals achievable by effort, time, and expert intervention (for example, your problem is motivational rather than an ability deficit, the condition is psychological rather than physical and organic, etc.). 3. The ready availability of culturally sanctioned explanations (stress, PMS, in the genes, conspiracies, the Evil Eye, celestial or spiritual activities, etc.). 4. Priming by recent exposure to a particular stimulus class. 5. The wish to have the explanation absolve the Actor of personal responsibility for the discontinuity. 6. Choice of causal explanations that elicit attention, sympathy, support from others. 7. Selection of a cause that maintains consistency with the Actor’s relevant beliefs, values, or ideology, even if false. 8. The reliance on learned, reinforced explanatory categories to guide the Actor’s search for meaning. The operations in both categories of (4) and ( 5 ) are combined in those for whom the discontinuity is a dysfunctional lifestyle that is misattributed to their being abused as children. The significance of the final category of learned search frame biases is elaborated next because it is one of the unique aspects of this theoretical approach. 3. Explanatory Search Frames Biasing Discontinuity Explanations
Many of us get reinforced for using particular classes of explanation when faced with a discontinuity to be explained. I term this construct explanatory search frames. We may imitate explanatory modes that are preferentially used by family models, or learn them from our exposure to other models in religion (Proudfoot & Shaver, 1975), education, the media,
DISCONTINUITY THEORY
375
friends, or others. These classes of explanation can be conceptualized as being arranged in a personal hierarchy of availability or accessibility (in the Tversky & Kahneman, 1973, sense), readily activated by the perception of a discontinuity. They determine how the search for meaning is framed initially, by focusing on either physical or social sources of discontinuity, and what questions are likely to generate answers relevant to that realmWhat? for the physical versus Who? for the social potential cause. They may direct searches either internally to aspects of body or self, or externally to the environment, to other people, or agencies. It is instructive to realize that anomalies of experience may occur in many domains of cognitive functioning, such as attention, imagery and perception, recall, recognition, judgment and belief, and anomalies in experiences of the self (Reed, 1974). To get a sense of how people explain a wide range of possible discontinuities, I asked several hundred of my students to respond to thirty hypothetical scenarios involving a variety of unexpected events that each had a positive or negative outcome. Their open-ended responses were coded and then given to another large class of students who reacted to the discontinuity scenarios by choosing one of the fixed-response alternatives that best explained the anomalous experience. Specifically, they were invited to imagine each scenario as having just happened to them, and then to ask: “How could this have happened, or why?” They were instructed then to “provide whatever possible explanation(s) you can come up with.” The gist of these scenarios can be gleaned from the following examples: 1. “You get together with a group of people to work on a joint project. Everyone else seems to be whispering. When you ask them why they are talking so quietly, they deny that they are doing so-but keep it up.” Explanations:(a) The others are talking about me and/or trying to exclude me from their group. (b) There must be something wrong with my hearing that I was unaware of. (c) the background noise and room acoustics are creating conditions that make communication difficult. 2. “You’ve been doing an excellent job at your place of employment. Promotions are being passed around, but your boss ignores you and your good work is not acknowledged.” Explanations: (a) The company criteria for promotion place more emphasis on being socially attractive than on being accurate and reliable. (b) My boss doesn’t like me and is out to undermine my confidence. (c) It’s divine intervention because the promotion would not be in my best long-run interests.
From this exploratory study, it was apparent that across this set of quite varied types of discontinuities, eight categories of explanation exhausted most of the variance. as shown in Table I.
376
PHILIP G . ZIMBARDO
TABLE I EXPLANATORY SEARCH FRAMES UTILIZED TO ACCOUNT FOR
A
VARIETY OF DISCONTINUITIES
Internal focus 1. My body and health status 2. My self-attributes,dispositional tendencies, personality traits (such as being “unlucky,” or a “shlemiel”) 3. Something from my past External focus 4. The currently experienced physical environment 5. My social environment, another person, or group 6 . Mystical or spiritual forces (such as gods, angels, devils, witches, extraterrestrials) 7. Structural or ideological systems (such as racism, sexism, Communism, liberalism) Dispersed focus 8. Chance, coincidence, “dumb luck,” or merely correlated events
The last category of chance or coincidence was the least frequently proposed as a cause for these varied and often extreme instances of discontinuity. Instead of “dumb luck” operating in a discontinuous experience, people are more likely to interpret it dispositionally (“I am a lucky/unlucky person”) as the causal mechanism. Incidentally, this exercise helped me to understand a pedagogical puzzle, why so many bright students in my classes do not learn the maxim that “correlation does not imply causation.” I now believe that the mind is designed to transform perceived correlations into causal experiences, so it becomes difficult to accept the fact that they are really different processes. In the final part of this section, I will advance another special feature of Discontinuity Theory, namely the prediction of specific types of pathology based on which of these characteristic explanatory search frames are habitually used when dealing with personally significant discontinuities. But before doing so, I will discuss the more general role of reasons, attributions, and narrative stories in formulating decisions. 4. Reason-Based Decisions about the Best Discontinuity Analysis
Our prototypical Actor perceives a discontinuity that qualifies as “significant” according to the criteria noted previously, reflects upon alternative causes, and makes the decision that one of them is most probable. Having chosen that option, the Actor then seeks evidence to bolster that decision (Lord, Ross, & Lepper, 1979). Although some Actors exert considerable mental effort in searching for a compelling rationale for choosing one alternative rather than any alternatives, other Actors’ choices may be imme-
DISCONTINUITY THEORY
377
diately and nonconsciously influenced by affective judgments (Kihlstrom, 1984,1985; Marcel, 1983; Rachman, 1981; Rosenberg, 1960; Zajonc, 1980). Research on reason-based choice (Shafir, Simonson, & Tversky, 1993) shows that the reasons considered in making a choice may be influenced by the effects of context, salience, compatibility, framing, and elicitation procedures, which are all manipulable. These variables “highlight different aspects of the options and bring forth different reasons and considerations that influence decision” (1993, p. 34). In the situations we have been considering so far, classical decision theory does not hold because people do not have stable values or preferences; rather these are constructed “on-line” during the process of dealing with a given discontinuity scenario. These cognitive scientists (Shafir et al., 1993) conclude their experimental investigation of how reasons guide choice with a statement quite compatible to those advanced earlier in this chapter: We often search for a convincing rationale for the decisions we make. whether for inter-personal purposes. so that we can explain to others the reasons for our decisions, or for intra-personal motives. so that we may feel confident of having made the ‘right’ choice. (p. 33)
This cognitive approach to the role of reasons that precede and guide choice contrasts with the postdecisional processes of generating rationalizations and justifications that has characterized much social psychological theorizing and research (e.g., Aronson 1968; Bern, 1972; Cooper & Fazio, 1984; Festinger, 1957; Wicklund & Brehm, 1976; Zimbardo, 1969). Discontinuity Theory, as outlined here, proposes that the role of reason operates at both pre- and postchoice stages of generating alternatives, choosing among them, and then constructing adequate rationales for why that choice was best. I prefer to think about these two temporally distinct phases as the context of discovery and the context of justification, respectively. Precisely because Actors must often consider the necessity of convincing others of the reasonableness of the causal argument selected, they need to create accounts, narratives, or mini-theories that enrich the central reason(s) guiding and justifying their position. A similar view has been argued by Billig (1987), who advances a rhetorical approach to understanding such issues, noting that, “our inner deliberations are silent arguments conducted within the single self” (p. 5). Surely that notion captures the process going on inside the mind of the Actor coping with the tension created by a significant personal discontinuity. We will see evidence for such a claim in both the case studies that follow this general section, and also in some of the personal accounts collected during the experimental testing of various aspects of our theory.
378
PHILIP G. ZIMBARDO
D. MADNESS AS IR-RATIONALITY PLUS AB-NORMALITY: ENACTING ONE’S THEORY IN SEARCH O F DATA -The
mind is its own place, and in itself/Can make a heaven of hell, a hell of heaven. -Milton, Paradise Lost
When the Actor does generate reasons, or a story, in the form of a personal theory that seems to account for the discontinuity, then that theory takes on a centrifugal force that sucks up all evidence into its vortex of justification. Only confirming evidence is noticed, catalogued, and packed into the emerging pattern of self-evident truth, while critical disconfirming data are set aside as irrelevant. This person gets into trouble in two ways. First is the failure to be an objective, diligent, bottom-up monitor of available data, and the second is the failure to generate a top-down set of the most reasonable hypotheses with alternatives that must be disconfirmed and discarded before accepting the most probable one. As a biased theorist enamored of one’s little theory, one’s conjectures and misattributions soon become “facts” on which other decisions are based. Under some circumstances, these mental transformations cause the initial perturbations from a discontinuity to escalate into pathological reactions. If a discontinuity is personally significant and ongoing, rather than transient, and the Actor’s searches for rationality and normality do not adequately achieve a sufficient degree of anxiety reduction, then the possibility exists that more drastic psychological processes will be called into action. There are several forms this action can take. Failing to discover order, logic, and a satisfactory reasoned analysis of the original discontinuity, the person may impose a spurious kind of order and organization on the rest of one’s life through reliance on compulsive thinking and obsessive behavior. She or he may develop a ruminative thinking style that replays aspects of the old discontinuity scenario, exaggerates self-blame for negative outcomes, and thereby encourages a depressive disorder (Nolen-Hoeksema, Parker, & Larsen, 1994). If there is the danger that others will reject the Actor’s explanation and not be socially supportive, then social withdrawal and isolation handles that challenge to one’s self-esteem. If the discontinuity-generated anxiety cannot be reduced by the cognitive, social, or behavioral searches, then it can be eliminated by suppressing all forms of arousal and becoming motiveless, or affect-less. If there is not consensual validation for the Actor’s explanation, nor acceptable external validity checks, he or she can reverse the process by making internal reality the standard against which to validate the external. Finally, if standard forms of reasoning have proven ineffective, then reliance on nonstandard forms of logic, reasoning, and language can
DISCONTINUITY THEORY
379
prevent refutation of the Actor’s proofs. Such tactics, operating at various levels of conscious and unconscious processing, obviously describe the cognitive, affective, motivational, and behavioral characteristics of serious mental disorders. In some cases, we can observe the development of a rigid character structure that comes to dominate the person’s lifestyle. Shapiro (1981) notes that “such a continuously deliberate, purposive, and tense self-direction involves a special kind of self-awareness or self-consciousness. Rigid people direct themselves with a greater and more extensive awareness than others have of what they are doing, and how they are doing it” (p. 70). In the final experiment to be presented in this chapter clinicians judged many of our ordinary, “normal” research participants as suffering from “pathological” disorders. I believe their judgments were swayed by the extreme rigidity with which these students maintained untenable explanations to account for the discontinuity of their unexplained arousal. Sociologist Edwin Lemert (1962) concludes his analysis of the conditions that lead to paranoid reactions among usually normal individuals with this summary of social and personal discontinuities: A number of studies have ended with the conclusions that external circumstances-
changes in norms and values, displacement, strange environments, isolation, and linguistic separation-may create a paranoid disposition in the absence of any special character structure. (p. 70)
We can now advance the strong form of our argument: All forms of madness that do not have their origins in organic, brain, or hormonal defects, are the perversion of the most natural and basic human functions, the needs to know, to belong, and to gain esteem. In the process of engaging in the fundamental aspects of being a thinking, relating human being who desires to understand the unknown, to order chaos, and to enrich the human connection, some of us get led astray. We start down the wrong path to getting the answers we seek by using biased thinking strategies when trying to find causal explanations for experienced discontinuities. These explanations become the symptoms of faulty thinking, of pathological reasoning, of delusional beliefs of inappropriate affect. The symptoms of madness emerge from, and are sometimes isomorphic with, the explanations generated to provide a narrative account of certain discontinuities. Explanations need not be verbally stated propositions, but can be nonverbal messages that convey various meanings to the Actor and observers. For example, a headache can be an explanation for academic or sexual performance failures, or a muscle spasm may become the reason for sports failures-both are forms of nonconscious self-handicapping reactions. Discontinuities that are unexplainable and not open to normalizing through social affiliations may trigger a transformation of the disturbed
380
PHILIP G. ZIMBARDO
psychic energy they create directly into somatic symptoms, such as asthma (Eastwood, 1975). We may “choose” physical symptoms that convert ineffable psychological distress into bodily problems that can be treated, discussed, and that often get sympathy (Pennebaker & Epstein, 1983). That happened to me during my junior year in high school. I had always been very popular in my Bronx neighborhood and in the various schools I attended, but was totally rejected for almost an entire school year when I went to North Hollywood High School. This social status discontinuity was alien to me, and did not make sense since I had not changed in any way, but it was evident that I was being shunned. After a month or so, I developed nighttime asthma, that made it difficult to sleep, caused me to miss classes or come late, and elicited much concern from my family. So much so, that they moved back to the Bronx at the end of the school term. I learned late in this unfortunate experience that a rumor had been circulated that I was a New York Italian connected to the Mafia, and the other students were afraid of me! Back in the Bronx, I was elected the most popular boy at James Monroe High School the very next term. That transformation based on situational context and not dispositional variability was a source of interesting dialogues with my classmate, Stanley Milgram. The secondary anxiety about appearing to be irrational, without a satisfactorily reasoned analysis for important discontinuities in one’s life, may also lead to some forms of social pathology as the Actor eschews usual reference groups for more exotic, bizarre, deviant ones that normalize idiosyncratic behavior in the unusual behavioral context they provide. We have also described how more extreme reactions can surface when the Actor becomes aware of seeming to be both irrational and abnormal-such reactions as social isolation, affect suppression, and reversed reality checks. Another form of personal pathology that can result from the processes described is the emergence of addictive and compulsive behaviors that get reinforced while trying to suppress the anxiety component of discontinuities by resorting to direct behavioral strategies. 1. Why Call it “Madness?” What is madness/To those who only observe, is often wisdomlTo those to whom it happens. -Christopher Fry,A Phoenix Too Frequent
Using the term “madness” permits inclusion of a wide spectrum of reactions across a diverse body of available literature in anthropology, sociology, law, drama, history, as well as psychiatry and clinical psychology (see Claridge, 1985/1995; Foucault, 1967; Franzini & Grossberg, 1994; Henry,
DISCONTINUITY THEORY
381
1971; Orford, 1976; Scheff, 1975; Shapiro, 1965; Szasz, 1973; Wing, 1978). Madness, in such a view, is not an objectively verifiable fact, as is cancer or stroke. Madness is always an attribution about a deviation from expected normative behavior. It is an attribution made by human observers and is particularly significant when they have expert or legal power. Morton Schatzman (1973) asserts such a view in his book, Soul Murder: “Many people feel persecuted, but no one ever feels paranoid. Paranoia is not an experience; it is an attribution one person makes about another” (p. 147). Such attributions, explanations, and accounts of unusual behavior must be understood in terms of the social transactions among people (see Antaki, 1981). To be labeled “mad” is dependent on cultural and contextual factors, on interpersonal power relationships, on the current value to the society of the Actor in question, as well as on the degree of social support or isolation of that person (Medvedev & Medvedev, 1971; Mezzich & Berganza, 1984). Standards of rationality and normality are culturally determined, varying as well over historical epochs as knowledge changes and as culturally sanctioned explanations for discontinuities and madness also change (Bell, 1980; Eaton, 1980; Levi-Strauss, 1966; Porter, 1989; Rosen, 1968; Rosenbaum, 1970; Van Gennep, 1960). Madness aIso varies across a broad continuum from mildly deviant, aberrant, eccentric in thoughts, feelings, and actions to “being out of one’s mind,” running amok, going berserk, certifiably crazy, and Diagnostic and Statistical Manual ( D S M )pathological (APA, 1980;Fabing, 1956;Luborsky, 1970). My predictions about the madness-engendering consequences of some discontinuities and their associated biased search frames do not extend to any pathologies that are organic in origin, or genetic, or developmental, but to learned adaptations or maladaptations. The normal side of madness involves “functional perturbations” in normal people in their obsessions (Kohler, 1958; Rachman & Hodgson, 1980); motivated irrationality (Frankel, 1973; Pears, 1982); benign paralogical thinking (Glasner, 1966); selfdeceptions (Goleman, 1985; Gur & Sackheim, 1979; Sarbin, 1981); mental epidemics (Young, 1927); “love melancholy” (Burton 1621/1977);“creative malady (Pickeri, 1974); everyday paranoid thinking (Artiss & Bullard, 1966); eccentric and bizarre behaviors (Franzini & Grossberg, 1994); suspicions that merge into delusions; occult beliefs (Singer & Benassi, 1981); and other forms of abnormal behavior in normal people, such as prejudice, vandalism, violence, self-derogation, and addictions (see Gruder, Gurwitz, & Reiss, 1977). We also need to recognize the extent to which apparently functioning members of society may still be carrying around a high degree of pathological symptoms. In a recent study of more than 600 students at three private and public colleges, 18% had average scores on a general symptom inventory that exceeded outpatient norms, and for 17%
382
PHILIP G . ZIMBARDO
the number of symptoms they reported exceeded even inpatient norms (Holman & Zimbardo, 1999). It can be argued further that madness decreases in a society to the extent that (a) the society provides a ready storehouse of sanctioned hypotheses, theories, or rationales for the appearance of discontinuities in thinking, feeling, and acting; and (b) the society intervenes to restore harmony by accepting some responsibility for its role in the discontinuity, as we saw earlier in the example of the rituals practiced by the Nyakusa (Wilson, 1962). For some kinds of discontinuities, readily available societal explanations abound in the form of truisms, homilies, folklore, wisdom of the elders, and white or black magic. When the culture has the answer, the individual’s search for a discontinuity-based reason is short-circuited. By accepting the dominant, prevalent explanation, and being open to whatever prescribed ritual practices accompany it, the Actor does not risk social rejection. Simply following the standard operating procedures ensures the Actor will be perceived as doing the “right thing.” Normality is thus assured, even if the treatment does not change the state created by the discontinuity. Conversely, madness becomes more prevalent to the extent that individuals are forced to generate their own explanations for experienced discontinuities, and also to provide necessary data sets that verify those attributions-to the approved standards of critical societal judges. This reasoning leads to the obvious conclusion that the frequency of psychopathology should be greater in more individualistic than collective societies. My perspective suggests that madness is not an “alien” aspect of humanity, but merely a perversion of human perfectibility. In understanding madness, we understand some of the most fundamental dynamics about the human mind: how we know reality, distinguish internal from external, represent sensory and social experiences in memory and reasoning, make causal attributions, and find evidential bases for our theories. Instead of focusing on inner “disease states” that need to be “cured,” or on “premorbid” personalities with a ready vulnerability to mental illness, the present model postulates that madness is t o be treated by first ruling out organic impairments and dysfunctions, and then by examining (a) the nature of the explanations for discontinuities that ordinary people generate; (b) how and why these explanations have become biased; (c) why people persist in maintaining them despite contrary evidence; and (d) what can be done to correct these biased explanatory searches, misattributions, and faulty conclusions. 2. Explanatory versus Confirmatory Processes in Symptom Development What is madness? To have erroneous perceptions and to reason correctly from them. -Voltaire, Madness, In Philosophical Dictionary
DISCONTINUITY THEORY
383
Mad is a term we use to describe someone who is obsessed with one idea and nothing else. -Ugo Betti, Struggle Till Dawn
So far, I have argued for the central role of discontinuity as a catalyst of various search processes, some of which swerve off the path of rationality and normality to create disorders in cognition and affect that may result in the formation of pathological symptoms. Further, I have maintained that biases in the explanatory search frames used to make sense of discontinuities mediate the development of unusual explanations that serve as symptoms, some functioning as self-handicaps (Kolditz & Arkin, 1982). However, I have combined two different processes that should be considered separately. While the explanatory search, triggered initially by the need to explain a discontinuity, leads to the development of theory building, the confirmatory search integrates the Actor’s current experience and situational information as support for the established personal theory about the nature of that discontinuity. This distinction was first made in a brilliant thesis by Lisa Butler (1993), working in my laboratory, as a way to understand how paranoid symptoms could reflect a disorder of affect (specifically, anger) and not a disorder of cognition, as traditionally believed. Although Butler’s distinction limits the two searches to her primary interest in paranoid processes, I believe it is more widely applicable to other pathological processes as well. Her analysis also distinguishes between bottom-up and top-down information processing as they operate in each of these two cognitive processes. Butler argues that the development of paranoid symptoms, such as suspicions and delusions, precipitated by discontinuities in explanatory or reactive paranoia, depends on a data-driven or ‘bottom-up’ process of explanation or theory building. She contrasts that process with confirmatory or proactive paranoia, in which “the expectations of a paranoid personality or the delusional beliefs of a paranoiac drive an inspection and ‘top-down’ interpretation of features of the social and physical environment with its goal being to reconcile experience and information with the prevailing predispositional theory” (p. iv). Paranoia is a mental disorder that ought to be of considerable interest to social psychologists because of the critical role played by people schema and misattributions about actual or imagined others (Maher, 1974a,b; Maher & Ross, 1984). It also shares features with prejudice in its creation of adversaries, resistance to change of belief structures, the central position of the self in paranoia, and of one’s identity group in prejudice (Fenigstein & Vanable, 1992). John Kihlstrom effectively develops the view that paranoid delusions are explanations of anomalous experiences that are best understood within
384
PHILIP G. ZIMBARDO
the framework of attribution theory and social cognition (Kihlstrom, 1979; Kihlstrom & Hoyt, 1988). He elaborates upon Brendan Maher’s (1974a,b) analysis that delusions are the product of normal reasoning processes initiated by an unexplained perceptual anomaly, one that arouses anxiety and demands explanation. In this formulation unexplained anomalistic experiences are primary while delusions are secondary. Similar views have been advanced by Jaspers (1962) and Reed (1974), who argued that primary delusion was a contentless, diffuse, ineffable perceptual experience. Delusions are then developed as belief systems-false beliefs-that provide the content, structure, and meaning for the anomaly. They are selfreferential ideas about oneself, whose “content is crucially related to the individual’s fears, needs, or security” (Reed, 1974, p. 83). Such delusional beliefs “reflect the patient’s attempts to describe, from attributions about, and explain primary delusional experiences. These explanations draw on the person’s fund of world knowledge and repertoire of inferential processes” (Kihlstrom & Hoyt, 1988, p. 83). In a sense, these delusions are reminiscent of the explanations given by Gazzaniga’s (1998) split-brain patients whose verbal, left hemisphere generates explanations, sometimes bizarre, to account for the perceptual or sensory experiences of the right hemisphere. Just as it is difficult to convince such patients that “believing is not seeing,” so too, it is difficult for normal individuals and schizophrenics alike to selfcorrect their delusions and irrational explanations as long as the anomalies persist that they purport to explain. This is so for all the reasons that they do not behave like professional scientists in using optimal hypothesis-testing strategies, searching for diagnostic information, seeking disconformation, and minimizing subjective biases. Mark Leary and Rowland Miller (1986) effectively articulate the general utility of social psychology for understanding the origins, diagnoses, and treatment of dysfunctional behaviors, usually the sole province of clinical psychology and psychiatry. They elaborate on a number of the processes and constructs outlined in this chapter that link basic aspects of social psychology to those of “abnormal” psychology (long ago linked before JASP became JPSP). A new, social-information-processingmodel of paranoid social cognition, elegantly developed by Rod Kramer (1998), highlights the importance of the informational and social context in which an individual’s perceptions and cognitions are embedded. This analysis of the formation of suspicion and distrust at all levels of social organization proposes three input variables: feeling uncertain about one’s social standing, or “tenure status,” perceiving one’s self to be distinctive, to stand out, in a group setting; and perceiving one’s behavior under evaluative scrutiny. They combine to promote a dysphoric, negative self-consciousness, which in turn, leads to
385
DISCONTINUITY THEORY
hypervigilance and rumination. When filtered through a series of judgmental biases, these selective, obsessive thoughts emerge as paranoid delusional beliefs. Here is another instance of a social psychological approach to understanding paranoia that deserves experimental testing, as well as application to clinical populations. The social psychology of paranoia is interestingly revealed in one of our experiments (to be presented later on) that induced paranoid symptoms in normal college students through several alternative discontinuity manipulations (Zimbardo, Andersen, & Kabat, 1981). In a matter of minutes, these participants, chosen for their mental and physical health, were behaving in ways that peers and clinical test results agreed was deserving of a diagnosis of a paranoid disorder. 3. Predictions Relating Biased Explanatory Search Frames to Forms of Pathology A man’s worst enemy can’t wish him what he thinks up for himself. (Yiddish proverb)
Finally, let’s move from abstract generalizations to more specific predictions about how particular types of pathology are linked to each of our eight categories of biased explanatory search frames. Table I1 outlines the TABLE I1 PREDICTED PATHOLOGICAL REACTIONS AT INDIVIDUAL A N D SOCIAL LEVELSBASEDON BIASED USE OF SPECIFIC EXPLANATORY SEARCH FRAMES TO ACCOUNT FOR PERCEIVED DlSCONTlNUlTIES Biased focus Internal orientation Bodylhealth status Self-attributes (Negative) Personal past (Aversive, Negative) External orientation Physical environment People MysticaVspiritual forces Structural/ideological agencies Chance, luck, fate (Present fatalism)
Individual pathology
H ypochondriasis, Somatoform disorder Depression Depression
Phobias Paranoid disorders paranoid disorders, spirit possession Fanaticism, suicidal martydom Learned helplessness, addictions, depression
Social pathology Mass hysteria Prevalence of shyness Prevalence of “false memory syndrome” Vandalism Prejudice, discrimination Witchcraft, satanic cults Terrorism, war Maintenance of low socioeconomic status
386
PHILIP G. ZIMBARDO
expected relationships between each type of biased explanatory focus and the forms of individual pathology to which it gives rise. This perspective is extended further by also relating biased search frames to forms of social pathology that are a product of collectives of people employing similar search frames when some of them are experiencing discontinuities. The experimental research section of this chapter presents evidence for the validity of some of these linkages, based on studies that manipulated various biased search frames and measured resulting pathological outcomes. For each type of biased focus, or explanatory search frame, there is a corresponding extreme reaction when it is overused as the only source to account for a significant discontinuity, and is rigidly maintained despite evidence to the contrary. The correspondence rule specifies that when thoughts, feelings, motivations, and behavior are organized around a particular biased focus, then the fundamental feature of that bias acts as an organizing schemata which leads to specific pathological disorders that share the properties of those central schema, The determination or diagnosis of pathology is typically made by those with expert status in the realm of mental health, psychiatrists and clinical psychologists. But such evaluations are routinely made also by those familiar with the Actor, and at times by the Actor as well. Here pathology refers both to the professional judgments of mental disorder and the lay evaluations as well. It is also possible to extend these ideas to social situations where many people share the same discontinuity bias, despite differences in the origins of their discontinuities. Social pathology refers to actions by a collective of people, acting individually or in concert, to engage in behaviors that are nonnormative, socially undesirable, or that violate some higher-order ethical standards. The predictions I feel most confident about are those positing the following correspondences: 1. Those judged to have hypochondriacal, or somatoform, disorders, will most likely be those whose internal biased search frame centers around their BodylHealth Status. 2. Those with phobic disorders will be overrepresented among those whose biased external search frame centers on aspects of the Physical Environment. 3. Paranoid disorders will be traced back most directly to those whose external biased focus is on People. In addition, the biased focus on MysticaVSpiritual forces can also contribute to paranoia if the Actor either personalizes those gods and demons, or assumes that others are influencing evil spirits to act against him or her. These three predictions were directly tested in experimental research to be presented later in this chapter. Manipulated search frames focusing
DISCONTINUITY THEORY
387
on environment, body or health, and people among those coping with unexplained arousal resulted in mental and behavioral reactions that were classified as pathological for some research participants. We found correspondences between environment framing and phobia, health framing and hypochondriasis, and also people framing and paranoia. Returning to the remaining predictions, from Table 11, I cite the following: 4. Depression is most likely to be linked to both an internal biased focus on Self Attributes, if that focus centers on personal inadequacies, deficits, or failures to achieve goals. It can also be linked to an explanatory search bias that focuses on one’s past, if that construal of the past is primarily negative or aversive. A Past-Negative focus centers around ruminations of trauma, abuse, neglect, mistreatment, and failures, which combine to lower self-esteem and raise depression. 5. The external biased focus on Structural/Ideological Agencies, as a causal agent in discontinuities, is most likely when the Actor has a salient identity as a member of a particular group in opposition to competing or more dominant groups in the society. Then the “isms” come to play a significant role in attributions of one’s personal failures and troubles, and for the inequities experienced by one’s group. At an individual level of pathology, the extreme involvement in this form of thinking leads to fanaticism, authoritarianism, and suicidal martyrdom. 6. Those who attribute discontinuities to the operation of chance, luck, or fate feel that their lives are controlled by random forces or are predestined by religious or mystical agencies, rather than by their personal agency. That orientation leads to the pathologies of learned helplessness (Seligman, 1975), pessimism, depression, and addictions to immediate pleasures regardless of long-term costs. Before turning to examine the predictions relating these explanatory search frames to corresponding forms of social pathology, I want to mention another line of research that identifies individual time perspective profiles, using a new metric, the Zimbardo Time Perspective Inventory (Zimbardo & Boyd, 1999). In one of our validation studies, those whose time perspective scores mark them as high on the Past-Negative factor (focus their thinking on aversive aspects of their past life) were lowest in self-esteem and highest o n depressive symptoms, and also on trait anxiety, of any group, with robustly significant correlations for each of these negative outcomes. The Social Pathology side of the agenda extends the conceptualization that went into formulating the individual pathologies to societal forms of dysfunctional behavior. These predictions are made more tentatively than those for the Individual Pathologies, but are perhaps more interesting.
388
PHILIP G. ZIMBARDO
1. Instances of mass hysteria represent a form of emotional contagion that occurs only among those whose primary search frame is Body/Health. When people are in proximity to others in a collective setting, any of whom experience and display reactions to a current discontinuity that is symbolically related to a prior discontinuity, such as a personal loss, a social psychogenic process will be triggered. 2. The focus by individuals on Negative Personal Attributes could, when aggregated, contribute to the high levels of shyness being reported in our society and in others (Carducci & Zimbardo, 1995). Shyness, like paranoia, is as much a social process as it is an individual one. 3. Overemphasis on the negative aspects of one’s Past, as a biased cognitive focus, when combined with a currently dysfunctional lifestyle, and mass media exposure to cases of traumatic abuse, could contribute to the recent phenomenon of False Memory Syndrome versus recovered memories of abuse (see Freyd, 1996, Hechler, 1994; Loftus & Ketcham, 1994; Ofshe & Watters, 1994; Pope & Brown, 1996; Schooler, Bendiksen, & Ambadar, 1998; Winegrat, 1985; and entire issues of Current Directions in Psychological Science, Lynn & Payne, 1997,and Child Mistreatment, Vol. 2, May, 1997). 4. Destruction of aspects of the environment through acts of vandalism is the social pathology of those with an extreme Environment bias (Zimbardo, 1976). 5. Prejudice is the social pathology aspect of individual paranoia. Instead of simply fearing others are out to harm them, the collective of such individuals focuses on others whom they can harm or exploit because these others are inferior relative to their own dominant status as the discriminators (J. Jones, 1997). 6. Practicing witchcraft, and other Satanic cult-like activities, assumed to harness the power of mystical forces for personal or collective gain, is an instance of the social pathology dimension of a biased focus on Mystical/ Spiritual Agencies (see Baroja, 1973; Middleton, 1967; Sebald, 1978). 7. While the positive dimension of this orientation toward explaining discontinuities in terms of Structural/Ideological Agencies is a constructive activism that seeks to change systems of assumed oppression or dominance hierarchies (Pratto, 1999), the negative dimension is anchored by those who form collectives that endorse terrorism and war, the ultimate social pathologies. 8. Finally, an insidious form of social pathology is the maintenance of low socioeconomic status (SES) for entire groups of poor people as a consequence of their adopting a fatalistic, present-orientation that leaves vital decisions to the operation of chance and luck, rather than a developed sense of self-efficacy (Bandura, 1997). These ideas and predictions about social pathology are based neither on compelling evidence, nor rigorously derived from Discontinuity Theory.
DISCONTINUITY THEORY
389
Rather, they are presented for their heuristic value in provoking consideration of a range of dynamic relationships between the cognitive and social psychological constructs outlined in this theory, clinical perspectives on psychopathology, and sociological views of collective forms of pathology. Each biased search focus, hypothesized to lead to particular pathologies when misused in the process of accounting for anomalies in one’s experience, also has an “idealized” version. Such a focus may function productively and constructively to promote value orientations, guide decisions, and initiate individual and collective actions in socially approved directions. When they are practiced by many people within a society, these biased frameworks help form the foundations of many “approved disciplines” and institutions. Consider one set of parallels between such socially shared biased search frames and the approved societal disciplines they engender. 1. Body/health focus supports medicine 2. People focus supports social sciences 3. Environmental focus supports physics, ecology 4. Self attribute focus supports drama, personalitykharacter analysis 5. Temporal focus supports history 6. Temporal focus plus people focus supports psychoanalysis 7. Structural/ideological focus supports politics, military 8. Mystical/spiritual focus supports religion, shamanism 9. Ideological focus plus spiritual focus supports right-wing religious politics We next consider a set of illustrative case examples of various aspects of Discontinuity Theory before turning to our experimental evidence. 111. Case Studies Illustrating Aspects of the Theory
I wish now to introduce a more clinical perspective to this presentation by describing case study examples that enrich the conceptual side of the theory by illustrating its applicability across a spectrum of real-life experiences. Some come from people I have worked with directly, or known personally, others derive from the literature and media sources. They are grouped under headings that relate to different aspects of Discontinuity Theory. A. FAILURE T O RECOGNIZE OR ADMIT THE TRUE SOURCE OF ONE’S DISCONTINUITY
The first of these two cases is perhaps the single best instance of the power of this new approach t o understand and deal with psychological
390
PHILIP G . ZIMBARDO
disorders that stem from the failure to recognize or admit the true source of a discontinuity. The second is a reinterpretation of a classic case study in clinical literature, that of Dr. Schreiber. The memoirs of his psychotic experiences formed the basis of Freud’s theory of homosexual projection in paranoid disorders (Freud, 1911). 1. Gary, the Almost Perfect Student Gary, a freshman in my Introductory Psychology course, came to an office hour to request a referral to a Student Health Clinic therapist. He reported having three problems that started recently and were interfering with his concentration and study, and thus were negatively affecting his grades. Before making that referral, I asked if he first wanted to discuss his problems with me so that my referral would be more informed. In response, he reported the following: (a) that eating difficulties were making him nauseated and unable to eat much food, causing weight loss and a general weakness that made studying difficult; (b) that sexual impotence made it impossible to get or maintain an erection, causing much concern over his sexual identity and masculinity: and (c) that he was experiencing uncontrollable hostile impulses and violent ideation at the slightest sign of disagreement with anyone, and these were very bothersome since he had always been a mild-mannered person. This combination of acknowledged physical and mental problems was responsible for his recent low grades in many courses since they impaired his concentration both in class and studying in the dorm, according to Gary’s reasonable self-analysis. Indeed, this set of “presenting symptoms” deserved expert clinical treatment, which I arranged for him to receive from a colleague, who happened to be psychoanalytically trained. But before leaving, I asked Gary two more questions about the specific timing of the onset of these events, his teachers, and grades. Because these experiences were recent and fresh in his mind, Gary was able to state with certainty the week or so when each of them first were noticed during the current school term. “I am a perfect student” was his answer to the grades question. He had always gotten straight-A grades, graduating at top of his class (in a small rural high school), which made his parents so proud they typically referred to their son as “Perfect Gary.” My colleague in Student Health confided to me that Gary’s was a very serious case, requiring long-term treatment, more than could be handled through their services. His diagnosis integrated all three seemingly disparate problems around the theme of repressed homosexuality impulses beginning to emerge into consciousness (briefly, the impotence was around contact
DISCONTINUITY THEORY
391
with females, eating food was symbolic fellatio, and the hostile imagery was projection of his fears of being discovered). Meanwhile, I had checked with each of Gary’s teachers who told me the dates on which he received his low test or paper grades in their courses. When Gary returned the next week, I invited him to verify the timeline of the origin of each of his three personal problems that were causing his academic problems. He was surprised to discover that in every instance, his poor grades preceded the onset of any of the troubling symptoms of distress. The causal sequence was totally reversed! I explained that the path of my reasoning started with the violation of a vitally important aspect of his self-image-being the perfect studentwhen he first started performing subpar in his courses. Such difficulties are common among many new students at top academic institutions where most other students were also selected because they had been outstanding. Lower grades are a fact of life where grading is “on the curve,” evaluating students’ performance relative to each other, rather than to absolute norms. The true source of his problems might be traced to the discontinuity occasioned by his sudden fall from academic grace and perfection, which had more to do with the new elevated comparison group of “academic sharks” among whom he was now swimming than a loss of brain power. After further discussion of the impossible burdens imposed on him by having to sustain that image of perfection, and learning that the college no longer sent student grades home to their parents, Gary left feeling much less anxious. I checked with him several weeks later and he reported that all those terrible problems had stopped, and his grades were beginning to pick up. He never went back to the therapist, and graduated four years later with satisfactory, average grades from a quite respectable college. Among the conceptual morals of this story relevant to Discontinuity Theory are the following: 1. When the true source of a discontinuity is perceived as negative, chronic, and unchangeable (I am not smart), alternative misattributions are generated to create sources that are perceived as modifiable by expert treatment (I have a psychological problem, rather than an intellectual deficit). 2. Self-handicapping substitutes a motivational basis for failure or anticipated evaluation in place of an ability-based explanation (Arkin & Baumgardner, 1985). 3. Physical symptoms can be the nonconscious, nonverbal explanations for discontinuities (Pennebaker & Epstein 1985) 4. There may be anticipated “secondary gains” in the preconscious choice of symptoms (e.g., Gary got attention, sympathy, and acceptance of the “reasons” for his academic problems)
392
PHILIP G. ZIMBARDO
5. The longer the time gap between experiencing the original discontinuity and seeking expert or lay help for it, the harder it is to discover the true causal timing sequence of discontinuity triggers and consequences, since memory distortion and other intervening discontinuities add to the causal attributional confusion. 6. While many students have similar problems, few develop such extreme solutions as Gary’s, because academic performance had become the most important, or only, yardstick for measuring his self-worth. 7. Biases of observer-listeners to the accounts of someone experiencing a discontinuity will influence their interpretation of its likely causal factors and sequences, and in turn, will get communicated to the Actor as a “working script” for understanding why it is happening.
I will elaborate briefly on two of these points. Most people who seek professional help for mental problems often wait months or years from the onset of the problem to their first visit, a time span that makes it difficult to do what I was able to do to create Gary’s “miracle cure” by uncovering a precise causal-sequence timing. Had he come to see me the next year, he would likely recall only vaguely when his problems began, and his alternative causal sequence would have been difficult to challenge and disprove, especially since it was so reasonable. Secondly, notice that not only did the psychoanalytic training of the therapist direct him to frame Gary’s symptom triad solely within that Freudian theoretical perspective, but he failed even to recognize what my academic bias focused on immediately-the significance of grades in Gary’s narrative (for a report of attributional biases among clinicians, see Plous & Zimbardo, 1986). (In a subsequent section, I will report a more detailed example of the way that such a theoretical bias distorted the understanding by a clinical expert of why my research participants were experiencing intense distress.) 2. Dr. Schreber Is Mad Because H e Can’t Tell the World that Dr. Schreber Abused Him A second instance of misattributing the cause of a discontinuity to a nonveridical source has a less happy ending than our first case. Daniel Paul Schreber (1842-1911) was an eminent German judge who became a paranoid schizophrenic mental patient at age 42, spent 13 of the next 27 years in asylums, and died there. During this time period he collected the notes about his thoughts and experiences into a published book, Memoirs of M y Nervous Illness (Schreber, 1903/1955).Schreber’s memoirs provided Freud with extraordinarily clear descriptions of the patient’s thinking pro-
DISCONTINUITY THEORY
393
cesses and fantasies, according to analysts Kanzer and Glenn (1980, p. 329). So much so, that they became the basis of one of Freud’s most famous cases, his first literary-historical writing of “pathography,” and the basis for his theory of the origin of paranoia as a defense against homosexual love (Freud, 1911). There are a number of curious facts and inferences about this case that fit aspects of Discontinuity Theory. First, all portions of the “Memoirs” related to Schreber’s childhood were deleted by Schreber’s editors prior to publication. The second related curiosity is that Freud exercised remarkable “analytic restraint” in not using a major source of available information about Schreber’s childhood, namely the voluminous writings of his father, Dr. Daniel Gottlieb Schreber. His father was the preeminent pedagogue of his era. He authored 18 books that were widely read and reprinted, covered a wide range of topics, but primarily focused around hygiene, physical culture, and rigorous educational systems for parents and teachers to raise strong, disciplined, obedient, and moral children. He was the medical director of an orthopedic institute in Leipzig, and the founder of “therapeutic gymnastics” throughout Germany, and of Schreber Associations, which Freud clearly lauded in his writing (1911). So great was his influence, that one biographer said of Dr. Schreber, the father: “Every age produces a man who expresses its spirit as if with the power of Providence . . . the generation of our century demanded and created a man like Schreber” (Politzer, 1862, p. 2). Schreber, the mad son, suffered from painful bodily experiences which he wrote were “miracles” performed on him by “God,” who sent “rays” upon his body. However, these were malevolent miracles, “of such a threatening nature that I thought I had to fear almost incessantly for my life, my health, or my reason” (Schreber, 1955, p. 131). He had sufficient insight to suspect that others reading about these miracles would doubt his sanity because they “will naturally sound extremely strange to all other human beings, and one may be inclined to see in it only the product of a pathologically vivid imagination” (p. 132). However, his insight was blindsided when it came to identifying the true source of his lifelong suffering-most obviously Dr. Daniel Schreber, his father. Extensive research by William Neiderland (collected in Kanzer & Glenn, 1980), and elaborated by Morton Schatzman (1973), compares the published child-rearing doctrines of Schreber, the father, with the persecutory delusions of Schreber, the son. For each of his “supernatural miracles” of head or chest compression, of heat or cold, of being tied to earth, or fastened to rays, there is a direct correspondence with a physical device or special treatment recommendation in his father’s writings about how to educate children to be physically and morally straight. Drawings of these devices
394
PHILIP G. ZIMBARDO
resemble tools of the medieval torture chamber more than of the classroom. Nevertheless, the father proudly reported testing each of them on his own children, some beginning as early as when they were three months old. Of the many interesting aspects of this case, four deserve special mention here. First, Dr. Schreber’s goal of training a child to be “unconditionally obedient” to authority was carried to fruition later on as a primary feature in Hitler’s Nazi ideology. It also presaged George Orwell’s “doublespeak” in 1984. The elder Dr. Schreber, wrote that “unconditional obedience was nobly independent,” “submission is freedom,” and “self-control is control by parents” (see my analysis of the psychological dimensions of 2984, Zimbardo, 1984a,b). The second point of direct relevance to Discontinuity Theory is the son’s inability to indict his father as his abuser and tormentor, selecting instead a narrative that makes God’s malicious miracles the agent or agency of evil, and the cause of his madness. The mad son comes tantalizingly close to acknowledging the sadistic abuse by his father, time and again in the Memoirs, but then deduces a safer source instead. Interestingly (given the use of hypnosis as the experimental paradigm in my research described in the next section), son Schreber compares his nervous state of what he calls “soul murder” to that of an hypnotic subject whose own will power is imprisoned by the hypnotist who takes possession of his soul. He concludes, “At a time when my nervous illness seemed almost incurable, I gained the conviction that soul murder had been attempted on me by somebody” (p. 55). Not by some God, but by some body, namely, the person who was treated by his contemporaries as the god of pedagogy. The third point is the relevance of this case to understanding one of the dynamics in current cases of recovered memories of early childhood abuse. It speaks to the powerlessness of children to speak up against powerful adults who abuse them, and their lack of societal support when society does not want to believe such horror stories about respectable parents. Finally, what should we make of Freud’s failure to recognize the obvious links between this Father’s tyrannical training methods that he applied to his children, and the Son’s corresponding symptoms of mental disorder? I believe it reflects Freud’s theoretical bias that motivated him to overextend the notion of repressed infantile psychosexual fantasies-over the realities of parental abuse, and actual seduction of children-to apply it to his analytic patients, in explaining hysteria and obsessional neuroses. (For a fuller account of this provocative dimension of Freud’s deceptive accounts of his psychoanalytic reconstructions, see Esterson, 1998.) We are also forced to note Freud’s unquestioning acceptance of the public esteem of Dr. D. G. Schreber (perhaps the same reason editors deleted those un-
DISCONTINUITY THEORY
395
flattering memoirs dealing with Daniel Schreber’s childhood). Such esteem is frankly surprising, when that widespread acceptance in Germany of Dr. Schreber’s child-rearing strategies and tactics, and his lofty reputation as an esteemed educator, are weighed against these stark facts: one of his sons had to be institutionalized for his madness, and his other son (curiously also named Daniel Schreber) committed suicide at an early age.
B. MISATTRIBUTIONS O F ORGANIC, PHYSICAL SOURCES TO PSYCHOLOGICAL SOURCES O F DISCONTINUITY Many physical illnesses with an organic basis also have symptoms that show up as behavioral, cognitive, or affective disturbances (such as hypothyroidism, pernicious anemia, or brain tumors), while others worsen an existing mental disorder. Without adequate medical examinations, patients may be misidentified as having a psychiatric illness when the cause of their problem is a disease o r organic impairment. They may be treated by psychiatrists, clinical psychologists, and even sent for treatment to mental hospitals when their primary problem is an undetected physical one, requiring medical care. Strong evidence for this point comes from a study that administered thorough medical evaluations to more than 500 patients from a variety of California mental hospitals. The researchers found that many of these mental patients had active, important physical diseases, a large percentage of which were not detected. “The mental health system had recognized only 47% of study patients’ physical diseases, including 32 of 38 causing a mental disorder, and 23 of 51 diseases exacerbating a mental disorder” (Koran et. al., 1989, p. 733). If health-care professionals so often fail to identify the origin of an unusual psychological reaction as due to physical causes, we can assume that many lay individuals make the same mistake. Let me illustrate some of the consequences of this failure to identify the physical, organic origin of a discontinuity that instead is experienced as psychologically induced, in four case examples. 1. The Case of Anna 0: Hysteria or Tuberculosis?
Freud and Breuer used the case of Anna 0. as the basis for their earliest diagnosis of psychogenic illness as an hysterical conversion reaction (Breuer & Freud 189Y1955). Among her various sensory and motor problems was a severe cough, which began when she was taking care of her sick father. Diagnosed as a “nervous” cough, its dynamic origins were explored under hypnosis with Anna 0.as the first patient to ever experience
396
PHILIP G. ZIMBARDO
the new “talking cure.” However, this extraordinary women, Bertha Pappenheim, who became a leader in women’s rights and a pioneering social worker, terminated her unsuccessful analytic treatment with Breuer and entered a sanitarium. She was later released largely free of her initial symptoms-of tuberculosis. It is most likely that she had been suffering from an organic illness contracted from her close contact with her father and not primarily a psychological one at all, although she may have suppressed her frustration about having to care for her father so long and about the lack of opportunities for woman. It is ironic that the vital concept of psychogenic illness was developed from a case of organic origin that was misconstrued as psychological (see Rosenbaum & Muroff, 1984; Thornton, 1984). Montaigne (Essays 1580/88) told us long before the psychodynamicists, that, “The mind has great influence over the body, and maladies often have their origin there.” But we might add, which There? 2. M y Colleague Confuses Lack of Sleep with Lost of Interest in Psychology
A colleague known for his dedication to, and love of, psychology, recently began to question whether he should stop doing research or maybe even change his field. That conclusion was based on his apparent loss of interest in research his students were presenting during lab meetings or colleagues during colloquia. The evidence he used in making this inference was his yawning, and overwhelming feelings of sleepiness, taken as symptoms of being bored. It was not until his wife detected he had a sleep disorder, which was verified by the local sleep disorder clinic as sleep apnea, that he received the right treatment for it. With that change in attribution, he resumed his overproductive research career. 3. Old Joe Goes Ballistic on the Pharmacy Line
In another personal case, an elderly man in front of me on the waiting line for hospital prescriptions told the nurse he was late for work and needed his medication as soon as possible. She turned from him to ask the pharmacist how soon he could get his prescription filled, and before turning back to help me as next in line, she said they would rush his order. As she walked back to the waiting line and began talking with me, Old Joe began to scream at her, shouting obscenities and racial epithets about his repeated lack of respect by her kind of people. Naturally she got furious, and started to yell back at him. I grabbed his shoulder and said in a loud voice as he
DISCONTINUITY THEORY
397
stared at me, “She is doing you a favor, and will get your medicine right away, just sit down and wait a few minutes.” Old Joe calmed down, apologized, and the tension was broken, and the line moved on. I realized that Joe had not heard the nurse talking with the pharmacist when her back was turned to him, so he must have a hearing disorder. My remedy was simple action to make it loud and clear to him that he was getting what he asked for. But it was evident to me that from his perspective of not having heard her solicitous exchange, he was being disrespected. It must happen enough for Joe to have formed a categorical conclusion about Black nurses who act in a hostile way toward him. His ultimate remedy is, of course, wearing a hearing aid, but there are many reasons why people do not use them even when they realize they are hearing impaired. This is the stuff of paranoia, reasoning correctly from an initial misperception, and the basis of one of our early experiments (Zimbardo, Andersen, & Kabat, 1981), elaborated on in the next section of this chapter. 4. Terry Paints a Mad Portrait of Himself at Bud Plastics
While the last two examples ended with positive outcomes, a tragic conclusion emerges from another instance of mistaking an organic problem as a psychological one. The Bud Plastics Company in Ohio makes the bodies for Corvettes using the help of many different union workers to handle various aspects of its manufacture. A few years ago, individual workers began to experience a variety of severe symptoms, memory loss, emotional volatility, blind rages, and exhaustion that made them sleepy all the time. Many of the men were quite young, in their twenties, and had been perfectly “normal” when these bizarre discontinuities started to surface. Neither they nor their spouses could understand what was happening, but it was sufficiently weird and extreme that they tried to conceal their symptoms from others. As their conditions worsened, they got medical exams, which were negative, so were referred for psychiatric evaluation. Some ultimately went into therapy, and others were sent to a mental hospital. But nothing helped improve their condition. It was not until a relatively large number of these workers reported sick that a union supervisor took notice of a potential workplace issue. The resulting investigation uncovered a tragedy in the making. These men were suffering from incurable neurotoxic brain poisoning caused by their chronic exposure over many years to low levels of neurotoxic substances in the paints and plastics they were using. OSHA safety tests and evaluations of these substances had not included long-term exposure to substances that had been proven benign
398
PHILIP G. ZIMBARDO
in briefer exposures (reported on National Public Radio, “Chemicals in the workplace ,” 10/21/I985). This case raises several issues relevant to Discontinuity Theory. First, it is possible to misread a physical cause as a psychological one and to be mistreated for it if the attendant symptoms are behavioral or psychological. Second, discontinuities in behavioral reactions demand explanations not only by the Actor, but also by all those personally connected to him or her, especially when they are obvious, atypical, and cause social shame. Third, when many individuals are experiencing the same symptoms but do not disclose them to others, they create a state of pluralistic ignorance that fosters dispositional attributions, “what is wrong with me?” (Prentice & Miller, 1996; Wetzel & Walton, 1985). That orientation leads them to seek individualized medical treatments, rather than advancing situational attributions of “what is wrong with the job?” that is affecting many workers. Doing the latter, calls into action a public health approach rather than a medical one. This case led me to conduct a study on disclosure of a discontinuity among students in a dyad, each experiencing similar symptoms of unexplained arousal, to be presented subsequently. A final instance of the interaction of organic, medical conditions and psychiatric complications comes from an old study that first chronicled the development of psychopathology in previously mentally normal patients following their open-heart surgery (Freyhan, Giannelli, O’Connell, & Mayo, 1971). Every one of 150 patients needing cardiac surgery, over a three-year period, was referred to preoperative psychiatric and EEG evaluation, and was then followed up during postoperative recovery using interviews and a wide spectrum symptom profile. Open-heart surgery is obviously stressful, and the associated experiences of mortality represent a unique kind of discontinuity for the patient. Nevertheless, the data from this carefully controlled study are astonishing. Over 50% of the patients manifested psychiatric complications postoperatively, which the authors say is “probably a conservative estimate” for a number of reasons. The “symptoms resemble a kaleidoscopic view of mental illness’’ (p. 187), with 42%of the patients being symptomatic for more than 3 weeks, 75% rated as moderate or severe in severity of the psychiatric symptoms, and 59% having multiple syndromes-delirious syndrome, paranoid-hallucinatory syndrome, and/or mood-disorder syndrome. Of relevance to one of my experiments (to be reported on the generation of paranoia in normal subjects experiencing unexplained deafness) is the observation made by these psychiatrists and physicians: “Patients developed typical ideas of references, suspecting that nursing and medical staff were secret agents observing and recording their activities. The monotonous
DISCONTINUITY THEORY
399
noises of electronic monitoring equipment lends itself to such notions particularly in individuals whose cognitive functions are already impaired.” (p. 189)
C. IGNORANCE OF TRUE SOURCE O F DISCONTINUITY REPLACED BY PREVAILING EXPLANATIONS Sometimes the inability to generate the correct causal attribution is the result of current lack of public knowledge about the nature of the true causal factors involved. When the discontinuity is severe, and shared by many people, alternative hypotheses will center around whatever prevailing theories are accepted in a given community. Many of history’s most bizarre and inexplicable societal episodes may fall within this category. 1. Was Salem Witchcrafi Caused by Eating LSD-Laced Rye Bread?
Although witchcraft persecution trials were a widespread effect of the European Inquisition, they occurred in North America in only one town (Salem), in only one year (1692). The set of intense symptoms that flared up in a group of farm girls (fits, spasms, tremors, feeling pinched or bitten, hallucinations-the discontinuity to be explained) was explained by the prevalent religious/demonic interpretation (Hansen, 1969). A likely alternative explanation appears to be food poisoning, ergotism, a fungus that grows on rye grain and produces a natural form of the hallucinogen, LSD. This interpretation was first proposed by psychologist Linnda Caporael (1976), and supported with additional archival data by historian Mary Matossian (1989). Most of the reported symptoms of bewitchment are identical with those of ergotism, and also were found among Salem’s farm animals. Available evidence indicates that the cold, wet climate that year favored growth of this fungus that was ingested by poor farm families in their heavily rye diet. Matossian argues persuasively that serious outbreaks of such microbiological contamination of the food supply have played a significant role in shaping social behavior, and thus explains many social discontinuities recorded throughout Europe for hundreds of years. Wherever peasants relied on a rye-based diet, and the climactic conditions generated fungus poisoning, mass reactions to the shared discontinuity of their bizarre symptoms led to revolutionary riots in 1789 by French peasants gripped by “the grand fear,” religious revivals in the 18th century, and many other social epidemics. This example illustrates another way in which a discontinuity experienced at an individual level can have social psychological implications when aggregated across a population.
400
PHILIP G. ZIMBARDO
Incidentally, the Inquisition was given its impetus by the publication in 1486 of a theological treatise on The Witches’ Hammer or The Malleus Maleficarum (Kramer & Sprenger, 148611971). It is said to have been on the bench of every judge and the desk of every magistrate in both Catholic and Protestant countries of Europe that were driven to combat the “Society of Witches.” Curiously, its distorted reasoning begins with reflection on a cosmic discontinuity, namely the existence of evil in a world created and controlled by an omniscient, omnipotent, God. By turns it focuses on human agents of the Devil, witches, who facilitate evil by undermining human virtue and resolve, thus must be found and destroyed. D. DISCONTINUITY-MOTIVATED SEARCH FINDS CAUSALITY IN CHANCE, CONSPIRACY IN COINCIDENCE As nature abhors a vacuum, the human mind abhors chance in human affairs. It finds patterns in randomness (Chapman & Chapman, 1967,19691, order in chaos, meaning in nonsense, causes for correlations, and planned conspiracies behind coincidental events. Thus, we should not be surprised that when a person has experienced some significant discontinuity, she or he will be readily able to generate a “reasonable” causal explanation, and sustain it despite all evidence to the contrary. When that erroneous explanation leads to social action, it could become dangerous. 1. Mrs. Lincoln Fears That I A m Controlling Her Mind Mrs. Lincoln was a middle-aged, African-American housewife, whose apparent discontinuity experience was hearing voices of several people talking to her. At first, she reacted to them with benign interest, curiosity, and some amusement. But when she started acting out their suggestions, she got in trouble at work, first losing her job at the post office (when the voices told her to take frequent breaks), and later as a bus driver, when the voices suggested new routes for Mrs. L. to follow. She was very concerned about getting her job back because she had a family to feed, and so came to see me at Stanford University to implore me to intervene-since she believed I had been responsible for starting up her voices! Prior to our initial conversation, she had sent me a New Year’s greeting card, had learned much about my background and family, and sat patiently outside my office all day for many days (during winter recess) until my secretary noticed her there and made a formal appointment with me. Our conversation began with her asking whether I taught a course on mind control, which I did, and then whether research was being done at
DISCONTINUITY THEORY
401
Stanford University on mind control, which I said was not likely, given the recent ethical guidelines against any such research. Nevertheless, Mrs. L. insisted that I was somehow controlling her mind, was responsible for the voices she was hearing, and wanted me to silence them so that she could get her job back at the post office. Despite my reasoned arguments, she insisted there was a definite link between her voices and being a subject in mind control research at Stanford. The meandering link was created from the following circumstances. She had come to the university earlier for diagnostic screening of her eligibility for inclusion in shyness treatment groups that we were starting at our Shyness Clinic. After taking a battery of psychological tests and a brief interview, she was told by the therapist that she was not really shy enough to warrant therapy. Five months later, her voices began to speak up, and Mrs. L. associated this unusual experience with the only other recent unusual event in her life, visiting Stanford University and being studied by psychologists under my supervision. This “ego-correlation,’’ pairing personally experienced unusual events, became ever more viable the more she thought about it. This belief had persisted for a full seven years, up to the time of our first appointment, when she wanted to know my intention in implanting those voices in her mind (see Malle, in press). She returned repeatedly, sitting outside my office for hours on end, always with a smile, but refusing to leave until she spoke with me again. I arranged with a female Afro-American clinical psychologist to talk with her, who encouraged her to see her therapist and take her medications, and provided rewards for doing so, like taking her to lunch. Nevertheless, Mrs. L. insisted on my writing a letter of recommendation to the post office, which I finally wrote, sending her a carbon copy. However, because her copy was unsigned by me, she got angry believing the original was also not signed or even sent. Mrs. L. then refused to leave my office until I wrote a better letter, signed and mailed in her presence, that got her job back. At that point we all became concerned about the potential for violence and informed her therapist that he had to talk her into leaving or university security would intervene. I never heard from her again, but have not forgotten Mrs. L. and what her voices might tell her to do next. At about the same time as Mrs. L. appeared and disappeared, a colleague in the Math Department, Karel deleeuw, was brutally murdered by a graduate student as he sat working in his office. His killer, Theodore Strewleski, was “unnaturally sensitive to what he called ‘put downs,’ and once he felt someone had insulted him, he brooded on it for years” (Ciotti, 1980, p. 42). Strewleski’s discontinuity may have been the failure to complete his dissertation 18 years after starting out as a promising graduate student
402
PHILIP G . ZIMBARDO
in an outstanding program. Professor deLeeuw was but one of many faculty on Strewleski’s list to be murdered for their arrogance and show of disrespect to students. Thus, sometimes discontinuities can lead from thoughts to feelings to actions, in some cases, destructive ones. A sad, remarkable parallel to the Strewleski case occurred recently when another graduate student confessed to fatally shooting his engineering professor as he was in the process of collecting final examinations at Wayne State University (Reuters News Service, 1998). A colleague hinted at a possible motive, saying, “The math is extremely difficult. (The suspect) was just unable to do it. He did not know how to evaluate himself correctly. He wasn’t at all able to get the degree.” (p. A2). This 48 year-old student, turned premediated murderer, was also like Strewleski in having been around his university for many years, unsuccessfully trying to graduate with a doctoral degree. Another illustration of a chronic discontinuity with lethal consequences.
E. SOCIAL SEARCHES FOR NORMALITY MAY END IN BIZARRE PLACES WITH STRANGE BEDFELLOWS People experiencing discontinuities that have a noticeable effect on their behavior may become more concerned over their apparent abnormal appearance than in generating an adequate rational explanation for it. Regardless of the origin of the discontinuity, if people feel that it is affecting their behavior and demeanor sufficiently to mark them as “different” by their usual reference groups, they may seek out new bases of social comparison with similarly deviant others. The first case illustrating this social aspect of Discontinuity Theory has a medical basis for her suddenly aberrant behavior that drove a conservative college student to become a punk rocker. The second reveals what can happen in a similar situation when the person experiencing a discontinuity has sufficient personal power to persuade others to act in ways that normalize his discontinuity. 1 . Mary Jane, The Punk Rocker Graduate Student, Needs Sugar
As an undergraduate, Mary Jane began experiencing sudden, dramatic mood swings from anger to rage to depression to elation within a few hours. She was easily irritable, temperamental, and hyperkinetic at times alternating with listless apathy in the mornings. Her dorm roommates found her hard to deal with since she became hostile and argumentative as well, but as suddenly could function normally and be sweet and happy.
DISCONTINUITY THEORY
403
She first attributed these atypical reactions to her sister’s death from a drug overdose since they began shortly after that salient event. It is significant to note that she also had started a vegetarian diet at that time to cope with her feelings of listlessness, and would faint during religious services if she fasted. Mary Jane began to worry about “what’s wrong with me?” as her anomalous reactions persisted for a year after a reasonable mourning period. Her mother interpreted her negativity as an “attitude problem” and referred her to a therapist, a family friend, who also focused his causal analysis on her mixed feelings regarding her sister’s death. She “bailed out” of therapy when her symptoms continued to worsen. Her family felt that she was “creating problems for herself,” and that her problems were obviously mental, even though her father was a physician. Her roommates got increasing irritated by her staying up all night reading and studying then sleeping in all morning, not to mention her abusive behavior that got her labeled as “weird.” She also started smoking, along with drinking lots of coffee and alcohol, and began to get interested in hard-core punk rock music and rockers. She attended concerts regularly (X, Sex Pistols, Chaos), would slam dance, get battered, punch others, scream obscenities, and be physically exhausted, but then would sleep deeply and wake up feeling better. She also donned punk attire totallyspiked Mohawk, leather, chains, studs, piercing, and other regalia. Mary Jane then gave up entirely on her former friends, and associated exclusively with punk rockers who, like her, were angry and existentially depressed. They needed no explanations nor gave any, and she fit in perfectly like her cut-down glove. A t that point in time, she made a character attribution, “I am just an angry person.” At the end of senior year she outgrew the punkers, realizing that she hated that kind of music and joined a new social set which embraced total apathy, negativity, and depression. She reported later basking in their depression and being “bummed out,” feeling no affect at all, and not needing to explain why to anyone. They listened only to “gloom and doom” music (like “Care” and “Joy Division”), dressed totally in black, and were united in their anti-establishment social activism. She felt that finally she was able to “get away from Me; I was a Group now!” Despite this lifestyle, Mary Jane continued to focus on her studies, got energized by studying, enjoying snacking a lot as she did, and got good enough grades to be admitted to Stanford University that fall. When I met her, Mary Jane was still in a socially alienated mode, and clearly “different” from our traditional, rather conservative graduate students, but obviously smart and surely “interesting.” She had just discovered that her vague associations of her mood swings and diet “sugar rushes,” “starch highs,” were the result of her chronic, severe hypoglycemia. Her abnormally low
404
PHILIP G . ZIMBARDO
blood sugar levels were worsened by a vegetarian diet, by fasting, coffee, alcohol, and a long night’s sleep. They were elevated by snacking junk food. Although her father was a physician and her mother also had been hypoglycemic, they did not interpret her obvious medical symptoms accordingly because of the vividness of the alternative explanation of grieving over her sister’s death. With a new diet and new interpretation, Mary Jane began to change her dress and demeanor to resemble that of a more typical graduate student. Over time she performed well enough to earn her a fine job as an assistant professor in a traditional liberal arts college. In retrospect, as we discussed her earlier behavior, she interprets joining the punkers and the nihilists as “social facilitators” that immediately established her unusual reactions as usual, made explanations unnecessary and provided needed social affiliation to offset rejection by former friends. The association with these particular groups also served to exchange an external focus for her prior egocentric preoccupation and failed internal search mission for an adequate explanation of the discontinuities she was experiencing. I would argue that a comparable social psychological process may be at work among many ordinary individuals who join various social/religious/ political cults (Zimbardo, 1997). Steven Hassan, a former member of the Unification Church (“the Moonies”) and author of Combatting Cult Mind Control (1988) informed me that the majority (as high as 75%) of those recruited into cults are experiencing some form of “discontinuity” in their lives, often of a spiritual or mystical nature, but it may also come from a personal loss (personal communication, January 9, 1995). At times the attraction of cults may derive in part from the group’s unusual demands for extreme behaviors that are alien to the usual lifestyle of the recruits, thereby allowing them to submerge their feelings of differentness or inadequacy within the cult’s new normative structure. 2. Prophet Applewhite Castrates His Followers on Their Way Through Heaven’s Gate It was strange enough for the world to learn that 39 members of a cult had committed mass suicide in a posh mansion in San Diego, California, March 26, 1997, but the story got ever stranger as more information was uncovered about the Heaven’s Gate cult (Chua-Eoan, 1997; Newsweek, 1997). This group, composed of highly educated, technologically sophisticated members (with a group web site), varying widely in age, were totally obedient to their leaderlprophet, Gordon Applewhite. He imposed a rigid lifestyle, with detailed rules and routines that controlled their every action,
DISCONTINUITY THEORY
405
so as to “prepare their seed for harvesting” when they were transported beyond earth to Heaven’s Gate by a UFO. As they waited patiently for a celestial sign, they first assumed their signal for imminent ascendance was the movie Cocoon, with its UFO transport ending. But when that failed to materialize, they later found a better signal in the advent of the wellpublicized comet, Hale-Bopp. Applewhite, a.k.a. “Leader Do,” told his followers that a hidden UFO was trailing behind the comet and would be their designated heavenly transport. To be carried out of this world, they first had to give up their corporal bodies in their well-choreographed group suicide (Gleick, 1997). My special interest in this case stems from the leader’s social power to rectify a discontinuity that he was experiencing by imposing a severe demand on his followers that normalized his unacceptable impulses. Applewhite is reported to have struggled for years against the emergence of his homosexual impulses that were unacceptable to him. A central goal of leader Applewhite was to eliminate the negative “vibrations from the vehicle” (body), primarily by controlling sensuality in all of its forms. He advocated first abandoning gender distinctions, then all sexual behavior, even masturbation (Taylor, 1997). Later, when apparently that was not sufficient for him, he urged his male followers to “remove the battle” they were fighting within by castrating themselves, as he did. Many of his blindly obedient followers also had themselves castrated (CNN Special Report, 3/30/97; Miller, 1997; Roberts, Hollifield, & McCarty, 1998).
F. CHRONIC, SHARED DISCONTINUITIES CAN LEAD TO MASS SOCIAL REACTIONS The phenomenon of Cargo Cults was presented earlier as an instance of a widespread societal reaction to the perceived discontinuity caused by the sudden disappearance of the abundance that had been created by the sudden appearance of foreign ships. Part of that analysis includes a socially shared perception of the discontinuity, and a common way of responding to it that is socially acceptable in that setting, even when it entails bizarre reactions. Our final case study involves a socially shared pathological reaction among children who earlier had all experienced a similar kind of personal discontinuity. 1. Mass Hysteria among Schoolchildren Experiencing Chronic Loss
and Family Disruptions An outbreak of illness spread swiftly among a large group of schoolchildren attending an assembly at an elementary school in a Boston suburb
406
PHILIP G. ZIMBARDO
(Small & Nicholi, 1982). Thirty four of the 224 children at that school function had to be hospitalized with severe dizziness, hyperventilation, headache, nausea, abdominal pain and general weakness. However, extensive epidemiological investigation failed to detect an organic cause, and since the symptoms went into remission as quickly as they surfaced, and were more common among girls than boys, the investigators concluded it was a case of “mass hysteria.” When a sixth grade boy, a class leader, fell from the stage during the play being performed for the assembly, immediately several other students became ill and in minutes clusters of children throughout the auditorium were similarly stricken down. Two aspects of this case are particularly relevant to our theory. First the initiating setting was the last gathering of the student body for the year, with the sixth graders moving on to new schools, leaving some of their old friends behind. This may have been perceived as a significant loss to them that could not be individually expressed since graduating was supposed to be a joyous event. Second, evidence was found that the majority of children hospitalized had suffered a previous loss that may have made them vulnerable to the current loss and predisposed those children to mass hysteria. Comparisons of the hospitalized children with their nonhospitalized peers revealed that they had a much higher incidence of family disruption, assessed as significantly higher rates of parental divorce and death within the family. These sources of discontinuity loom large in a child’s life. They may be stored as chronic perturbations in consciousness that can influence behavior in settings that share in symbolic significance, in this case, an anticipated new source of loss. Having outlined and elaborated on features of Discontinuity Theory and illustrated some aspects of it with a variety of case study examples, we are ready to turn to a series of experiments conducted over many years to examine its empirical support. Several of the studies have been published, but some additional data will be presented here, while several more are unpublished and will be presented for the first time.
IV. Experimentally Creating Discontinuities, Search Frame Biases, and Their Consequences This section begins with a summary of the basic paradigm that guided each of the series of experiments that will be elaborated subsequently. I then describe in more detail the ways in which hypnosis was utilized as our methodological tool for inducing the experience of discontinuity. As part of this overview, it is helpful to lay out other common features of our
DISCONTINUITY THEORY
407
research, such as the assessment of the psychological and physical status of our participants, the procedural sequence and standardized controls used, the types of outcome and process measures recorded, how the extensive debriefing phase was developed, and the follow-ups conducted.
A. BASIC EXPERIMENTAL PARADIGM Earlier research from our laboratory that attempted to replicate and extend the conclusions of the Schachter and Singer (1962) two-factor model of emotion revealed that the discontinuity created by hypnotically induced unexplained arousal (Maslach, 1979) had many advantages over the discontinuity induced by misattributions of arousing drugs, such as epinephrine (Marshall & Zimbardo, 1979; Maslach, Zimbardo, & Marshall, 1979). These advantages are listed in Table 111. I am aware of the potential limitations and controversy of using posthypnotically induced discontinuity experiences and search frame biases. Hypnosis is still not well accepted among mainstream researchers, nor are the mechanisms by which it works well understood. Nevertheless, I decided to adopt that procedure for my basic paradigm because I felt confident in its research utility for testing the predictions from my conceptual model based on my personal familiarity with hypnosis along with successfully using it over many years professionally (see Marshall & Zimbardo, 1979; Maslach, 1971; Maslach, Marshall & Zimbardo, 1972; Monteiro & Zimbardo, 1987; Piccione, Hilgard, & Zimbardo, 1989; Zimbardo, 1969;Zimbardo, Marshall, White & Maslach, 1973; Zimbardo, Maslach & Marshall 1971,1972). I have also relied on the successful use of hypnotic programming of research subjects by Blum (1979), amnesia studies by Evans (1979) and by Kihlstrom,
TABLE 111 RESEARCH BENEFITS ASSOClArED WITH HYPNOTICALLY INDUCED AROUSAL COMPARED TO AROUSAL BY EPINEPHRINE INJECTIONS Arousal effects are equally strong Onset and offset are immediate and stimulus cued Less variability across participants within the same condition Can be limited to specific and minimal arousal symptoms Effects do not carry over beyond the research setting Less suspicion and disbelief among participants Does not require elaborate and costly medical cautionary procedures (demanded by Medical School Panels for Human Research)
408
PHILIP G . ZIMBARDO
Evans, Orne, & Orne (1980); and the knowledge gained from my long association with Jack Hilgard’s hypnosis laboratory at Stanford. The major relevant criticism to contend with is the subject-selection factor, since only about a fourth of the population is highly hypnotizable. I’ve tried to deal with that constraint by randomly assigning these subjects to the various conditions in each experiment, using within-subject controls in one experiment, hypnotic simulators in some studies, and by comparing high and low hypnotizable subjects in another study. In this program of research, discontinuities were operationally defined as experiences of “unexplained arousal” induced in highly hypnotizable subjects by means of posthypnotic suggestions to feel changes in their physical functioning, (e.g., increases in heart rate and respiration) at the sight or sound of a pre-determined cue. Suggestions for source amnesia in the experimental treatment added the “unexplained” component to the standard arousal manipulation for these participants, while controls remained aware that their arousal was hypnotically induced. A second posthypnotic suggestion was used in some studies to provide participants with explanatory frameworks to guide their attributional search for the meaning of their aroused state. These “biased” search frames included other people (social), the physical environment (external), or their bodylhealth (personal, internal). Following the standardized hypnotic induction and suggestions, participants continued to engage in various prearousal tasks, then received the arousal cue, and then continued to work on their task alone, or sometimes with another peer or confederate. Participants completed personality scales and other measures during, and sometimes after, this discontinuity experience. In some studies, they were also wired for psychophysiological recordings. While still in a state of arousal, each participant was interviewed by the experimenter or a clinical psychologist (blind to their experimental condition) to assess her or his judgments, attributions, emotions, and memory. Subject reactions were sometimes videotaped for later analysis, rated by judges, also by clinical psychologists, and made available to our Human Subjects Institutional Review Board on request. Extensive debriefing ended the experimental session, which typically lasted about 90 minutes. 1. Participant Characteristics
Our research participants were Stanford University undergraduates (assumed to be both intelligent and well educated), who were recruited from Introductory Psychology courses and given experimental course credit and modest payment for their services. Prior to their research involvement, we determined that they were normal and healthy, both physically and
DISCONTINUITY THEORY
409
psychologically. We did so by means of interviews and self-report questionnaires that revealed they all were not in therapy, or on psychiatric medication, had no current medical or psychological problem, were not anxious (all were within one SD of the nonclinical norm on the Manifest Anxiety Scale), and showed no evidence of paranoia, phobia, or hypochondriasis on relevant Minnesota Multiphasic Personality Scale (MMPI) subscales (within the normal range). Finally, all had to report feeling “good, well, or great” at the start of any experiment to be included during that session. 2. Hypnotizability and Hypnosis Enhancement Traning Participants were selected on the basis of their scores on measures of hypnotizability (using a slightly modified, shorter version of the Harvard Group Hypnotizability Scale, Form A, Shor & Orne, 1962, and in some studies, also individual assessment via the Stanford Hypnotic Susceptibility Scale, Form C, Weitzenhofer & Hilgard, 1962). Two groups were used: highs scoring 8-10, and lows scoring 0-2. In addition, it was necessary for those who were highly hypnotizable to also demonstrate their ability to experience posthypnotic amnesia, as shown by high scores on the amnesia section of the above scales, and validated in specific tests during hypnosis training sessions. Hypnotizability is a reliable individual difference measure of responsiveness to hypnotic suggestions to experience a range of phenomena not typical in ordinary consciousness (Hilgard, 1965). It is this ability that creates the hypnotic experience and not some special talent of the hypnotist, who is merely the coach or guide. Neither gender differences nor any valid dispositional correlates have been found for hypnotizability (Hilgard, 1982), although it does decrease with age and education. Hypnotizability is a highly reliable trait, as shown by our 25-year test-retest correlation coefficient of .71 (Piccione, Hilgard, & Zimbardo, 1989). A special feature of this research program was the use of a training procedure designed to enhance hypnotizability by enabling participants to enter a deep hypnotic state quickly, by means of instructions administered by me in person or via tape-recording, or later on by self-administration. During the training, participants attended one or two small group sessions (3 hours of training in total). Among the features of this training was practice by participants in rapid induction, deepening of hypnosis, and the experience of a wide range of hypnotic phenomena, such as sensory imagery, positive and negative mood arousal, amnesia, carrying out posthypnotic instructions, age regression, pain control, self-esteem boosting, and concentration for studying (only some were relevant for our research; the last four were for the personal benefit of the students). This training also included
410
PHILIP G. ZIMBARDO
repeated self-ratings of subjective arousal on a 10-point SUA scale (Subjective Units of Arousal, where 0 = totally relaxed and 10 = extremely tense), a measure used later in all our studies. Participants also received training in how to write about and discuss their reactions during each phase of the hypnotic training. It was important that participants realized that they were able to communicate their feelings and thoughts while deeply hypnotized or during a posthypnotic phase, so that this knowledge could be transferred to the experiment proper, when they were actively engaged in various tasks, or in debriefing. To control for this pre-experimental exposure time and contact with the researchers, comparable training was used for low hypnotizable participants in the one study where they were included. For hypnotic simulators (used in another study), the preliminary small group sessions, also held in the same laboratory, included watching a video on hypnosis, discussion about hypnosis phenomena and research, and experience with induced relaxation and role-playing being hypnotized.
3. Common Procedural Features To save space in describing the methodology for the individual studies that follow this section, those procedural features common to all studies will be outlined here, so that only special aspects particular to those studies need be mentioned later. All participants were randomly assigned to experimental and control treatments. The treatments included posthypnotic suggestions for somatic arousal (cued to a specific stimulus) with or without explicit amnesia, or awareness, of this suggested arousal source. The arousal cue was varied between studies to minimize any specific associations of a particular stimulus, and thus to increase its generic value. The cues included a bell sounding, a timer being started or stopped, or a specific word projected on a slide or on a memory drum. Hypnotic inductions and experimental treatments were standardized and presented via tape recordings, written instructions, or both. Instructions for somatic arousal were limited to the two basic symptoms of most arousal states, increased heart rate and respiration. Pretesting with other, additional symptoms, such as tense muscles or moist palms, revealed no greater effect than did these two basic symptoms on any of our assessments. Interestingly, many subjects spontaneously reported feeling those, and other symptoms when they were “generically aroused.” Either the entire procedure, or final interview, was video-recorded, both for data analysis and so that the Human Subjects Research Committee could monitor the research protocol. The Amnesia Suggestion stated, “You will NOT remember the reason you are feeling aroused, and will have no desire even to try to remember
DISCONTINUITY THEORY
41 1
the suggestion made to you. You will feel aroused and not associate it with the suggestion you got.” The Awareness Suggesrion stated: “Of course, you will remember fully that the reason you are feeling aroused is because of this suggestion 1 made to you to alter your heart rate and respiration. You will feel aroused and associate it with the suggestion as you try to understand the reason for your feelings of arousal.” The engagement of an explanatory search process was encouraged by further instructions in the Amnesia condition, to the effect: “However, it will be important for you to try to figure out why you are feeling as you are, and to try to understand the reason why you are feeling aroused.” Experimentally induced search frames were added to the above instructions in some studies, by inviting the individual to consider the possibility that what she or he was experiencing “might have something to do with other peoplehhe physical environment/your body or health.” Obviously any subject was given only one of these search frames to consider using during the attributional search process. In pretesting, we also explored the effects of the induced biased search frame of “with mystical or spiritual forces.” A wide variety of assessment measures were used in various studies, including self- and peer-ratings (especially subjective units of arousal [SUA] ratings); videotaped behavior ratings; Thematic Apperception Test (TAT), Rorschach, MMPI, emotion circle, interpersonal circle ratings; psychophysiological measures of heart rate, respiration, galvanic skin response (GSR), electromyography (EMG), and EEG. Each study concluded with a long three-phase debriefing of participants both in and out of hypnosis. Subjects had to make written and oral statements of their understanding of the nature of the experiment and the manipulations; they were dismissed only when they gave evidence of full understanding; and they had to have returned to near their initial base rate levels on the final SUA rating (the debriefing process will be more fully described in the Ethics section of this chapter). All participants who showed particularly strong reactions during treatments were called by me or the clinical psychologist that evening or the following day, in order to assess and help resolve any negative sequelae (there were none). A written mail survey was conducted several weeks later in one study, and a year after another study to determine negative or positive enduring effects, and the value of participating in this research. (Some of that data will be presented in the next section on Ethical issues.) B. PILOT-TESTING FEATURES AND OBSERVATIONS Extensive pilot-testing was conducted with several dozen participants to determine the ideal procedure for creating discontinuities using posthypnotic suggestions, the minimal number and type of symptoms needed to
412
PHILIP G. ZIMBARDO
create a general state of arousal, and the effects of various induced search frame biases. In addition, piloting helped us to evaluate the reality and validity of the amnesia induction, as well as to determine and deal with any negative sequella of this unique research experience. The posthypnotic suggestions used in early pilot studies instructed participants that at some time during the day’s session, when they saw a brown circle on the cover of a book, they would begin to experience feelings of anxiety. The book was about hypnosis (Fromm & Shor, 1979), which later I asked them if they had read. This procedure was abandoned when I learned that some students were getting aroused outside the experimental setting by seeing brown circles on various objects. Thus, we learned that the arousal cue had to be specifically limited to the experimental setting. We also found that directly suggesting “anxiety” as the arousal state created too wide a range of negative consequences since it tapped into multiple interpretations, including anticipations of upcoming future events over which one might feel anxious, as well as the recall of past events that had been anxiety-provoking. Consequently, we settled upon limiting the suggestion to specific physical symptoms associated with generalized arousal. It also becomes more conceptually interesting (from the perspective of the current model) if such a generic arousal state leads to misattributions that are in some way judged to be pathological. We had to discontinue using the biased search frame of “spiritual or mystical forces” as a possible explanation for the arousal, even though it generated fascinating accounts of religious influences, UFOs, intelligent extraterrestrials, and remote mind control by unseen technologies, among other accounts. (Recall my case study of Mrs. L. who resolutely believed I was controlling her mind at a distance.) This interesting search frame was not utilized in our formal studies because the after-effects it created in some students were disturbing. Some science and engineering students remained upset after being debriefed when they discovered that they could even entertain such preposterous beliefs that were alien to their scientific orientation and value system. It was evident that our various arousal manipulations had a powerful, observable impact on most pilot students, as shown in their strong emotional reactions, unusual thinking and overt behavior. However, the nature or power of the amnesia induction was less clear. Although almost all participants reported not recalling the experimental instructions, it was impossible to determine whether that was evidence for a true deficit in memory retrieval (i.e., a dissociative experience in highly hypnotizable individuals), or merely an active suppression of memory, or a socially motivated act in which they appeared as though they had forgot. Although “genuine” amnesia has been found among high hypnotizable subjects in other studies who
DISCONTINUITY THEORY
413
met the criterion of amnesia that we also used (Evans, Orne & Orne, 1980), it was important to demonstrate the reality of our amnesia conditions in our laboratory setting. Initially, we tried to do so by seeing if an experienced clinician, totally blind to the experimental treatments, could breach the induced amnesia during a diagnostic interview, and thus discover the nature of the manipulations. A practicing therapist was invited to assist me by dealing with a problem I told her I was having in a new research project, namely, some subjects were becoming upset during the experiment. She was paid to interview each of seven high hypnotizable students who shortly before had been given the arousal induction in a sound chamber (while selected psychophysiological measures were being taken). When next in an adjacent room with another student, they were exposed to the arousal cue word that appeared on a memory drum as part of a memory task. The clinician’s goals were: to find out what the student-subjects were feeling, and why they were feeling that, and to try to make them feel better-all while the subjects were still in the midst of their induced amnesic arousal. Analysis of the tape-recorded interviews revealed some encouraging outcomes, as well as some discouraging findings. Amnesia was never breached; the therapist was totally unable to discover the true explanation of the arousal symptoms as suggested by our experimental instructions. This result encouraged us to continue using our arousal-amnesia induction procedure. The therapist did find out that the students were aroused, upset, anxious, nervous, distressed, and were experiencing a variety of physical symptoms. She also discovered that hypnosis was somehow involved in the research. But curiously hypnosis was not entertained as the causal candidate for the negative reactions being experienced because the students insisted it could not be the responsible agent since hypnosis in the past had always made them feel good, relaxed, focused, and positive. That is the good news we had hoped to find. We were surprised, however, to hear that in every case, a suitable alternative explanation was generated based on “leading questions” derived from the therapist’s psychodynamic orientation. For example, when told that the first experience of distress came while the student was in the sound chamber, she asked if he were alone there, and/ or felt abandoned by Dr. Zimbardo, and then she explored his prior feelings of rejection. To another student who said she got aroused when taking the memory test in the large lab room along with another student, the therapist explored issues of performance anxiety and excessive competitive needs. Desires to please his demanding parents as a source of high levels of “text anxiety” seemed appropriate as her causal attribution proposed for yet another subject’s discontinuity.
414
PHILIP G. ZIMBARDO
Although the therapist was unable to make most of the student participants feel better in the relatively brief diagnosis session (30-45 minutes), she was effective with several of the last students for whom she was more assertive in proposing the dynamic reason for their distress-which they accepted as “reasonable.” There are several messages here to note. The experience of a significant personal discontinuity may motivate not only the actor to engage in cognitive detective work to come up with an appropriate explanation, but family, friends, co-workers, and professional caregivers may also offer their interpretations. Those views may reflect prevailing societal perspectives, or individual search-frame biases based on personal experience, or they may be a natural consequence of the formal theoretical training of professionals in health-care occupations (see Plous & Zimbardo, 1986). These alternative explanations may add to the Actor’s confusion when they are wrong or misleading, yet may be accepted if they have the authority power of a persuasive, credible communicator. For example, it seems that some of the client-patient testimony in “false memory syndrome” cases may be understood as a matter of attributing a variety of personal discontinuities to the alleged recovery of repressed memories of childhood abuse, as suggested by influential therapists, social workers, or the popular media. We have continued to develop more sophisticated assessment of hypnotic amnesia using EEG event-related potentials (ERP) of memory, as well as other validity tests during hypnosis training. In one of our recent studies, with a reasonably large sample of 44 subjects, (LaBerge & Zimbardo, 1999) of E R P correlates of suggested hypnotic amnesia for recently learned words, high-hypnotizable subjects reporting amnesia show significantly increased P300 in response to words covered by the amnesia suggestion compared to control words. In contrast, subjects not experiencing amnesia, whether high or low hypnotizables, or whether simulating amnesia or not, show no such difference. The increased P300 waveform may well index the surprise of unconscious recognition or unexplained arousal in subjects experiencing amnesia. Although it is possible to breach hypnotic amnesia with expectancy manipulations (Silva & Kirsch, 1987), the expectation of amnesia following deep hypnotic relaxation (as in all of our research) has been shown to produce dramatic reductions in recall for a set of stimulus words.
C. OVERVIEW OF EXPERIMENTAL FINDINGS The individual experiments that follow will show the following general outcomes:
DISCONTINUITY THEORY
415
1. High hypnotizable participants respond differently than those who are low hypnotizable to amnesia versus awareness suggestions of the source of induced arousal. 2. Hypnotic simulators respond less intensely than hypnotic participants to unexplained arousal inductions. 3. There are significant autonomic, psychological, and behavioral consequences of these induced discontinuities. 4. Concomitant mood states are typically negative in affective quality. 5. Plausible alternative explanations are generated by subjects to account for their experienced discontinuity. 6. Paranoid-like thinking is generated by biased attempts to explain discontinuities based either on unexplained deafness or unexplained arousal with a social search frame bias. 7. Research participants construct explanations for their induced discontinuities that are predictable from their induced search frame categories 8. Predictable types of psychopathological reactions (phobias, hypochondriasis or somatoform disorders, and paranoia) emerge in a significant number of these normal individuals, as assessed by self-reports, peer ratings, objective clinical tests, and expert judgments by clinicians. 9. Subjects typically fail to disclose the strong negative feelings and thoughts they have about their unexplained arousal to another person (peer or confederate) in the arousal context, although when they do, disclosure has beneficial effects. This interpersonal disclosure setting is one social dimension of Discontinuity Theory that raises the issue of the attribution of one’s discontinuity either internally to idiosyncratic dispositions or externally to shared properties of the immediate stimulus situation being experienced by coacting others.
D. MULTIPLE CONSEQUENCES OF UNEXPLAINED AROUSAL This study (Zimbardo, LaBerge, & Butler, 1993) compared the emotional, cognitive, and autonomic nervous system reactions of subjects experiencing physiological arousal with and without awareness of its source. High hypnotizable subjects (and an equal number of low hypnotizable subjects as controls) were used in a unique within-subject design in which hypnotically induced arousal (heart rate and respiration increases) was experienced by each subject both with awareness and again a second time with amnesia for its source (in randomized sequence). This two-phase
416
PHILIP G. ZIMBARDO
procedure enabled each subject to have arousal experienced as explained and also as unexplained. Unexplained arousal was operationalized in this study as the induction of generalized physical arousal symptoms with source amnesia. We assumed that such unexplained arousal would be experienced only by those subjects who were highly hypnotizable and thus could internalize experimental suggestions for hypnosis and for arousal with amnesia for its true source. We predicted that the experience of discontinuity, in the form of unexplained arousal, would have the following demonstrable effects: (a) both autonomic and psychological measures of arousal would be comparable in their patterns of change over the time course of experiencing relaxation, arousal, and debriefing; (b) the level of autonomic and psychological arousal would be significantly greater than in other conditions (either for explained arousal among those high in hypnotizability, or in either the amnesia or awareness condition for those low in hypnotizability) because of the added effects of uncertainty-caused anxiety superimposed on the initial general arousal; (c) subjects would report concomitant mood state changes in the negative direction; and (d) subjects amnesic for the source of their condition would be more likely to generate plausible alternate explanations, that is, causal misattributions. As predicted, for the hypnotizable subjects, unexplained arousal produced significant and dramatic effects when compared with explained arousal (for high or low hypnotizable subjects). We found elevations in self-reported and physiological measures of arousal, negative mood states, and also causal misattributions that characterized those experiencing this discontinuity. Eighteen subjects completed the experimental procedures, nine in each hypnotizability condition. Among our highly hypnotizable subjects, five were male and four were female; of our low hypnotizable subjects, three were male and six were female. Another eight students were utilized in pretesting phases to train research assistants, test aspects of the procedure, and assess outcome measures. Of the 280 students taking the group hypnotizability test procedure, 55 (20%)scored as highly hypnotizable, 29 (53%) of whom passed the amnesia item, while 32 (11%)scored as low hypnotizable. Our basic design involved repeated assessment of several physiological and psychological variables at selected times during a 30-minute sequence of events in each of two similar but separate experimental periods. Under the guise of a study of hypnosis and signal detection, all subjects received identical, standardized taped instructions that included responding to a variety of faint audio signals, hypnotic suggestions for relaxation, and then induced arousal with or without amnesia for the posthypnotic suggestion, followed by various dependent measures. Subjects were randomly assigned
DISCONTINUITY THEORY
417
to the arousal-amnesia or arousal-aware treatment for the first experimental period. Using a within-subject design, each of the low and high hypnotizable subjects then experienced the reversed aware/amnesia treatment in the second period. This within-subject manipulation was introduced to reduce error variance in the physiological measures and to provide a more effective test of the effects of unexplained arousal uncontaminated by individual differences between subjects. Three kinds of physiological measures were used: heart rate, respiration rate, and EEG recordings (technical details are available in Zimbardo, LaBerge, & Butler, 1993). Three different psychological measures were utilized to tap perceived arousal, mood, and subjects’ attributions for their arousal. A primary dependent measure was the subjective assessment of experienced arousal reported as SUA. These repeated SUA ratings were made five times during the first period of the procedure: (a) initially after the protocol was explained and subjects had completed consent forms; (b) after being fitted with sensors that would record their heart rate, respiration, and EEG reactions, and the recording equipment had been calibrated; (c) following hypnotic relaxation suggestion; (d) following posthypnotic cued arousal; and (e) as the final baseline measure at the end of debriefing in the first phase. During the second phase of the study, SUA ratings were taken three more times corresponding to the repetition of events 3,4, and 5 in the first period (data presented in Figure 4). Additional psychological measures were taken during the critical period when subjects were experiencing the effects of arousal following the amnesia or awareness manipulation. These included a selection of items from the Profile of Mood States (POMS) (McNair, Lorr, & Doppleman, 1971), and a sentence-completion task designed to check on memory for the hypnotic suggestion and causal attributions for the arousal. Both were administered verbally via an intercom between the experimenter in the control room and the subject who was isolated in a dimly lit, sound-attenuation chamber within an adjacent laboratory. As the experimenter read each of the 34 mood descriptor terms (e.g., “forgetful,” “alert,” “annoyed”), the subject replied aloud with a number from a 5-point scale that indicated the extent that mood was currently being experienced (where 0 = not at all; 2 = moderately; and 4 = extremely). Next, the experimenter read aloud each of ten sentence stems (e.g., “Right now I feel . . . ,” “The answer to my confusion i s . . .” “I feel the way I do right now because . . .”) and the subject’s spontaneous stem completions were recorded. Subjects were led to believe the study concerned differences in signaldetection ability between those who differed in hypnotic susceptibility and degree of hypnotic relaxation. This cover story was used so that subjects’ attention would be focused on randomly presented, external acoustic signals
418
PHILIP G. ZIMBARDO
in order to create a uniform mental set across subjects. I t also would enhance their responsiveness to the acoustic arousal-cue signal. Baseline physiological measures were taken t o calibrate our recording systems and to provide a pre-experimental index for subsequent change. Subjects were instructed to try to increase their heart and (HR respiratory rates (11) as much as possible (without moving around in their seat) for one minute. This Arousal-Demand period allowed the subjects to experience intentionally directed arousal in a nonthreatening way in this novel experimental situation, prior to their hypnotically induced arousal experience. Following their baseline assessment, subjects heard a 2-min tape-recording to induce a state of hypnotic relaxation, followed by the third SUA rating and then taped arousal instructions: “In a short while you will have an unusual experience. When you hear a bell ring, like this (BELL RINGS), you will act as if you are aroused, so that your heart rate will increase and your respiration will increase. You will continue to respond this way while performing various tasks. You will maintain that arousal until you hear, ‘Now you can relax.’ Then you will no longer feel the symptoms of arousal; as the signs of your arousal go down to their usual level, your heart rate and respiration return to their typical normal levels.” This posthypnotic arousal suggestion was combined with a second suggestion designed to establish either Awareness (AWR) or Amnesia (AMN) for the source of the cued arousal experience-as I described in the earlier Overview section. The arousal cue followed a 60-sec baseline period of no activity by the subject, then came the fourth SUA rating, with physiological data collected for 120 sec after the arousal cue. The mood measure and sentence completion measure were then administered, followed by removal of the posthypnotic suggestions for arousal and amnesia, and a fifth SUA rating. After a brief rest period of about 5 min, this procedure was repeated with the awareness or amnesia suggestion reversed for each subject in the two hypnotizable conditions, with the order counterbalanced within condition. At the end of this two-phase procedure, each subject was given an extended, personalized process debriefing (see later Ethics section for details). 1. Results and Discussion
The major results are described first for the psychophysiological measures, then the SUA ratings, the mood ratings, and finally, the sentence completions (see Zimbardo et al., 1993, for the detailed statistics of these findings). Mean values of HR and RR were analyzed by a mixed ANOVA (Hypnotic Level X Treatment X Pre-Post Arousal Cue) and significant main effects were found for both H R (F(1, 16) = 15.9, p < .001, and R R (F(1,
DISCONTINUITY THEORY
419
16) = 9.1, p < .01). A significant Hypnotic Level by Pre-Post interaction indicated that high hypnotizable (HI-HYP) subjects increased their H R more than did the low hypnotizable (LO-HYP) subjects from before to after arousal (F(1, 16) = 5.9, p < .03). Arousal change scores associated with the arousal cue were computed and H R was found to be five times greater for the HI-HYP group than for the LO-HYP group under the Amnesia condition ((16) = 3.1, p < .01). Moreover, the comparison between Amnesia and Aware conditions showed, as predicted, significantly greater respiration rate increases for the HI-HYP group relative to the LO-HYP group ((16) = 2.6, p < .02). There were no group differences in either measure for the Aware condition (nor any gender or order effects). After the data were Windsorized to deal with outliers, t-tests showed that again, as predicted, the HI-HYP group compared to the LO-HYP group showed more arousal in the Amnesia than Aware conditions for both heart rate ((12) = -4.2, p < .OOl) and respiration change scores ((12) = 2.8, p < .02). Figure 1 shows this effect as box plots for both H R and RR differences. It is apparent that the big action is in the H R arousal effects for the highly hypnotizable subjects experiencing amnesia (our Discontinuity condition). The immediacy and power of the arousal induction on the HI-HYP subjects compared to the minimal impact on those in the LO-HYP condition is apparent from the HR data plotted across the experimental time line, as shown in Figure 2. The second obvious effect, in support of one of our hypotheses, is the sustained higher arousal of those hypnotizable subjects when they were in the Amnesia condition than when they were Aware of the source of their arousal. By contrast, the lowest arousal is seen for the Amnesia-LO HYP group. Although not presented in our earlier publication, there was also a provocative EEG finding, which I think is worth mentioning here for its possible value in speculating about brain mechanisms involved in these amnesia effects. As can be seen in Figure 3, there was relatively greater activation in the right parietal lobe (P4) than the left (P3) for EEG Alpha waves in only one condition: HI-HYP experiencing Amnesia. Although the effect is small and needs replication under conditions where subjects are not carrying out as many tasks as in this study, it could be interpreted to mean that the right cerebral hemisphere was being activated during the time these hypnotizable subjects were experiencing amnesia. Recall it is the right hemisphere that is the “silent partner” to worldly experiences which are interpreted by the left cortex (Cazzaniga, 1998). Thus I would expect this effect to be reversed during the time when those experiencing such discontinuities are trying to generate causal attributions to make sense of them,
420
PHILIP G . ZIMBARDO
Fig. 1. Box plots for heart rate and respiration rate differences by condition. The box plots (Cleveland, 1985) are read as follows: The line through the middle of the boxes marks the median values; the tops and bottoms of the boxes mark the 75th and 25th percentiles; the lines extending above and below the boxes mark the 90th and 10th percentiles; the circles mark data points above and below the 90th and 10th percentiles. (From Zimbardo et al., 1993. Copyright 0 1993 by the American Psychological Association. Reprinted with permission.)
with greater left hemispheric activation. That evidence will have to wait for our next program of research. The SUA data were analyzed with a 2 X 2 X 2 (Hypnotic Level X Treatment X Pre-Post Arousal Cue) mixed analysis of covariance
0 Amensia, hi-hypnosis
88. o Awareness, hi-hypnosis
rn Amnesia, lo-hypnosis 0
Awareness, lo-hypnosis
Arousalcue
30
/
60
p/
120
90
150
180
Sec Fig. 2. Mean heart rate under amnesia and aware conditions by hypnosis group. (From Zimbardo et al., 1993. Copyright 0 1993 by the American Psychological Association. Reprinted with permission.)
.2
.15
1-
-.05
i
Amnesia
0 Lo-hypnotizable IHi-hypnotizable
I
T
I I
Awareness
-.1 Fig. 3. Mean relative electroencephologram alpha activation in right and left parietal lohes.
422
PHILIP G . ZIMBARDO
(ANCOVA), with Post-Hook-up SUA levels as the covariate and with Treatment and Pre-Post cue scores as the repeated measures. Figure 4 portrays this elegant set of data arrayed by the two hypnosis groups each separately for their Amnesia and Aware conditions. We see that their initial baselines are nearly identical, as are their Pre-Cue low levels of arousal. The arousal cue has a clear effect across the board-everyone showed some degree of arousal (F(1,16) = 49.3, p < .OOOl). However, the arousal for the HI-HYP subjects was greater (“very aroused”) than for the LOHYP group (“moderately aroused”), (F(1, 16) = 8.2, p < .Ol). Arousal ratings were highest when HI-HYP subjects experienced the Amnesia condition compared to their reactions when in the Aware condition (@) = 2.7, p < .03). A near significant triple interaction with p < .07 (Hypnotic Level X Treatment X Pre-Post Arousal Cue) confirms what can be seen in Figure 4, arousal was greatest when hypnotizable subjects were experiencing Amnesia. It is also important to point out that these differences between the Amnesia and Aware conditions in self-reported arousal were highly correlated with similar differences in both HR change scores (r(18) = .54, p < .02), and also with RR change scores (r(14) = .71, p < .005). Thus, as expected, both systems were responding comparably to our manipulation. Two final comments about the SUA ratings deserve passing mention. One index of the effectiveness of our debriefing procedure is shown by the substantial decline in arousal in all conditions back down to, or even below, the initial, normal level of arousal that students brought into this experi-
0LO-HYPNOSIS HI-HYPNOSIS
s
cn
Pre Cue
3. 2.
BASELIME
AMNESIA
AWARE
Fig. 4. Mean subjective arousal (SUA) ratings by experimental condition over the course of the experiment. (From Zimbardo et al., 1993.Copyright 01993 by the American Psychological Association. Reprinted with permission.)
423
DISCONTINUITY THEORY
ment. The power of the hypnotic arousal induction can also be gauged by the HR increases in some subjects by as much as 20 beats per minute in a short interval, with SUA ratings reaching as high as the scale maximum. From these arousal measures we do not get a sense of the hedonic quality of the experience; for that report we must turn to our mood measures. They show us that induced unexplained arousal not only has clear psychophysiological effects, but it has a predictable impact on mood states, as shown in Table IV. This effect is not experienced as an hedonically neutral, generalized activation, but rather is manifested as a syndrome of negatively valenced affects. Those subjects in the Discontinuity condition (HI-HYP + Amnesia), showed significant elevations in their degree of Negative Arousal ( 4 8 ) = 2.79, p < .02); Confusion ( 4 8 ) = 2.90, p < .02); Anger ((8) = 2.45, p < .04); and a trend in the predicted direction for Depression ( 4 8 ) = 2.00, p < .08). Paired t-tests revealed no significant differences in the mood states of LO-HYP Subjects in the Amnesia compared to the Aware conditions. Main effects of hypnotizability were obtained for Negative Arousal (F(1,16) = 4.4, p < .05), and for Anger (F(1, 16) = 4.5, p < .05). There was also a main effect of condition on several of the scales, with the mood ratings in the Amnesia condition being consistently higher than when the same subjects were in the Aware condition
TABLE IV MOODSTATESCORES: MEANS(AND SDs) BY CONDITION AND GROUP" ~~~
~
~
Awareness conditionh
Amnesia condition
LO-HYP
HI-HYP
LO-HYP
HI-HYP
Negative arousal (tense, restless, nervous, anxious, on edge)
1.51 (.81)
2.20 (1.06)
1.62 (.92)
2.73 (99)
Vigor (lively, energetic, active, vigorous, alert)
1.82 (30)
2.20 (.72)
1.96 (.65)
2.60 (.62)
Confusion (forgetful, muddled, bewildered, confused, uncertain)
.89 (.67)
1.07 (1.W
1.09 (.76)
1.91 (1.18)
Anger (grouchy, annoyed, resentful, rebellious, angry) Depression (unhappy, blue, miserable, sad, discouraged) Fatigue (exhausted, fatigued, listless, worn out, sluggish)
.24 (.26)
39 (1.09)
.44 (.46)
1.53 (1.37)
.27 (.25) 1.09 (.71)
.69 (.81) 1.58 (.81)
.47 (.49) 1.11 (.74)
1.09 (1.21) 1.71 (37)
" From Zimbardo et al., 1993. Copyright 0 1993by the American Psychological Association. Reprinted with permission. LO-HYP, low hypnotizability; HI-HYP, high-hypnotizability.
424
PHILIP G. ZIMBARDO
(Negative Arousal: F(1,16) = 6.7, p < .02; Confusion: F(1, 16) = 9.4, p < .008;Anger: F(1,16) = 9 . 9 , ~< .006;and Depression: F(1,16) = 6 . 1 , ~< .03). Attributions for the arousal were obtained indirectly by means of the sentence-completion test and qualitative analysis of the subjects’ answers. The apparent differences in the general cognitive-emotional state of HIHYP SS when they were in the Amnesia versus the Aware condition, and the lack of such differences in the responses of LO-HYP subjects across the two manipulated conditions (which were essentially the same) were tested by having independent judges predict from which experimental condition each response came. They were able to do so to a significant degree (binomial test, p = .009) for the HI-HYP subjects, but not for the LOHYP group ( p = .30). Consider next the different interpretations given by the same highly hypnotizable subjects when they were aware or had amnesia for their arousal as shown in this sample of reactions to the stem, “I feel the way I do right now because . . .”
S1 (AMN): I’ve been running and I’m trying to relax. (AWR): I’m following hypnotic suggestion. S11 (AWR): He told me to do it, I think. (AMN): I don’t know why-I think something is happening. S14 (AMN): ’Cause I’m nervous. (AWR): I’m in an experiment. S15 (AMN): Because she lied (he reported discovering his girlfriend was unfaithful) . (AWR): I’m wired to a machine. S19 (AWR): The EEG cap on my head is too tight. (AMN): Exams are coming up; I’m stressed. This overall pattern of results (despite the relatively small sample size) is in line with the theoretical predictions from Discontinuity Theory. Creating discontinuity experimentally by means of unexplained arousal triggers a range of significant and dramatic effects on emotions, cognitions, and physiological functioning. Furthermore, these results show that the effects of unexplained arousal can be assessed at both autonomic and psychological levels. These effects were not only statistically significant, but were large enough in an absolute sense to represent dramatic increases in arousal intensity that were enduring over the course of the study and even personally disturbing to some of the subjects. The lack of an immediate and situationally appropriate explanation for their sudden arousal clearly heightened the generalized arousal being experienced by our hypnotizableamnesic subjects. Our findings also show that unexplained general arousal is a personally experienced discontinuity that is perceived as hedonically negative arousal, and as “aversive” in the moods it invokes. Those feelings are characterized by the research participants’ self-reports as being angry, grouchy, annoyed,
DISCONTINUITY THEORY
425
resentful, rebellious, unfriendly, anxious, nervous, tense, and restless. I think it is this negative affective state that motivates a search for an appropriate causal explanation, one biased in the direction of selectively noticing or recalling cues that support a negative interpretation of the anomalous experience. Of further interest to the study of emotion was the complex of multiple negative affect states created by our somatic arousal with amnesia. We are reminded here of Polivy’s (1981) demonstrations of several affects resulting from experimental attempts to induce one particular emotion. Her anger manipulations created not only hostility, but also high levels of correlated anxiety and depression-comparable to the pattern observed in our data reported above. It should also be apparent that the obtained pattern of results cannot be attributed to experimental “demand characteristics” in which subjects simply give us back what we had instructed them to experience in our suggestions, or were elicited because of implicit communication between subject and experimenter. The pattern of psychophysiological responses shown by the highly hypnotizable subjects who experienced unexplained arousal goes well beyond the specifics of the suggestions they were given (merely H R and R R increases). Moreover, the taped instructions and all treatment of the subjects were identical for the high and low hypnotizable subjects throughout the experiment by the experimental assistant who was blind to their condition. This was equally true of the identical treatment of those in the amnesia and aware conditions, with the sole exception of the taped posthypnotic suggestion to be forgetful or knowledgeable of the source of the arousal. With this encouraging study deposited in the data bank, we move to consider the next, larger experiment, some of whose assets have not been reported previously to the authorities.
E. MULTIPLE PATHWAYS T O PARANOIA
The first of our studies to focus explicitly on inducing psychopathological reactions in normal individuals was conducted with Susan Andersen and Loren Kabat (Zimbardo, Andersen, & Kabat, 1981). Here I summarize the highlights of that study and add a treatment and additional features of the methodology and results that were not reported in the original article (see Zimbardo & Andersen, 1999). They were tangential to the limited focus of that brief report, but are central to the thesis being advanced in this chapter. Recall Old Joe on the pharmacy line “going ballistic” when he misconstrued his inability to hear the nurse’s kind comments to him as her indifference and possibly prejudice toward him? His partial deafness created a
426
PHILIP G . ZIMBARDO
perceptual disorder, which must have made his world confusing at times, and which could be resolved by developing causal attributions about other people acting malevolently toward him. It could be argued that he correctly reasoned (about social agents acting hostile toward him) from an initially false perception (being unaware that his hearing disorder was organic) given the confirmatory evidence he was able to uncover in his social setting. That line of reasoning may generate beliefs that qualify as paranoid delusions when they are held with conviction despite contrary evidence (Cameron, 1943). Such a view about the etiology of delusional thinking has been persuasively argued by Brendan Maher (1974a,b; Maher & Ross, 1984). His analysis is supported by clinical observations that paranoid reactions are often seen among the elderly when their hearing loss is gradual, thus they may be unaware of it as the source of perceptual anomalies (Cooper, 1976; Cooper & Curry, 1976; Cooper, Kay, Garside, & Roth, 1974; Post, 1966, and others). As I mentioned earlier, of all mental disorders, paranoia should be of most interest to social psychologists. It involves real or imagined transactions between people, as well as fascinating self-attention and social perception-cognition processes (Butler, 1993). Moreover, much of paranoid thinking is characteristic of everyday thought (Artiss & Bullard, 1966; Fenigstein & Vanable, 1992). Consider this scenario. If you could not hear what people were saying because they seemed to be whispering, and they denied that they were when confronted, it would be reasonable to conclude that they were lying or covering up something. When challenged, they would react with confusion, possibly anger, that could escalate into a hostile interaction. Observers, unaware of the hearing disorder, judge the hard-of-hearing person’s actions as bizarre and as evidence that a “dangerous” thought disorder exists. Their interpretation, though false, constitutes a sufficient basis for excluding the person from their company. Perceiving signs of being socially excluded is a new source of ambiguity and confusion for that person, to which he or she responds with hostility and ideation of a now “obvious” conspiratorial threat (Lemert, 1962). Becoming isolated, the person loses opportunities for corrective social feedback, and any delusions of persecution become self-validating in a closed, autistic system. Thus, a once-normal person may spiral down this path to paranoid pathology. Our research was designed to broaden the analytical focus of these earlier correlational and conceptual studies in several ways. First, we put the concept of perceptual anomaly (deafness without awareness) within the larger generic category of discontinuity. That led us to consider other paths to paranoia, such as that which might be based on a somatic anomaly of unexplained arousal, coupled with a socially focused biased search frame.
DISCONTINUITY THEORY
427
Second, we eliminated obvious alternative explanations based on biological aspects of the elderly with hearing deficits, because we studied young, healthy individuals as our subjects. Third, we experimentally created in the laboratory the conditions believed to play a causal role in the etiology of some forms of paranoia. The purpose of this study was to demonstrate that it is possible to initiate paranoid thoughts and feelings in young, healthy, normal subjects by inducing either of two types of discontinuity, unexplained deafness or unexplained arousal, under certain conditions. One condition was to create a context for socially focused misattributions by having several others present and interacting in ways that could be interpreted as excluding the subject, or even making fun of him. The second condition was to add the socialfocus explanatory search frame to the unexplained arousal manipulation. We predicted that both experimental treatments of discontinuity would give rise to significant elevations in paranoid thinking as measured by standard and specially devised measures, as well as in self-ratings and judges’ ratings of the subjects’ actions and feelings. The subjects’ reactions were compared with two control groups, one of explained deafness, and another for the effects of following posthypnotic suggestions. We did not make differential predictions about the ways the two experimental groups would respond on our various dependent measures. There were some special features of this study’s methodology (in addition to those common elements described in the earlier Overview) that deserve mention. The 24 highly hypnotizable male students who participated in the training and testing phases of this study were randomly assigned to four treatments: Unexplained Deafness; Explained Deafness Control; Unexplained Arousal (with Social Bias), and a Posthypnotic Suggestion Control group. Amnesia was assessed and validated for all subjects both on the Harvard Group Scale of Hypnotizability and in individual assessment on the Stanford Scale, Form C. During hypnosis training, those assigned to the partial hearing loss conditions demonstrated their ability to experience hearing impairment by failing to repeat five or more of eight standard tape-presented phrases. The last item of subject selection was establishing their normal range of functioning on several scales of the MMPI (Dahlstrom, Welsh, & Dahlstrom, 1975) and on our medical evaluation form. The paradigm used in this study was one in which individual subjects were led to believe we were studying creativity and problem solving of individuals and teams under varied hypnotic conditions, but they were unaware that the two others in the group were confederates. In phase one, the hypnotic induction of treatments, when the subject arrived at the laboratory, another “subject” was already there and the second confederate
428
PHILIP G . ZIMBARDO
joined them soon after, and all three participants were given hypnotic relaxation suggestions-to establish their comparable status. Next, they were told that they would be listening to special music to deepen their relaxation state, and would each go to separate listening rooms equipped with hi-fi stereo headsets. The real subjects then heard tape-recorded music followed by a tape of one of the four treatments, pre-selected by the research assistant (so all other researchers and confederates were ignorant of their condition). The discontinuity-induced post hypnotic suggestions for all conditions used the same triggering cue: “When you see the word ‘Focus’ projected on the screen in the next room, and only in that room . . .” 1. Unexplained Arousal (Social Bias): Subjects were given our standard suggestions for that cue to induce increases in heart rate and respiration with amnesia for their source, to which were added, “You might find that what you are experiencing has something to do with other people.” 2. Unexplained Deafness: Their instructions continued, “That will be a signal to you that your environment has changed so that sounds are reduced, making room noises and what other people are saying very low. It will thus be very difficult to understand what other people are saying because it will seem as if they are talking so low as to be nearly inaudible. You will be concerned by the fact that others appear to be talking so low you can’t hear them well, as if they were whispering.” Source amnesia was then suggested. 3. Explained Deafness Control: These subjects were treated identically to the Unexplained Deafness condition, but were explicitly told that they were aware of the suggested source of their experience of partial deafness. Those in both deafness conditions were also told that their hearing would return to its usual level as soon as the experimenter touched their shoulder, and then they would be able to understand perfectly what others are saying. 4. Posthypnotic Control: Their instructions continued, “You will begin to experience a change in you. Your left earlobe will begin to feel very itchy, and will continue to itch until you scratch it. You will feel compelled to scratch it. As soon as you do, you will feel fine again. You will not remember having been given this suggestion until the experimenter touches your shoulder. Then you will remember everything. You will not remember this suggestion until then.” This hypnosis control was included to assess whether some dynamic quality of carrying out a posthypnotic suggestion with amnesia might be sufficient to generate the predicted results. Data from the pair of control groups also help to rule out experimental demand characteristics (they were treated identically throughout the experiment and confederates and researchers were blind to conditions), subject selection traits (they were
DISCONTINUITY THEORY
429
equally hypnotizable and amnesic), and context features that might facilitate irrational reactions (they were the same). In phase two of the experiment, the subject entered the familiar large laboratory room, (where hypnosis training sessions had been held), taking the empty seat near the two confederates who were sitting closer to each other. The experimenter gave a cover story about creativity, problem solving, and hypnosis, explained that all task instructions and most stimulus materials would be projected on the screen, there would be an initial problem-solving task (10 anagrams) during which they should decide to work alone o r to collaborate on the next task of analyzing a picture (TAT card), after which there was a second creativity task to be done alone, followed by final reactions of each of them, also to be done separately in adjacent rooms. Seeming to check out the projector before leaving, the experimenter activated the first slide, “Focus,” the posthypnotic cue. As the subject started working on the anagram task, the confederates engaged in a well-rehearsed, standard interaction. Their dialogue was designed to emphasize an alliance or comraderie as they recalled having met at a party, laughed at an incident mentioned, made funny faces, and eventually decided aloud to work together on the next task. Their verbal exchange and mannerisms were also an occasion for misunderstanding by subjects in the experimental conditions. Before turning to the anagram task, one of them casually asked the subject if he wanted to work with them on the next task and recorded any answer unobtrusively, as a measure of desire to affiliate. During the session, the confederates asked the subject three other questions, and they and a pair of observing judges (behind a two-way mirror) recorded the answers, if any. After completing the various timed tasks, subjects were instructed to return to their relaxation music listening rooms for the next part of the study. That gave the confederates the opportunity to immediately and independently give their reactions to the subject on a series of mood adjectives (which were averaged in the data presentation in Table V; their inter-rated reliability across all these measures was r = S7). Subjects’ behavior, thoughts, and feelings were assessed on the following dependent measures: number of correct anagram solutions; content analysis of “creative” stories written alone to the TAT picture; their self-assessed creativity on this task; the Multiple Affect Adjective Checklist (MAACL) (Zuckerman & Lubin, 1965), ratings on a series of self-descriptive adjectives; and the short form of the MMPI administered earlier in selection. O n that form, subjects rated themselves on 207 items representing scales measuring Paranoia, Suspiciousness, and Grandiosity, as well as on two scales on which we did not expect to find treatment differences, Schizophrenia, and Preliminary Hypochondriasis. Indeed, no differences were found on those two measures, In addition, they also completed the Paranoia
MEANSON
A
TABLE V VARIETY OF MEASURES DISTINGUISHING EXPERIMENTAL FROM CONTROL GROWS Treatment Experimental
Dependent measures
Unexplained arousal, social bias (n = 6)
Control
Unexplained deafness (n = 6)
Explained deafness (n = 6)
Posthypnotic suggestion (itching ear) (n = 6)
F
- .28 -.17 .67 - 1.00
7.48*** 1.72 1.73 1.63
4.63*** * 2.13** 1.71* 1.oo
5.17
3.00
3.12****
r-value"
Paranoia measuresb Paranoia clinical interview form MMPI-Paranoia MMPI-Suspicion MMPI-Grandiosity
.33 2.33 2.83 -.83
.30 1.50 1.oo 1.33
MAACL-Hostility
5.33
11.oo
- .09 .33 .00 -.83 2.40
75.00 67.50
83.35 42.83
16.55 68.33
33.50 73.33
3.68* 4.66*
3.23**** 2.47**
2.67
3.00
4.17
4.83
1.19
1.87**
8.22'*** 6.75'***
3.55*** 2.36***
Thematic Apperception Test Affective evaluation Self-assessed creativity Anagrams Number correct Affiliation (%) During experiment Volunteered for future
17 17
17 0
50 50
100 67
-
$
Self-rated feelings Creative Distracted Confused Relaxed Agitated Irritated Friendly Hostile Happy Judges' ratings Confused Relaxed Agitated Irritated Friendly Hostile Happy
54.17 74.33 55.00 27.17 74.33 55.83 33.83 56.83 74.33
34.17 60.83 73.33 43.33 73.33 70.00 26.67 38.33 45.83
55.83 41.67 39.17 81.67 14.17 25.00 53.33 13.33 51.67
65.83 19.17 35.00 78.33 15.33 7.00 56.67 13.33 65.00
1.72 3.81* 2.17 7.56*** 22.72*** * 10.61**** 2.16 3.83* 1.25
-1.65* 3.02**** -1.65* 4.61**** 8.25**** 5.34** * * 2.48** -3.16**** -1.61*
33.00 36.67 38.17 28.50 35.50 8.00 33.34
40.83 34.17 51.25 45.84 23.34 18.75 21.25
27.08 54.59 24.59 18.92 48.34 5.00 40.00
17.67 65.42 13.75 11.25 65.00 1.67 56.25
.77 3.06 2.62 2.75 4.87* 2.00 3.75**
1.31 -2.89**** 2.54*** 2.46** -3.39**** 1.92** -2.76***
'All &tests are one-tailed and derived using a single planned comparison (Hays, 1965, p. 465). Reported paranoia means represent difference scores (post minus pre). MMPI, Minnesota Multiphasic Personality Inventory; MAACL, Multiple Affect Adjective Checklist. Chi square comparisons. experimental versus control groups. * p < 10, ** p < .05,*** p < .01, **** p < ,005
432
PHILIP G. ZIMBARDO
Clinical Interview Form, a specially designed instrument for this study that contained 15 declarative self-descriptive statements adapted from a British clinical study of paranoia in hard-of-hearing persons (Kay, Cooper, Garside, & Roth, 1976). All subjects went through an elaborate debriefing expertly administered by Susan Andersen (see Ethics section for details). All those showing strong reactions during the study were contacted the next day for reassurance, and again a month later when we readministered the relevant MMPI paranoia scales and the Paranoia Clinical Interview Form. As we expected, and I am pleased to report, every subject’s scores returned to their normal, nonparanoid, pretest values.
1. Results and Discussion The manipulations appeared to have worked as intended. The Unexplained Arousal subjects were significantly more agitated ( p < .001) and less relaxed ( p < .005) than the Controls, but equal to those in the Unexplained Deafness condition (See Table V). All those in the two deafness conditions informally reported difficulties in hearing the other subjects (confederates) during the experiment, felt their hearing was not “keen” on the final postexperimental rating (mean of 21 compared to 78 for the two nondeafness conditions, p < .001), and they answered fewer of the four questions posed by the confederates during the experiment than did the other subjects (averages of 1.4 versus 3.8, respectively, p < .001). All those in the Itchingear, posthypnotic control condition scratched their left ears, as noted by both observers, and all of them also reported that they had experienced an itching sensation in their left earlobe. Finally, variations in the amnesia manipulation led all six subjects in the Deafness Explained condition to report (during debriefing) being fully aware of why they were having hearing difficulties, while none of the other 18 subjects reported any memory for that suggestion. Table V summarizes the primary results for most of our measures. As predicted, experiencing a discontinuity, either perceptual or visceral, with amnesia for its source, produced significant changes in cognitive, emotional, and behavioral functioning. Those in the two discontinuity conditions became more paranoid as shown in a number of ways. Most directly, they showed significant elevations on the Paranoia Scale of the MMPI ( p < .05) and on the clinically derived paranoia interview form ( p < .OOl), and they showed marginally greater scores on the MMPI Suspiciousness scale ( p < .lo), one aspect of paranoid thinking. It is instructive to consider the total lack of face validity in the content of some items on the MMPI Paranoia Scale where discontinuity subjects changed in the paranoid direction from pre- to posttesting, while the controls did not: “If I were a reporter I would
DISCONTINUITY THEORY
433
very much like to report news of the theater”; “I have had no difficulty in starting or holding my bowel movements,” and ‘‘I would like to wear expensive clothes.” Something deeper than experimenter expectancy biases or subjects wanting to please the experimenter appears to be operating here. Another aspect of paranoia is feelings of hostility that experimental subjects displayed at significantly higher levels on the MAACL than did controls ( p < .Ol)-notably the enormously high hostility level expressed by the Unexplained Deafness subjects. That finding, when coupled with the absence of differences on other MAACL scales of anxiety and depression, supports the notion that these discontinuity inductions evoked specific affective responses considered most representative of preliminary stages of paranoid thinking (see Beck, 1974). There were also a number of indicators of the social interplay between the negative feelings and actions of the experimental subjects and the reactions to them of the confederates, which contribute to the process of developing paranoid thinking and the labeling of someone as paranoid. The subjects were highly and significantly agitated, irritated, hostile, and distracted, while feeling neither friendly nor relaxed. The confederates responded to them in kind, perceiving the discontinuity subjects as agitated, irritated, hostile, unfriendly, and not relaxed or happy, all significantly more so than in their judgments of the control subjects. The restricted sense of social involvement accompanying paranoid reactions is seen in the low percentage of experimental subjects who chose to work together with the confederates on the T A T (where collaboration was expected to improve performance, according to the experimenter’s depiction). Only 17% chose to affiliate on this task compared to 75% in the control conditions (xz(1) = 8.22, p < .01). When invited to participate in a future study with the same partners (the confederates), the majority of controls accepted (58%), while the majority of the experimentals declined (92%) (,$ (1) = 6 . 7 5 , ~< .Ol). The agitated-distracted state of the experimental subjects took its toll on cognitive processing required in the anagram task, where they solved significantly fewer anagrams than the controls ( p < .05). They also felt that they were not as creative on the TAT task as did the controls ( p < .OS). However, one aspect of their TAT performance provides a subtle, indirect assessment of paranoid thinking that I especially like. All TAT stories were rated blind and independently by two student judges on a number of a priori dimensions (exact agreement of 83%).One such dimension was the degree to which they were “evaluative,” both positively and negatively, regarding the actors in their stories. One hallmark of paranoia is the confident assessment and evaluation of other people, even in ambiguous behavioral situations, and this was exactly what our discontinuity subjects revealed. Their scores on this measure of affective evaluation of the TAT
434
PHILIP G. ZIMBARDO
actors were more than three times greater than those of the controls ( p < .00.5). Despite the overall similarity in the reactions of the two discontinuity conditions that took different paths to the common end-point of paranoia, there are several discrepancies to note in passing. I do so only for their provocative value in stimulating further, more process-oriented research on alternative dynamics in the creation of paranoia. The induction of paranoid reactions by means of unexplained deafness compared to unexplained arousal was marked by somewhat greater grandiosity, but less suspicion, much higher hostility, but lower feelings of creativity. Those experiencing unexplained arousal provided a paradoxical portrait of being least relaxed, most hostile, and yet most happy of any of the student-subjects (a high mean of 74 on a 100-point scale). The confederates did not pick up on their feelings of hostility (rating the unexplained deafness subjects higher, as they also did for agitation and irritation). Perhaps their negative feelings were masked by a show of happiness, which the confederates did judge as being higher among the unexplained arousal subjects than among those in the unexplained deafness condition. The small sample size in this study constrains enthusiastic generalizations, yet points up the power of our manipulations (and the utility of pre-post, within-subject change scores) to generate so many significant effects that, without exception, support the predictions derived from Discontinuity Theory. It should be apparent that it is the labor-intensive nature of this research that limits sample sizes and not the researchers’ lack of industriousness. We turn next to an experiment that expands the pathological consequences of inappropriate resolutions of discontinuities beyond paranoia to phobias, hypochrondriacal mental disorders, and more.
F. BIASED SEARCH FRAMES CREATE PREDICTABLE DIAGNOSTIC PATHOLOGIES This study (Zimbardo & Piccione, 1999) explored the relationship between particular explanatory search frames and specific types of pathology that might result from relying on them to account for the experience of sudden somatic arousal without awareness of the cause of this discontinuity. We extended our basic paradigm to include the posthypnotic induction of three different biased search frames superimposed on the unexplained arousal manipulation: People, Environment, Body or Health. A host of reactions of these subjects were compared with those in two control conditions; explained and unexplained arousal, without any induced search frames. We assessed emotional, cognitive, perceptual, communicative, and
DISCONTINUITY THEORY
435
search behaviors by means of standard scales, projective tests, self-reports, observer ratings, and content analyses. In addition, videotape segments of interviews with the subjects were evaluated by clinical psychologists on a number of dimensions, most importantly, evidence of deficiencies in interpersonal functioning, pathology, and specific DSM-ZZZ diagnoses. We predicted the following: 1. All groups would show evidence of arousal. 2. The hedonic quality of that arousal would be largely negative. 3. Those given the Environment-search bias would show evidence of visually searching the laboratory environment, create misattributions based on physical features of the current behavioral setting, and be high on the MMPI Phobia scale. 4. Those given the Body/Health-search bias would show evidence of searching their bodies, create misattributions based on aspects of bodily functioning and health status, and be high on the MMPI Hypochondriasis scale. 5. Those given the People-search frame should create people-based misattributions, show emotional-cognitive signs of envy, jealousy, anger, suspicion, and vindictiveness, give more Rorschach human anatomy responses and more rejection responses to the Rorschach (an index of pathology), and also be high on the MMPI Paranoia scale. We also expected that judgments by clinical raters would reveal evidence of symptomatic behaviors congruent with each of these explanatory biases. Finally, we were simply curious about the extent of “pathology” that these trained observers would find among our “normal” sample of college student-subjects. The participants were 50 highly hypnotizable-amnesic students (of both sexes) from an Stanford Introductory Psychology class. They were randomly assigned to one of the five conditions described above, after having successfully completed several hypnosis-training sessions. Excluded from this sample were all students who were above the average on the Manifest Anxiety Scale, had reported any current medical or psychological problem, or did not pass several amnesia tests during the training sessions. One special form of preparation for the current study included having all the subjects-in-training read a letter describing a posthypnotic arousal cue that would make them feel euphoric and happy, but they would not remember the suggestion, and might find that this unusual feeling had something to do with their past. The components of this letter parallel a letter they all would receive later during the study, some suggestions for the unexplained or explained arousal, and search biases. This procedure helped to determine the effectiveness of the manipulation, while also giving
436
PHILIP G . ZIMBARDO
subjects a specific positive association with the laboratory in which they would later be tested. The guise of “a study of creative problem solving under varied conditions of hypnosis” enabled us to administer a series of projective tests and tasks to the subjects, after they had been deeply hypnotized and given the posthypnotic suggestions. These standardized taped instructions were delivered through earphones, and at the point where the subject opened her or his eyes, the first experimenter handed over a personal letter to be read silently while simultaneously listening to its contents. The letter specified the terms of each of the five conditions. Since the letters and appropriate tapes were randomly selected by an assistant, the experimenter, as well as the second experimenter who conducted the debriefing, were blind to treatments. The somatic arousal suggestion was again for heart rate and respiration to increase on cue, but this study included a third arousal component of muscles becoming tight. In the Explained Arousal condition, subjects were told they would remember having read and heard this letter, while the No Bias Arousal subjects were informed that they would not remember these suggestions. But they, like the three biased groups, were told, “that it is important that you try to figure out the reason why you are feeling these reactions.” Those in the biased conditions were given the added suggestion that, “they might find that the explanation for what you are feeling has something to do with the physical environment” (Environment-Bias); “your body or health” (Body-Bias); “people” (People-Bias). The experiment was conducted in the same large room where hypnosis training had taken, subjects were in self-reported good health and mood at the start of the session, and they were alone during the procedure since all instructions and test materials were presented by an automated slide projector. Thus there were no obvious negative aspects of this behavioral context that could serve as a ready attributional focus for any of the three search biases. After writing a story to a TAT card for 3 min, a buzzer sounded the posthypnotic signal. After a 1-min interval (for the arousal to start and be noticed), subjects completed the MAACL, a set of 10 Group Rorschach cards (Harrower & Steiner, 1951), and three scales from the MMPI: Phobia (PHO, Wiggins, 1966), Hypochondriasis (Hs’, Welsh, 1948, and Simple Paranoia (Pa-S, Wiener & Harmon, 1946). After writing one more TAT story, they completed a personal reactions inventory that asked about their current feelings, and reasons for what they were experiencing. The second experimenter then conducted a semistructured interview (for 10 to 15 min), asking subjects to elaborate and clarify those thoughts and feelings. Whenever subjects generated an explanation that fit one of the three biases, the
DISCONTINUITY THEORY
437
interviewer challenged it by inviting them to consider other possibilities from the other two categories. A lengthy debriefing followed, and no subject left the experiment until the clinical psychologist (Piccione) felt assured that there was no evidence of symptomatic behavior. Subjects who had reacted strongly during the experiment were contacted the next day to treat any negative carry-overs (there were none), and all subjects were contacted again in a 1-year follow-up (the results of which are described in the Ethics section). The entire procedure was videotaped and parts of it were analyzed by raters who were ignorant of the hypotheses. Pairs of observers also rated subjects’ search behaviors as they were directly observed from behind a special one-way window. The last feature of this study, and perhaps its most important one, involved having ten clinical psychology interns from local Veterans Administration Hospitals review the videotapes of the final interviews between subjects and experimenters (prior to debriefing). Each pair of them viewed 10 of the 50 edited tapes (roughly equivalent lengths with the sequence of conditions randomly arranged). Each clinical intern separately provided a series of ratings and evaluations on a Clinical Observation Form that invited them to identify any students who should not be allowed to continue to participate in a second study on stress and its management. While carrying out this primary task (and ignorant of the experimental context of the interviews), they were also asked to note or rate the information category the subject predominately provided to the interviewer (our three biases plus several others); the subject’s personal and interpersonal reaction styles (1 1-point scales on eight dimensions, such as sadistic-friendly, phobiccurious); individuals who seemed most “normal” and most ‘‘pathological,’’ and finally, if justified, to give a DSM-111 diagnosis of any student whose symptoms merited such a clinical judgment. (The interns participated at the encouragement of their supervisors, were given a fee for their services, and a lecture later about Discontinuity Theory, this experiment, and their collective results.) 1. Results and Discussion Obviously this study generated a huge amount of data, only some of which can be reported briefly within the already overtaxed constraints of this chapter. I’ll start with the good news that all of our predictions were supported, with two exceptions to be described. The majority of subjects (70%) rated their experience as negative, 24% as a mix of negative and positive, and only 6% evaluated their arousal positively. Thus for 94% of the subjects this hypnotically induced arousal was interpreted as either totally negative or mixed. It was most clearly
438
PHILIP G. ZIMBARDO
negative for those in the People-Bias condition (90%, 10%mixed)-a powerfully negative reaction among these subjects that will resurface across many of the measures. The observational data offer another index of subjects’ general distress and tension. Of the average number of 31 identifiable coded reactions per group (range: 27-36), fully 74% were of tension and emotional and behavioral signs of distress, whereas only .05% were of positive reactions, with the remainder being searching the environment, one’s body, or other reactions. A factor analysis of subjective mood appraisal on the MAACL generated five factors (principal components), on which only People-Bias subjects showed significant trends toward being least Positive, least Relaxed, and most Negative-Situationally, with no effects for Challenging or NegativePersonal. On specific adjectives, the People-Bias subjects were highest on Envy ( p = .05), Guilty ( p = .09), Angry ( p = .09), and lowest on Friendly ( p = .08), with no other clear trends. On composite scales, the PeopleBias subjects were again highest on Disgust (Disgusted, Scornful, Contempt, Resentful, Envious) of any group ( p < .lo), lowest on Happiness (Amused, Elated, Excited, Thrilled, Happy, Contented) ( p < .05), and Calm ( p < .lo). The only other effects of note were for the Body-Bias to be most Nervous ( p < .lo), and the No Bias group to be most Interested (Interested, Challenged, Expectant) ( p < .05). The rest of the data will be presented in terms of the predicted effects for each of the three biased conditions rather than by individual measures. Starting with subjects given the Environment-Bias, we found as expected, that they searched the experimental environment significantly more than any other group ( p < .005) by looking around, touching things, checking out the equipment. Their mean of 4.2 was nearly twice as high as that of any other condition, which were comparably low. The reasons they gave (on the postexperimental questionnaire) for their unusual feelings were coded into 12 subcategories related to the environment, with 80% of these subjects resorting primarily to environment-based explanations. Ratings by the clinical judges of the nature of the information communicated during the interview phase of the study, again show that the majority (55%) of what these subjects said was related to the environment, as seen in Table VI. But it is the quality of their misattributions that is more important than that they simply followed the posthypnotic suggestion appropriately. Consider some of their causal explanations: “I really think that the fumes of the projector kind of made me sick . . . the ink fumes . . . not the ink fumes . . . maybe it was just the warm air”; “Well, I was feeling kind of claustrophobic . . . I was [seated] in a corner and stuff”; “Maybe from being alone, in a room alone”; “It was the heat of the projector blowing on me . . . that really bugged me. . . Well, I think it might have been that
439
DISCONTINUITY THEORY TABLE V1 CATECORICAI. INFORMATION BIAS: CLINICAL JUDGES’ RATINGS OF BIASIN SUBJECTS’ COMMUNICATED INFORMATION Experimental conditions Category“
Body
People
Environment
N o bias
Explained
Body
Jl*
24
35
45
40
People
08
62*
0s
25
30
Environment
04
03
55*
20
15
Other
11
11
05
10
15
Percentage of “category of information predominantly presented.”
* Predicted highest percentage.
camera actually.” Others said their aroused state might be due to the uncomfortable chair, the stuffiness of the room, the rainy weather, the stimulus materials, or even the shape of the pencil they were given to write their reactions. Two other bits of evidence that mesh nicely are the clinical raters’ judgments of phobic symptomatic behavior (see Table VII) and the MMPI data (see Table VIII). Those in the Environment condition were highest on displaying phobic behavior; their mean score is significantly higher than any other condition ( p = .Ol). Moreover, their elevated mean
TABLE VII INTERPERSONAL STYLEWITH INTERVIEWER: CLINICAL JUDGES’ RATINGS ON ANCHORED DIMENSIONS“ Experimental conditions Dimensions
Body
People
Environment
No Bias
Explained
Hostile versus friendly
.I5
.4J*
.OO
.05
.05
Dominant versus submissive Mistrusting versus trusting
.23
.10
.0s
.15
.25
.08
.60*
.20
.15
.20
.58*
.23 .40
35 .45*
.I5 .25
.15 .30
Illness versus health bias Avoiding versus exploring environment
.15
a Eleven-point scale: For example, -5 = “hostile”, 0 = neutral, +5 = “friendly.” Ratings are mean number of ratings in negative direction. divided by total ratings. * Predicted highest value.
440
PHILIP G. ZIMBARDO TABLE VIIl MEANSCORES ON MINNESOTA MULTIPHASIC PERSONALITY INVENTORY SUBSCALES BY EXPERIMENTAL CONDITIONS (WITH COMPARATIVE POPULATION MEANSFOR EACH SCALE) ~
Scale: Hs’
Hypochondria Population Mean:
Conditions
Bodyhealth Environment Explained arousal People No bias Phobia Population mean Environment Explained arousal No bias People Body/health Paranoia-simple Population Mean People No bias Bodyheal t h Environment Exulained arousal
Scale: PHO Conditions
Scale: PA-S Conditions
1.7 2.7 1.3 1.1 0.9 0.8 7.3 7.4 5.7 4.9 3.8 3.8 5.8 7.5 8.1 7.8 7.3 6.4
score, which resembles the clinical outpatient population mean, is interpreted as “admitting to a number of fears, many of them of the classically phobic variety such as heights, dark, closed spaces, etc.” (Wiggins, 1966, p. 13). The Body-Bias subjects did not search their bodies, as we had predicted, any more than did the others. In retrospect this was a naive prediction since the induced action was happening inside their bodies and not externally. However, most (70%) of the reasons they gave for their feelings were in one of nine subcategories related to health or their bodily functioning, among them: not enough sleep, hungry, on a diet, exercised too much, drank coffee, ate something bad, or biological cycle. For example, “My muscles are a bit tense and I have a headache , . . I think it is because of today’s early swimming practice . . . or maybe from horseback riding.” Or consider these multiple attributions from another subject, “I’m uptight because my summer plans had not been all laid out, and also I’m coming down with a cold . . . I got a really bad headache on the right side. . . .
DISCONTINUITY THEORY
441
I haven’t been getting much sleep because I went to the Formal (dance).” While subjects in all conditions communicated a lot of information during the interview about their bodies since their discontinuity was based on a visceral change of state, Table VI shows that the Body-Bias group spoke about little else, with 77% of their talk in the body category. As expected, their excessive concern about their health led the clinical raters to judge their symptomatic behavior as primarily hypochondriacal (see Table VII). Finally, their MMPI scores reveal them to be not only significantly more hypochondriacal than any other group ( p = .0003), but even greater than the population comparison mean for the MMPI Hs’ scale of hypochondriasis. The People-Bias group is clearly the most fascinating for social psychologists because the manipulation was a powerful catalyst for attributions about the real and imagined social world. Although there were no actual people with them in the laboratory when they became aroused, and they had only positive prior contacts with the experimenters, these subjects “saw” people everywhere, and in very upsetting scenarios-in line with their negative mood state noted before. Nine of the subcategories in their interviews dealt with relationships with other people, which were the dominant explanatory category for 80% of them. Specifically, they felt guilty because they had “killed” the actors in their TAT stories; were sad because of how the TAT actors had died; were depressed by how the T A T actors looked; were upset by the “horrible, violent ending” of the TAT story; were angry at recalling recent confrontations with a boyfriend, roommate, and parents; and were angry at some mean people. One remarkable attribution of extreme jealousy may or may not have actually happened, but is surely unusual in this context: “I feel that I am under heavy stress right now. . . . It’s like I remember something about two people I care about, and I am like a third person. . . . I turn around and I see my girlfriend and my brother is swinging my girlfriend around. . . . There should be a reason to be jealous. . . . We broke up ‘cos my mom and dad asked me to because they didn’t want me to get serious. . . . I still like her a lot. . . and to see her with my brother playing around triggered something.” The subjects were also significantly more likely to report seeing human anatomy in the ambiguous Rorschach cards than any other group ( p < .05), as well as reporting seeing “nothing at all,” or rejecting a card, which clinicians regard as an index of pathological responding (7 of 10 subjects gave 24 rejection responses compared to only 3 of 10 in the other bias conditions who gave only 4 rejection responses). Table VI shows that their communication with the experimenter was judged to be primarily about people, whereas Table VII shows that the clinical judges rated their videotaped interaction with the experimenter as reflecting the highest levels of para-
442
PHILIP G. ZIMBARDO
noid, vindictive symptomatic behavior, but also very high degrees of hostile, sadistic symptoms. Curiously, these subjects were also relatively high on hypochondriacal symptoms. Note that among the two controls there was no predominant symptom category as perceived by these judges. The one fly in the otherwise soothing predictive ointment is the failure of our MMPI measure to reveal this paranoid orientation. Examination of Table VIII does show that the mean of the People-Bias subjects of 7.5 is well above the population mean for this PA-S measure, but so are all the others, with the N o Bias control highest of all! Resolution of this anomaly comes from our later discovery that we had used the wrong scale, since the Subtle Paranoia Scale has been found to have poor validity with non-psychiatric populations (Wiener, 1947). But we shall see on the final measures provided by our clinicians that somehow imposing a people explanatory search frame on the motivation t o understand unexplained arousal transformed the ordinarily normal cognitive and emotional functioning of these college students subjects into pathological realms. Although the clinical judges saw only a relatively brief behavioral sample of all of our subjects, they decided in 78 of their 100 total judgments (50 subjects X 2 judges) that a particular student-subject should be allowed to continue onto a second experiment involving stress, and in 22 cases should not be allowed exposure to stress. Next, when asked to select those who would be judged most “normal” and the most “pathological,” if any, from their set of 10 subjects, 30% were assigned to the pathology category and 70% to the normal. Finally, and most amazing, was their willingness to offer clinical diagnoses with certainty in a third of the cases, that is DSM111diagnoses of Axis-1 (clinical syndromes) and Axis-2 (Personality Disorders). Diagnostic uncertainty was shown in two-thirds of the cases, where they either gave no diagnosis or deferred/provisional ones. Figure 5 graphs the clinicians’ diagnostic evaluations of pathology of our “normal” subject population by the percentage of such ratings in each experimental condition. On this index, scores above 70 (when summed across both raters) reflect greater diagnostic certainty of psychopathology. Set against the low levels in the two control conditions is the increasing level of pathology from Environment-Bias (30%),to Body-Bias (50%),up to the highest level among People-Bias (80%). These differences are highly significant conceptually as well as statistically. The p value for the ANOVA is <.002, whereas the contrast of the three experimental groups with the two controls is greater than .0001, and the People-Bias condition is significantly higher than the other experimental conditions ( p < .OOOl) or all the other groups ( p < .0001). When these results and the entire study were presented to the clinical interns and their supervisors, concern focused on the “symptomatic” evi-
443
DISCONTINUITY THEORY
100
I
1
People bias
Body bias
Environment bias
No bias
Arousal exlained
Fig. 5. Percentage of “normal” subjects judged with diagnostic certainty above D S M - / I / pathology criterion.
dence that may have been used to draw these diagnoses of pathology among these ordinarily sane students. Four causal candidates were rigidity, agitation, multiple attributions by the subjects, and the clinicians’ expectations. Confidence in making pathology judgments seems to have been greater when interviewees rigidly stuck to a particular reason for being distressed and were unwilling to even consider the alternatives posed by the interviewer; were physically agitated (notably one woman in the PeopleBias condition who began banging on the desk while recalling several confrontations); or gave several different attributions in close succession. In addition, because they had expected these college students to be rather normal except for the usual stress of academic pressures, the clinicians may have overreacted to the degree of emotional distress and cognitive confusion they witnessed in many cases. In a sense, our situation was the reverse of the one made famous by my colleague, David Rosenhan (1973), in his classic study of being sane in insane asylums. There the expectation was that anyone in a mental hospital who was not staff, must be crazy. Here the expectation was that anyone in a prestigious university must be smart and well functioning. So when the clinical judges observed someone who was relatively incoherent and dysfunctional, a contrast effect exaggerated the extent to which the evidence supported a pathology judgment. When psychiatrist David Spiegel reviewed some of these same tapes earlier as an ad hoc reviewer for the Stanford Human Subjects Research Committee, he saw many signs of distress in them, but not the level of pathology
444
PHILIP G. ZIMBARDO
reported by these judges. However, he was familiar with the context of the experiment and the reason for the subjects’ arousal. One final bit of data to support the claim of relative normality of the population from which our subjects were drawn comes from data on the general symptom inventory (GSI) given recently to several hundred Stanford Introductory Psychology students (Holman & Zimbardo, 1999). The student mean of .61 is less than-less disturbed than-the nonpatient adolescent norm of .76 on the highly correlated SCL-90-R (Derogatis, 1982, 1983). Nevertheless, in less than 1 hour of experiencing the discontinuity created in this research paradigm, some of these normal, healthy young men and women went “mad”-in the judgments of clinical psychologists and by their own assessments. G. RULING OUT DEMAND CHARACTERISTICS, WHILE BROADENING THE RANGE OF DIAGNOSTIC EFFECTS The next study had two aims, the first of which was to establish in a definitive fashion that results found with our hypnosis paradigm could not be explained away by resorting to notions of experimental demand characteristics in which subjects were consciously behaving as we had told them to, or as they imagined they should in order to please us. Instead, I want to argue that the manipulations initiate the predicted central motivational and cognitive processes that in turn generate the obtained outcomes found in the studies reported here. The second aim, and partially related to the first, was to broaden the range of measurable effects of coping with discontinuities by including a new psychophysiological measure-muscle tension-and several new measures diagnostic of emotional distress, cognitive worry, and interpersonal style. One way to handle the issue of demand characteristics is to incorporate an hypnosis simulator control condition. Low hypnotizable subjects (from the same general student population) are given experimental instructions identical to those for the unexplained arousal subjects, and told to act as if they were hypnotized in carrying out any tasks and answering any scales or questionnaires. They need to be low in hypnotizability because highly hypnotizable subjects might spontaneously enter hypnosis while simulating the state. We expect these simulators to respond comparably to the experimental group they are mimicking when response measures have high face validity and it is obvious how one might react while following posthypnotic suggestions. However, on more subtle, nonobvious response indicators, they should not be able to match the experimental behavior of the unexplained arousal subjects. If so, then our confidence
DISCONTINUITY THEORY
445
increases in believing that something more is driving the behavior of the experimental subjects than mere demand characteristics or experimenter expectancies. After describing the details of the simulator condition, I will outline the procedural features of this study that were rather complex, since its design required repeated assessments across many measures to detect specific changes at discrete phases of the experiment, as well as using a new arrangement in the onset-offset of the arousal manipulation. The low hypnotizable subjects scored only 1 or 2 on the Harvard Group Hypnotizability Scale (Shor & Orne, 1962) compared to 9-11 for the high hypnotizables. Like the highs who met in small groups for hypnosis-enhancement training, the lows met for the same amount of time to view a film on hypnosis that stressed the importance of simulator controls in the scientific investigation of hypnosis, and to listen to me lecture on and lead a discussion on myths, methods, and evidence for hypnotic phenomena. They were also made aware of their low hypnotizability scores. When they arrived at the laboratory, these students were given sealed instructions that read: “You will be instructed to simulate or pretend to be hypnotized during the study; to try to convince an observer that you are indeed hypnotized and following the hypnotic suggestions as are high hypnotizable subjects. They will be given suggestions to enter an hypnotic state and while in that state to experience certain reactions and to complete a series of ratings, scales, and experimental tasks.” The informed consent statement given to all subjects further reinforced this direct comparison between hypnotized and simulating subjects: “You will listen to taped instructions designed to induce a state of hypnosis. You may be given posthypnotic suggestions to act, feel, or think in certain ways. You may be in a condition in which you might react to the hypnotic suggestions by being deeply relaxed, confused, aroused, or distressed for a brief period of time. You might be in a control condition in which you are instructed to simulate, or role-play hypnosis-to act as if you were hypnotized even though you are not.” Throughout the rest of the study, simulators and hypnotizable subjects were treated identically, and research assistants were unaware of the status of any subject. Simulators were later given the same induction letter and heard the same tape (for relaxation, arousal, amnesia, and People-search bias) as did the Unexplained Arousal subjects. The instructions for the Explained Arousal group were the same as described in our previous research. The reactions of the nine low hypnotizable Simulators (Sim) were compared to twenty high hypnotizable experimentals randomly assigned to conditions of Unexplained Arousal (UA) ( n = 11) and Explained Arousal
446
PHILIP G. ZIMBARDO
(EA) ( n = 9). We previously assessed the psychological adjustment of these and other potential subjects on the Weinberger Adjustment Inventory (Weinberger & Schwartz, 1990), using only those whose scores were within one SD of the normative mean. Those in the Sim and U A groups were given the suggestion that what they were feeling “might have something to do with people.” To demonstrate the arbitrariness of the arousal cue and to examine differential reactions to its presentation sequence by the Sim and UA subjects, it was designed to be activated for the E A subjects when they started an experimental timer, and stopped when they turned it off. But for the other two conditions, stopping the timer was the onset cue that was to be sustained until the end of the experiment. The subject was placed in a special chamber equipped for psychophysiological recording, and communicated throughout the study by intercom with the experimenter. Sensors were applied to the face to record facial muscle stress at various phases of the study. SUA ratings were recorded at seven different critical points in the procedure, to more closely monitor changes in arousal related to various experiences subjects were having. Two other measures that were each completed three times during the experiment were Plutchik’s Emotion Circle (see Plutchik, 1980; Plutchik & Conte, 1997) and the Interpersonal Circle (Wiggins, 1979; Wiggins & Broughton, 1984), each to be described when their data are reported. Emotional reactions were assessed once on the MACL, and also on Endler’s (1983) Personal Affect Reactions Questionnaire (PARQ). The PARQ also includes assessment of cognitive worry. Postexperimental feelings and attributions were collected prior to full debriefing. Finally, an ambiguous stimulus was shown that could be perceived as a human or animal, and also the paranoia subscale (Pb) of the MMPI-to assess possible effects of the People bias manipulation. 1. Results and Discussion a. Arousal. All subjects reported being aroused at the onset of the arousal cue. While the E A subjects’ arousal was immediately reduced at the offset cue, the U A and Sim subjects’ reported arousal gradually diminished over the course of the experiment. There were no differences between conditions at the pre- or postmeasurement epochs, and both were at similar low levels (once again showing the effectiveness of our debriefing to return subjects to their pre-experimental levels). However, significant differences were found within the five other experimental epochs: (a) hypnotized; (b) arousal on for EA; (c) arousal off for EA, on for U A and Sim; (d) working on tasks; (e) debriefing. A repeated-measures, split-plot ANOVA on SUA ratings of the seven epochs, by condition and sex (about
447
DISCONTINUITY THEORY
equally divided in conditions) revealed a significant condition effect ( p < .OS), a robustly significant condition by epoch interaction ( p < .0001), and no effect of sex (dropped from subsequent analyses). A priori orthogonal contrasts were carried out for the two hypnotizable groups for the five experimental epochs, with high values for the arousal-onset epoch. Both analyses were highly significant ( p < .002). The most important comparison to be made here, and in other data sets, is the highest level of arousal reached by the U A and E A groups, in order to assess predicted incremental effects of amnesia for the arousal. From a comparable, low SUA level of 1.6 for E A and 1.9 for UA, the groups both rose to high levels immediately following the onset of their arousal cue-up to 4.8 for EA, but to a significantly higher SUA of 6.6 for U A ( p < .Ol), as can be seen in Figure 6. Unexplained arousal adds something beyond the impact of the same arousal whose source is known. The data for the Sim group both mimic the UA, yet differ from it in important ways. Neither pattern of orthogonal contrasts for the experimental groups was significantly different when applied to the SUA means for Sim. In general, they looked much like the UA group, with two notable exceptions. When the experimental subjects were at low of 1.7 during hypnosis, the Sim subjects “relaxed” mean was 3.3, an increase of 1.5 units from baseline. Next, when the on cue was given for the E A group the Sim subjects reacted most strongly, their mean of 5.1 being even higher than the 4.8 for the E A group, but recall that turning on the timer was not the arousal cue for Sim. Turning off the timer was the arousal cue for the U A
10 15 /
I
I
I
m
I
I
-
(O,n) experimental
8 -
< 3
I
I
7 -
I
1
‘\
5 -
-
\
\3,
(on) / /
-
\ \
4
‘s\
I
-
‘ (off1 \
(
O
f
f
~
~
O
-
-
-
~
-
2
0
I
I
I
I
I
I
I
.
Epochs
Fig. 6 . Mean subjective arousal ratings for experimental conditions across epochs.
448
PHILIP G . ZIMBARDO
group, and they increased their SUA from 3.4 up to 6.6 ( p = .02), whereas the Sim subjects increased only 0.9 units (ns) in response to their arousal cue from their “anticipatory arousal” prior level. Another measure of arousal comes from the facial muscle EMG data reported in Figure 7 for the two experimental conditions. The pattern of tension as recorded from forehead and cheek muscles varies systematically in predicted directions across the experimental epochs. Similar to the SUA data, the EMG means decrease during hypnotic relaxation, increase with arousal onset, and decrease over the remainder of the study to return to levels at, or lower than, initial levels. The U A group shows a heightened anticipatory arousal effect, of tensing up when the clock is activated, although it is the onset cue for only the EA group. However, again the most important contrast here is the significantly higher level of facial muscle tension at the peak of arousal for those experiencing unexplained than explained arousal. b. Emotions. Two measures of emotion/mood/affect come from the MACL, and the Emotion Circle. The strongest felt moods (on MACL) by
:I 9 1
E M G tPV
F 0 R E H E 6 A D C H E
E K
% o
t PRE
HYPNOSIS AROUSAL
(A-1)
AROUSAL
TASKS
POST
(A-2)
TIME + Fig. 7. Mean facial muscle tension for experimental conditions across epochs.
DISCONTINUITY THEORY
449
the E A group were rather mild and positive compared to those reported by the UA and Sirn groups. They were in order of intensity, Friendly, Content, Happy, Anxious, and Excited, but at relatively low levels from 4.9 to 3.6 on this 9-point scale. By contrast, the U A subjects were most feeling Tense, Anxious, Frustrated, Nervous, and Excited, with a high averaged value across these mood states of 7 on the 9-point scale. The more negative experience of emotion by the UA than the E A was highly significant ( p < .003, by contrast analysis following a significant ANOVA). The Sirn group again mimicked the UA group closely, with no significant differences between them on this measure. However, their strongest felt negative moods were less intense than those reported by UA, 6.4 vs. 7.0. The biggest differences between these two groups were the greater levels of Frustration (7.1 vs. 5.1), Tension (7.5 vs. 6.8), and Resentment (5.0 vs. 4.3), in the U A than the Sirn group, respectively. The Emotion Circle consists of 16 emotions arrayed around an outer circle and 6 inner concentric circles of intensity of emotions being experienced, from neutral to extremely. Subjects placed one mark anywhere on the circle to indicate both their feeling and its intensity, at three times: baseline, arousal epoch 1 for EA; and arousal epoch 2 for U A and Sim. Here is the first measure on which the simulators fail to match the reactions of the unexplained arousal subjects. During the baseline epoch, the experimental groups were similar in choosing as their average emotion as Optimism/Anticipation, with mean ratings around 1.7 (between “slight” and “somewhat”). At that time, the Sirn group selected Anticipation weakly, 0.5 intensity. During the first arousal period, these ratings shifted as follows: the EA shifted away from Optimism to more a more negative Anticipation (2.1 rating); the UA group was virtually unchanged; the Sim group became more positive in selecting as their representative emotion Optimism/Joy (1.8 rating). Differences emerged clearly in the next epoch, Arousal-2, as follows: EA, no longer aroused, moved four sectors positively from near Anticipation across Joy and Love to near Acceptance (1.4 rating); the U A group, now aroused, moved negatively from Optimism/Anticipation to Anticipation/Aggressiveness (1.3 rating); and the Sirn moved into the dead center of the circle at Neutral. Because these mean ratings might distort individual changes in opposite directions, chi-square analyses were conducted on changes in individual subjects in emotions at the various rating periods. There were no differences on either the baseline or Arousal1 epochs, but strong effects surfaced for Arousal-2 epoch. First, more E A subjects (89%) selected positive emotions than did the other groups who did not differ ( p < .01). Second, more U A subjects (82%) selected negative emotional reactions from among the sectors of AngeriAggressiveness; Aggressiveness/ Anticipation; Disappointment/Sadness; and Fear/Submis-
450
PHILIP G. ZIMBARDO
sion than the E A subjects (11%)or Sim. subjects (22%) This difference was significant between all three conditions ( p < .02), and also between U A and Sim ( p < .05). Also comparing the use of these four categories versus all others from the Emotion Circle, the U A group was again statistically different from each of the other conditions. What most characterized the emotional reactions of the U A subjects on this measure was their high intensity scores on the Aggressivity dimension, where more of them gave ratings of 3 or higher (moderate, high, extremely) than did those in the other two conditions ( p < .05). So on this measure, the Simulators did not duplicate the more intense, negative emotional reactions of the UA group. The emotional and cognitive components of subjects’ experience immediately after the second arousal cue was assessed by the PARQ, 20 self-report items answered on a 5-point scale of how much the affect or cognitive process is being felt “at this particular moment.” After the ANOVA revealed highly significant between-group differences for emotional arousal and overall reaction ( p < .005), with near significance for the cognitive component of worry ( p = .059), individual group multiple comparisons were made. They show once again that the U A group differed significantly from its E A counterpart on all three indices ( p < .02 or greater). The emotional arousal level for U A of 37 was higher than Endler’s clinical norm of 23, equal to it for worry, and much above it for overall negative reactivity (67 vs. 53, respectively). However, on this measure where the face validity is high, the Sim group did not differ from the U A group on any of the PARQ components, although U A was higher on each of them. c. Interpersonal Style. The Interpersonal Circle measure invites subjects to “imagine talking to someone right now,” and to decide how they would relate to him or her by putting an X anywhere on the circle to indicate their interpersonal style. The opposite poles are Controlling/Dominant and Submissive/Docile (top to bottom), and Hostile/Antagonistic and Friendly/ Cooperative (right to left sides). Clockwise we find Advising between Controlling and Friendly, Accepting between Friendly and Submissive, Sulking between Submissive and Hostile, ending with Criticizing between Hostile and Controlling. Like the Emotion Circle intensity is indicated by five inner concentric circles ending at the neutral center. This measure was taken at baseline and immediately after each arousal epoch. Subjects’ single-point responses in this circular space have x, y coordinates, thus could be computed for each dimension by coding them separately on the x-axis (FriendlyHostile) and on the y-axis (Controlling-Submissive) for quantitative analyses. Baseline scores were covaried in the analysis of Arousal-1 epoch, and baseline plus Arousal-1 scores were covaried in analysis of Arousal-2 epoch. The ANOVA on axis-y scores is significant for the three groups ( p = .Ol), caused entirely by E A group becoming much more Controlling/Dominant
DISCONTINUITY THEORY
45 1
(+10.4), whereas the other groups that were not yet aroused, had similar low scores on this dimension. On the warm-cold dimension of FriendlyHostile, the significant overall effect ( p = .01) was due to the EA group becoming slightly colder, while the other two, as yet unaroused, groups became much warmer. But the most powerful effect (that was delightful to behold) was the significant Arousal-2 effect ( p = .03) as the EA group became very warm and friendly (+ 14.6), while the Sim group moved toward Neutral (-0.2), but the U A group became more negative and cold (-6.2). Their scores qualified them as being Hostile/Antagonistic-reactions comparable to those in several of our previously reported studies. This complex index of interpersonal style was thus able to differentiate between those experiencing unexplained arousal with a people-focused search frame from simulators given the same exact instructions. d. People-Related Measures. On the next two measures, simulators also responded quite differently than UA, and surprisingly so. Although there were no group differences on the overall paranoia scale, there was a trend worth noting on a submeasure of the scale that identifies critical paranoid items, with the Sim group being much higher (2.9) than the U A group (1.0) or the E A group (0.7), with p = .08. Here the simulators over reacted in believing the UA group would respond with paranoid symptoms. However, this particular experimental setting mediated against that possibility since the UA subjects had little time alone t o ruminate about the meaning of their discontinuity; they were in regular contact with the experimenter in the next room, calling out answers and being aware that their physiological reactions were being continuously monitored. Another unexpected difference was on the measure of seeing people or animals in the ambiguous “RatiMan” picture. The significant condition effect ( p < .05) was due to all of the E A seeing people, most of the Sim group (78%) also seeing people, while our U A group was evenly divided in seeing people and animals. Thus the effect of the People-bias, imposed on the experience of a discontinuity, does not lead to a mindless perception of people everywhere, certainly not in a cartoon-like ambiguous figure, but perhaps only of “people” who in some way might play a meaningful attributional role in understanding the cause of the unexplained arousal. e. Attributions. Presentation of these results ends with but one more vital set of evidence in support of Discontinuity Theory, the explanations advanced for what the subjects were experiencing. Those whose arousal was explained all attributed their feelings of arousal to the letter they had read. The simulators again overshot their mark, by giving exaggerated, pathological explanations that differed from the U A group’s explanations. Typical Sim attributions were “People are after me; out to get me, trying to beat me up”; “People are watching me.” Or, they reiterated the instructions,
452
PHILIP G . ZIMBARDO
“People have something to do with it.” There was a sharp contrast with the more thoughtful, varied attributions of those using the People-bias to help make sense of their unexplained arousal: “My girlfriend makes me feel jealous of her; it gets me mad because I feel so inadequate”; “It’s hard to know what people expect of me, so I find it hard trying to deal with people”; “I am having trouble with people at work”; “I’m starting to panic because there should be people around here and there aren’t any, so I am alone.” Following debriefing, all subjects understood the reason for their arousal and their emotional reactions returned to the normal level they were initially. This pattern of evidence helps rule out experimental demand characteristics as an alternative explanation for our previous results and the current ones. Simulators were able to match the reactions of the discontinuity group on measures where it was apparent how one should behave. But on some of these measures they overreacted in being much more extreme than the UA group, and on other more subtle or complex measures where it was not so clear how the UA might respond, the simulators were undone in their desire to mimic them. The second aim of this research was to extend the realm of observed differences between those experiencing discontinuities and those with similar arousal but aware of its origin. I think we have done so with the consistently significant differences found in this study on measures like: SUA, facial muscle tension; Mood Adjective Checklist (MACI); PARQ; Emotion Circle, and Interpersonal Circle. We did not see in this study the more pathological effects found in the previous studies for those with the phenomenological perspective of feeling aroused, without knowing why but thinking people might hold the answer. It is most likely due to the severe constraints imposed on the subjects by the restrictive facial physiological sensors, the many tasks, and the frequent interactions with the experimenter in the control room. All of which may have combined to limit time for reflection, memory search for available instances of relevant people, rumination, and some solitary incubation period that may be necessary to sow the seeds of pathological thinking and feeling.
H. FAILURES OF INTERPERSONAL DISCLOSURE OF DISCONTINUITIES This research was stimulated by the case study example noted in an earlier section about the workers in the Bud Plastics Company suffering from neurotoxic-induced deficits in a range of behavior, thinking, and emotions. While unaware of the subtle environmental cause (chronic exposure to toxic substances on the job) of their severe dysfunctional state, they
DISCONTINUITY THEORY
453
chose to conceal their anomalous reactions from co-workers. They probably did so out of embarrassment, concerns about appearing strange, and fears of getting laid off. But had they disclosed their reactions and confusions immediately, it would have been possible to identify common situational elements in the workplace that were the causal agents for their shared symptoms. Their treatment then would not have been the unnecessary psychiatric care for mental disorders, but the required medical care, and it would have prompted earlier corporate prevention actions. However, the failure to disclose maintained the idiosyncratic interpretations of each affected man that “something is wrong (only) with me.” When this view was aggregated, it created a state of general pluralistic ignorance that was a public health menace. Disclosure to others of strong emotions, anxiety, distress, and uncertainty may have the potential gains of discovering an external, situational cause for one’s symptoms, one that might be modifiable, as well as sympathy and understanding. On the other side of the scale are the anticipated costs of revealing a significant discontinuity to others in terms of vulnerability to negative social appraisals if these others do not share the same reactions, or may not show “attributional charity.” Getting such unsupportive information pushes one toward concluding the problem is personal, dispositional, and deviant. The issue of disclosing anomalous experiences raises a social dimension embedded in Discontinuity Theory. Many reactions to discontinuities are social to the extent that they involve other people who may notice and become concerned about the actor’s atypical reactions, or are personally affected by their consequences. I have assumed that a basic motivation of the actor in a discontinuity scenario is a striving to sustain a positive selfimage for one’s self, and to project it to others. Given the etiological ambiguity of some discontinuities, especially the ones created in our research, the decision to disclose or not pits concerns for self-image maintenance against concerns for cognitive clarity and developing problem-solving coping strategies (Lazarus & Folkman, 1984). Research on disclosure makes it clear that people do not make intimate self-disclosures to strangers (Snell, Miller, & Belk, 1988) because it would be judged inappropriate (Chaikin & Derlega, 1974). And disclosure is less likely as the dissimilarities in dyadic partners increase (Stephen, Wenzel, & Cornelius, 1991). However, since disclosure is often symptomatic of underlying anxiety (Stiles, Shuster, & Harrigan, 1992), it could be expected to decrease feelings of arousal and anxiety (Sholle, 1992) in some interpersonal settings. I wondered if it would do so in our experimental setting. Two related studies were conducted in our laboratory (Zimbardo & Williams, 1998) to determine the extent of disclosure by subjects experienc-
454
PHILIP G. ZIMBARDO
ing unexplained arousal, and to study the consequences of such disclosure. We also wanted to explore the specific tactics and strategies of interpersonal disclosure used by various subjects. For instance, would they first question the other person about their feelings before giving up any personal information, or begin by putting out some minimal self-revelation to prime the other to reciprocate? The first study examined pairs of subjects who were both experiencing our unexplained arousal manipulation, at the same time, in the same setting. The second study used different types of confederates to try to stimulate disclosure in subjects experiencing unexplained arousal. Both studies will be outlined only briefly because of their disappointing findings of failures to disclose discontinuities. However, their methodologies are sufficiently interesting, as are a few results, to warrant some consideration, especially in planning new research on this unstudied realm of disclosure. The first study sought to examine disclosure among student-subjects in twelve dyads where each person was experiencing unexplained arousal. Four of the dyads were male, four female, and four mixed gender. Under the cover story of solo versus team problem solving and creativity when hypnotized or in ordinary consciousness, individual students were given the standard hypnotic induction tape with posthypnotic suggestions for later cued unexplained arousal in part 2. Then pairs of students were brought together to work on a series of problems and tasks, some alone, in the presence of the experimenter, and some as a team, without the experimenter present during the disclosure phase. O n the solo tasks (writing creative TAT stories, solving word puzzles), a barrier separated the subjects. It was taken down as they started a shared anagram task on which they were encouraged to discuss solutions (to prime later verbal disclosure). The start of the next shared task included the posthypnotic arousal cue. The experimenter started an experimental clock for the 5-min task, left the pair alone, and discreetly videotaped their interaction. In general, there was almost no disclosure of the high levels of arousal and uncertainty that both members of each dyad were experiencing (as determined from their posttest ratings). In only 2 of the 12 dyads was there any evidence of disclosure of emotions, or of the agitation and confusion created by the manipulation. In one of these dyads, the disclosure of the first person was seconded by the other, but quickly attributed to their mutual concern for final exams, which ended the search for causes in the experimental context. The second disclosing dyad came close to discovering the true situational cause of their common, unusual feelings as somehow related to hypnosis. But after some discussion about the hypnosis training and its usually relaxing effects, they dismissed it as the cause of their current
DISCONTINUITY THEORY
455
tension. Although the latter pair were females and the former was a mixed gender pair, the results are too lean to make any gender inferences. Analysis of the tape-recordings revealed that distraction was the most common tactic employed during the period when the subjects might be privately disclosing to each other. They quizzed each other about schoolrelated topics and avoided mention of the research underway. Such distracting talk may be a tactic to lower one’s arousal, but it reduces the probability that the other person will then move from this banal communication mode to a more intense, personal mode involved in interpersonal disclosure of strong, confusing emotions. Also it allows the misinterpretation that the other person is not similarly aroused if they are casually talking about courses, sports teams, dorm food, and the like. The next experiment attempted to correct this problem by having an aroused subject interact with a confederate whose overt reactions clearly established an emotional similarity. In this study, 33 high hypnotizable students were randomly assigned to either a condition where a confederate acted tense (CT), or a confederate acted calm (CC) while the subject was experiencing unexplained arousal. The gender of the confederate matched that of the aroused student, 7 males and 10 females in CT, 7 males and 9 females in CC. We again used the cover story from the first experiment, but added repeated SUA measurements, and appropriate reasons for the confederate and subject to be separated at times and to be close together at other times. The first of five SUA ratings was made before starting an initial solo task. This was repeated after hypnotic relaxation, arousal induction, immediately after the disclosure period, and finally after debriefing. The taped hypnotic induction process was followed by the next SUA rating and the first of several mood ratings, Profile of Mood States (POMS) (McNair et al., 1971), and then oral answers were called out to word association and Stroop color naming tasks. Confederates were blind to the experimental hypotheses, and even to the knowledge that subjects were hypnotized, since they could not hear the taped hypnosis induction and were unaware of the contents of a letter subjects read privately informing them that they would experience somatic arousal (when activating a timer) and would have amnesia for the suggestion. Following the solo task, the confederate was seated next to the subject to begin their shared creativity task. For 4 min, they were to write a story about a T A T picture by each writing one sentence and passing back and forth their response sheet. Next they did a similar task to a different TAT card but solo for 3 min, with each starting their own timers (arousal cue). In the CT condition the confederate began to make sounds and gestures intended to convey that she or he was feeling tense, which continued
456
PHILIP G. ZIMBARDO
throughout the rest of the writing and disclosure periods. In contrast, in the CC condition, the (same) confederate acted calm, smiled, and was at ease during this task and subsequently. After completing another SUA and POMS, the experimenter pretended to have made a mistake in not having the next form available in the folders of the two students, and excused himself for a few minutes while he went upstairs to get copies of them. This was the occasion for the subject and confederate to be alone, and ideally to disclose their thoughts and feelings. During the 4-min Disclosure Period, videotaped by a concealed camera, the confederate enacted a scripted role by saying nothing for the first minute (if the subject did not speak) then asked, “Which of the two stories did you like writing best?” Any input by the subject was responded to in kind, but the confederate did not initiate conversation about hypnosis, emotion, or the experiment. If asked how the confederate felt, he or she replied either, “I feel fine” (CC), or “I don’t know, I guess 1 feel a little tense” (CT). When the experimenter returned with the missing forms to be completed separately, the postdisclosure SUA form and POMS were completed and the subject was then interviewed by me about her or his current feelings and awareness of the reason for them. Subjects’ arousal and amnesia were lifted, they were fully debriefed, and they completed the final SUA.
1. Results and Discussion The basic conditions for testing our hypotheses were clearly met: subjects experienced unexplained arousal, and the differences in the confederates’ behavior were perceived veridically. Arousal level, as measured by SUA ratings, went down during hypnotic relaxation (change from initial baseline = -3.2), went far up following the arousal manipulation (change from relaxation = +3.8), down after the Disclosure Period (change from arousal = -1.O), and further down after debriefing to a final level that was much lower than the baseline (change from disclosure = -1.4). (This is further evidence of the efficacy of our debriefing procedure, as described in the later Ethics section). Each change was highly significant ( p < .OOl). All but three experimental subjects gave evidence of complete amnesia for the arousal suggestions (and they were deleted from all analyses). Those in the CC condition rated their partners as quite relaxed (high mean of 67 on 100-point scale), and significantly more so ( p < .001) than those in the CT condition (low mean of 29). The tense confederates were seen as significantly higher in anxiousness than were the confederates who had acted calm (57 vs. 20, p < .005), in agitation (50 vs. 12, p < .OOl), and in irritation (35 vs. 9, p < .005). Clearly, the confederates’ scripted behavior
DISCONTINUITY THEORY
457
elicited divergent perceptions from the subjects in the two confederate conditions-of reacting like me (CT) or dissimilar to me (CC). Another dimension of the unexplained arousal manipulation was manifest in the content and affect of subjects’ solo TAT stories. A story’s content was rated “violent” if its central or secondary theme explicitly involved violence, murder, or rape. Affect of the stories was rated on a 10-point scale, from 1 = very negative to 10 = very positive. Comparisons of these ratings with those from a Control Group (of 18 high hypnotizables not given any treatment, 8 males, 10 females) indicated greater negativity in the stories of those with unexplained arousal (4.6 vs. 3.1, p < .001). There was also more violent content in the stories of the males experiencing discontinuity than in those of either all the Controls or the female Experimental subjects. Those comparison groups’ data were violence-free, while 42% of the stories of the aroused males were aggressive ( p < .05). Those who wrote such hostile stories during their arousal phase were also higher on the POMS TensiodAnxiety subscale (p < .05), and this POMS measure correlated significantly ( r = .49) with the degree of negative affect in those stories. Thus strong conditions were created that should have encouraged disclosure to another student who was in the same discontinuity situation and was also clearly demonstrating his or her similar arousal/anxiety. However, once again, very little disclosure occurred in any condition. Only 7 (23%) of the 30 aroused students disclosed at all, with no differences in gender or confederate condition. These few met the most minimal definition of disclosure, that is, the subject made a statement about what he/she was feeling-as reported by either confederates or the two videotape recording raters. And then subjects rarely followed through when the confederate acknowledged feeling similarly. This low rate of disclosure mimics our earlier results. Three additional results are worth mentioning for their value in future research. Those who disclosed were more highly aroused (7.0 vs. 6.2, p < .05) prior to the disclosure period than those not disclosing, and they rated themselves as more creative on the POMS than did non-disclosers (71.7 vs. 51.7, p < .05). The only really positive result from my perspective is the final measure of the consequences of disclosure for those few who did so. There was a significantly greater decline in their arousal level after having disclosed than among the non-disclosers (-2.0 vs. -0.8, p < 4.5). (The general slight decline in arousal over the course of the experiment is a function of adaptation to the arousal, which we have found in all of our research.) Why the failure to disclose in these studies? Several alternatives present themselves. First, it may be that the self-image costs of disclosing confusing
458
PHILIP G. ZIMBARDO
feelings of strong arousal/anxiety to a stranger are simply greater than the perceived gains in terms of enhanced understanding of a one’s discontinuity. We know that people rely on different self-presentation strategies when relating to strangers than friends (Tice, Butler, Muraven & Stillwell, 1995). But this tendency for greater reliance on self-enhancing strategies in interactions with strangers is put in competition with the potential risks of disclosing one’s unexplained arousal states, without even the assumed sympathetic ear of friends. We need to know more about this situation because of the practical consequences of learning how to create optimal conditions for effective self disclosures of strong, negative affective states that may be precursors to psychopathological reactions (Derlega, Margulis, & Winstead, 1987). Future research should focus on eliciting high degrees of disclosure of discontinuities by using novel experimental strategies, such as, (a) having friends be confederates trained to induce disclosures from their buddies, to make them feel better; (b) informing subjects that their partner has successfully navigated the same situation, is nonjudgmental, good at psychological problem solving, and has other traits that might lower the threshold for disclosure; and (c) make the experimental subject’s task to elicit or force disclosure from their partner-confederate under the guise that the partner is the target of research interest. A second reason for the failure to find disclosure in these studies may simply be traced to the power of the hypnotic manipulation. The suggestion of not being able to remember why they were feeling aroused may have created a mental barrier in the subjects against sharing any information with others that might breach the amnesia, and thus “spill the beans.” This potential constraint might be relaxed by new hypnotic suggestions that imply subjects might understand why they are feeling aroused while they exploring possible reasons with certain other people. Alternatively, other research needs to be done that does not use the hypnotic paradigm to generate the discontinuity to be disclosed. But these last suggestions all raise additional concerns about the ethics of such deceptive research and the elicitation of emotional arousal and cognitive misdirections. I address those ethical issues next.
V. Raising, Reconciling Some Critical Ethical Issues Since the research presented in this chapter was explicitly designed to make normal, healthy college students “mad,” if even for a short while, it raises serious ethical issues. It was necessary to address and resolve them
DISCONTINUITY THEORY
459
by working closely and creatively with Stanford’s IRB in order to develop special procedures that enabled this research to be conducted while protecting the rights, dignity, and well-being of our student-research participants. This section describes how that trade-off was handled. A psychological colleague, critical of some of the present research, as well as my earlier Stanford Prison Experiment (SPE) (Zimbardo, 1972; Zimbardo, Haney, Banks, & Jaffe, 1973), asserted, “It is difficult to imagine more extreme instances of deception than those provided by Zimbardo’s experiments” (Baumrind, 1985, p. 167). A professor of pediatrics raised both ethical and scientific issues in his critique of the paranoia study reported earlier: The question to be considered here is the advisability of inducing paranoid behavior in a previously healthy individual. Aside from the other ethical questions . . . there is the questionable act of inducing a state about which we do not know a great deal, (Lewis. 1981, p. 9).
Science writer Morton Hunt (1982) centered his entire New York Times Magazine essay, “Research Through Deception,” around our paranoia study (Zimbardo et al., 1981) while exploring the broader issue of social psychology’s use of deception paradigms. The debate within social psychology about the use and misuse of deception in experimental research has sensitized investigators about human subjects issues, and may be responsible for markedly reducing the number of studies employing deception in recent years. The advent of a cognitively focused social psychology also obviates the need to disguise experimental manipulations with “cover stories” and other forms of misdirection, since behavioral studies like those of Milgram (1974) and mine are replaced by deception-free, “quick-and-clean,” paper-and-pencil questionnaire studies, where respondents only imagine how they might react to hypothetical scenarios. I believe that experimental social psychology still requires some behavioral research that involves “subjects” more personally and profoundly in studying a host of issues that are vital to the central themes of our discipline. Readers are referred to my views on absolute versus relative ethical principles in human experimentation (Zimbardo, 1973) as summarized in reaction to published criticism of the SPE (Savin, 1973), and to my reply to Lewis (1981) about the ethics of the induced paranoia research (Zimbardo, 1981). Ethical issues of research using hypnosis are well presented by Coe and Ryken (1979). Here, I want to discuss briefly the following three points: (a) the necessity to employ some forms of deception in particular research paradigms when the phenomena of interest could not be investigated with-
460
PHILIP G . ZIMBARDO
out it; (b) the special demands imposed on researchers, who believe they must use deception and other ethically questionable procedures, to be extremely sensitive to all the ways they can minimize risks and promote the general well-being of their experimental participants, while also actively maximizing the potential of their research for scientific and societal gain; and (c) the need t o increase the “gain factor” in the gain-loss equation for any ethically sensitive research, by insuring that the participants (and sometimes researchers) themselves derive maximum benefit from the power of “experiential learning” that may occur in such experiments. A. WHY DECEPTION AND DISTRESS MAY BE NECESSARY T O STUDY THE “DARKER SIDE OF HUMAN NATURE” It is my belief that one cannot sit on the side of the angels and study phenomena in the realms of social pathology or individual pathology using experimental paradigms that seek to discover causal relationships between situational, dispositional, and behavioral variables. By their very nature, such investigations move us beyond the boundaries of conventional research and force social scientists to face daunting ethical and moral dilemmas. In trying to understand the nature of the dynamic processes involved in powerful societal experiences-such as failures of bystanders to intervene in emergencies, group conflict and its resolution among children, blind obedience to unjust authority, the effects of anonymity or dehumanizing labels on aggression, the power of prison environments to elicit social pathology in good men, or the initiation of normal people into pathological ways of thinking and acting-some critical dimensions of the relevant situations must be functionally re-created by researchers, and experienced directly by participants. Those subjected to such procedures typically cannot know in advance all that will occur, otherwise, self-selection of those who stay versus those who quit distorts conclusions intended to generalize to all. For example, to demonstrate one of social psychology’s enduring lessons, the power of the situation, individuals must experience its force first-hand from within its dynamic crucible. Reflective imagination cannot replace such direct involvement. Recall that 40 psychiatrists predicted that fewer than 1%of U S . citizens would go all the way up to 450 volts (only the sadists, they said) after they had listened to Milgram’s experimental scenario. Aside from their dispositional bias (by virtue of medical school training), from their uninvolved and remote vantage point, they could not fully appreciate the situational forces impinging on the majority of average, normal, research subjects who delivered maximum shock to the victim. O r consider the design of a
DISCONTINUITY THEORY
461
proposed study where deception is the independent variable in establishing whether it could be eliminated in testing certain categories of causal hypotheses. To show that deception was superfluous, it would be necessary first to include a deception versus no-deception pair of research conditions. But that would not be permissible by current standards. In my research, on the “psychology of evil,”-deindividuation, vandalism, prisons, and cults (Zimbardo, 1978), and more recently, institutionalized training of torturers (Huggins, Haritos-Fatouros, & Zimbardo, 1999)-I have been interested in the social conditions responsible for recruiting good people to engage in evil deeds. The research focus was on elucidating the process of human transformation among those who believed they would never do such bad things (“the just people”) into becoming those who crossed the line to do so (as “the unjust”). Such knowledge may be used to identify the situational variables and processes involved so as to avoid, escape, or change them. The current set of studies, which can be classified as experimental-social psychopathology, are designed to explore the earliest stages in the initial processes of arousal, confusion, and distress (in response to significant personal discontinuities) that I believe may be responsible for the transformation of normal individuals without “premorbid” personalities into those who begin to suffer from some forms of mental disorders. The goal is to recommend new diagnostic and treatment modalities to clinicians based on the underlying conceptions, as well as to better inform the public about how misattributions could lead to pathological reactions. Critics of the SPE have argued that everyone already knows prisons are violent places, so why do a study that shows it again? However, what our research showed was that it was the situational features of prison-like environments, not the “negative” dispositions of guards or prisoners, which creates violence. This is a more subtle, and I believe, more important insight. Critical reading of our induced paranoia study led to a similar rejection of the need for this empirical demonstration that deafness can lead to paranoid behavior among otherwise normal people. The counterargument being, “it is a generally acknowledged clinical observation that unexplained deafness in older people induces suspiciousness” (Baumrind, 1985, p. 172). However, clinical observation may inform, but it does not confirm. Systematic, controlled experimental research is essential to go beyond anecdotes, observations, and correlational analyses if understanding causality and designing controls (in the form of therapeutic interventions) are among the goals of psychology and psychiatry. The field of psychology, and especially those who claim the higher moral ground, must answer the question I have asked myself often. Would it be better if none of the following “unethical” studies had ever been done: the Milgram (1974)
462
PHILIP G . ZIMBARDO
obedience studies; Latank and Darley (1970) and Darley and Batson (1973) unresponsive bystander studies; or the Stanford Prison Experiment (Haney & Zimbardo, 1998; Zimbardo, Maslach, & Haney, in press)? Is that knowledge gained about the human condition worth the alleged harm to the subjects, to society, and to the profession of psychology (as argued by Baumrind, 1985)? At the core of the ethical dilemma for social scientists is creating a balance between what a given researcher believes is necessary for the conduct of socially or theoretically significant research and what will safeguard the well-being and promote the dignity of research participants. Since researchers’ self-serving biases may push them toward more of the former than the latter, external reviewers need to serve as ombudspersons for the relatively powerless participants. But these IRBs must also act in the interest of “science” and “society” to determine whether, and to what degree, some deception, emotional arousal, or other aversive states can be permitted, assuming the negative impact of such procedures is not likely to endure beyond the confines of the experiment. Let’s consider next how those competing interests were served in the present research.
B. SPECIAL CONSTRAINTSiDEMANDS IMPOSED ON ETHICALLY CHALLENGING RESEARCH Throughout the many years of the current research program, I worked closely with Stanford University’s Panel for Human Subjects in Behavioral Science Research to help develop and maintain the delicate balance between what I believed was theoretically novel and clinically significant experimental research on one side, and the necessity for minimizing risks to subjects while maximizing the educational value of their participation, on the other. Let me outline how and why I was allowed to conduct this research program and the ways in which I went beyond the tough demands of this IRB to raise the standard for debriefing and concern for the wellbeing of my research participants. It is my aim to communicate to both researchers and IRBs that ethically challenging research can still be conducted in an ethical fashion. 1. I appeared a number of times in person at the IRB to discuss the nature of the research, its methodology, and to consider alternative approaches. The entire committee met in my laboratory to view randomly chosen videotapes of pilot research subjects undergoing the experimental tasks. The committee also reserved the right to monitor any study in person or to view tapes on demand and to rescind approval if the monitor believed that subject reactions were extreme enough to warrant terminating the research.
DISCONTINUITY THEORY
463
2. I was required to personally conduct all the hypnosis training, and to be present as experimenter or observer in the control room during all experimental testing. In addition, I was required to be assisted by a clinical psychologist or to have one on call to cope with any emergency reactions that might arise. 3. An expert on hypnosis and psychopathology, Dr. David Spiegel of Stanford’s Psychiatry Department, served as the IRB’s ad hoc reviewer to evaluate the paradigm and the impact on participants based on viewing videotapes from various experimental conditions. He approved continuing the research. 4. Extensive pretesting of each aspect of our research methodology included lengthy discussions with participants about their immediate and delayed reactions, interpretations of the manipulations, and the adequacy of the debriefing procedure. All of this helped minimize risk and increase subject power. Those goals were achieved by (a) using the briefest possible duration of arousal (b) inducing arousal by means of minimal physical symptoms rather than “anxiety” priming; (c) cuing arousal to situationally specific stimuli unique to our laboratory setting; (d) developing “partially informed” consent forms (described below) for each separate component of our procedure; (e) utilizing a formally structured triple-phase debriefing; (f) not terminating the debriefing of any subject until the staff consensus was that he or she was responding again in a normal, healthy manner; (8) conducting 24-h phone call follow-ups for all experimental subjects, as well as long-term evaluations (up to a year later) in some studies; and finally, (h) enabling subjects to withdraw their data or to use it selectively after they had been thoroughly debriefed. 5. Subjects completed five separate consent forms, one each for hypnotic susceptibility screening; hypnosis enhancement training; any personality testing; the experiment proper; and use of their videotape data following participation. The partially-informed consent stated: “The researchers cannot fully disclose to you in advance all of the information about which you may experience. However, they promise to do so in a thorough debriefing session at the conclusion of your participation.” They were also told orally and in writing that they could quit at any time without penalty even after signing the consent forms. Each form was completed just prior to engaging in the task so its content would be in the subject’s awareness during the task. Consent to use the subjects’ audio or videotaped reactions was separated into three categories of increasing public use, from only scientific data reduction to educational use in classrooms, and to professional use at scientific meetings. A fourth consent was requested to use the rest of their test data for scientific research purposes. Students were further informed that even after having given their consent, they could change their mind
464
PHILIP G. ZIMBARDO
at any time in the future, and whenever they withdrew their original consent in writing, we would destroy their data. Although none did the latter, there was variation in the degree to which they were willing to give consent to the various uses of their data. 6. A highly structured, thorough debriefing (lasting longer than the entire experiment, usually 45 min to 1 h) was conducted by either myself, a clinical psychologist (Carlo Piccione), or an advanced graduate student working closely with me (Susan Andersen o r Lisa Butler). The three-page debriefing procedure outline made explicit that the researcher was to conduct the initial inquiry into the subject’s emotional and physical state and beliefs about the nature of the research “within a supportive interview context between researcher and student participant.” (For specific details about this and the other aspects of my IRB protocol, see Stanford Sponsored Projects Office Protocol: 8586-77; renewal of 845-101, 12/13/85). The first part of the three-phase debriefing was conducted with the subject in the alert state, following lifting of hypnotic arousal, amnesia, and any search frame biases. Part 2 involved rehypnotizing subjects and instructing them to “disconnect” their arousal (and search frame explanations, where given) from the experimental stimulus cue so that those reactions would no longer be elicited by similar cues in their everyday lives. To ensure that these instructions were understood and internalized, the disconnect instructions were administered again in Part 3 when subjects were again in the alert state. The content of the debriefing included a full presentation of the experimental hypothesis, and of the purposes and justifications for using deception. We ended with the statement of our personal concern: “We are sorry for not fully informing you of our procedures in advance, but we hope you can appreciate that it would not be possible to study the effects of unexplained arousal if we explained everything first.” Each subject was then fully “dehoaxed,” by informing him or her that any behavior they displayed was due t o the experimental manipulation and was not unusual or atypical in this setting, since most other students react comparably in the condition to which they had been randomly assigned (and thus any unusual reactions should not be seen as idiosyncratic or personally symptomatic). The “desensitization” then involved establishing that the mentalphysiological state of the subject was now comparable to what it had been initially, and if not, “the researcher must do all in his power to bring about that process of appropriate desensitization of distress and a return to normalcy.” Subjects then completed a form describing what they now knew about the nature of the research and the true cause of their reactions. If there was any indication that our debriefing had not been adequate, the subject was rehypnotized and given the relevant instructions to enhance memory,
DISCONTINUITY THEORY
465
mood, or understanding. Following all this, subjects gave their consent for data use, were asked to maintain experimental secrecy, and were given the office and home phone numbers of the principal investigator and clinical psychologist to call if they experienced any unusual reactions. 7. Further respect for our research subjects was shown by providing them with special training in several hypnotic phenomena unrelated to the research, but of potential value to them personally, such as concentrated focus during academic study and testing, stress management, pain control, and self-esteem building. 8. All potential subjects completed a Medical Evaluation Survey, Health Survey, and premeasure of manifest anxiety and/or selected scales of the MMPI. No student was included in the research sample who (a) had scored outside the “normal-average’’ range on any of these measures, (b) was experiencing any symptoms; (c) was in therapy or on psychiatric medication; or (d) who gave any indication of being in other than “good, well, or great” health at the start of an experiment. 1. Reframing the CosdBenejit Equation for Ethicul Analyses of Research In my 40 years of research and teaching experience, it has been my consistent impression that the majority of students engaged in research using deception-and given adequate debriefing-find it more interesting than most other research they participate in as part of an introductory psychology course requirement. They are less distressed at being fooled than they are surprised that they could be deceived. Moreover, they readily appreciate the reason why deception was necessary for testing a particular hypothesis, when the justification for the “cover story” is carefully presented. In addition, some of this research elicits “dramatic” individual reactions during the study, such as agreeing to take more painful shock when in a high dissonance condition, or giving shock to another woman when the subject is in a deindividuation condition. Those reactions become part of the experiential learning of the participants to understand how and why they behaved so. This is in contrast to most other research where any participant’s minimal experimental reactions are not individually salient but must be aggregated across all subjects to demonstrate any effect. In our research program, we conducted follow-up surveys several weeks and up to a year after each study, with subjects indicating the positive and negative effects of each part of their experience in the research. For example, in a 1-year follow-up survey of the “three bias study” study (with a high 74% return rate) none of 37 subjects reported any negative effects of the hypnosis training whereas 26 (70%) reported positive effects. When
466
PHILIP G . ZIMBARDO
asked about any lasting effects of participating in the discontinuity experiment proper, most reported that it had none, but 5 (14%) reported positive effects, and only one reported a minor negative effect of remembering a TAT slide that had made him upset during the experiment. Finally, students reported on the personal value, if any, of having participated in this research on a 100-pt scale, where “0 = of no value, preferred not to participate”; “50 = moderate value, about as much as typical psych. experiments”; and “100 = of great value, pleased that I participated.” The mean for these 37 students was a high 83.0, with a positively skewed range of 60-100. Thus, all subjects rated the value of being in one of the currently reported studies as higher than the other research in which they had participated (four other required studies in their introductory psychology course). Among the gains reported from having had this unique experimental treatment and debriefing are the following personally valuable changes reported by some participants in this follow-up survey: (a) “Now I can notice these ‘strange’ things in myself if they ever come up, also I can see these results applied to friends and relatives” (People Bias Condition); (b) “increased self-awareness, ability to focus, more self confidence and ability to control my emotions” (Body Bias Condition); and (c) “learned how to relax and how to prepare myself for potentially tense or stressful situations” (Environment Bias Condition). Such mindful reactions seem to me like more than simply dissonance-reduction reactions following a tension-filled experience. Following the SPE we got similarly positive feedback from our former prisoners and guards. They reported that they learned more about themselves from their atypical behavior in that mock prison than from most other ordinary situations they experienced regularly (see Zimbardo, 1975). Moreover, in some individual cases, participating in the study profoundly changed their entire lives in prosocial ways. For example, one ex-prisoner became a forensic psychologist working in the California prison system to improve prisoner-guard relationships (for other examples see Zimbardo, Maslach, & Haney, in press). Also noteworthy is the impact of such research on the researchers who may also be changed in positive ways by their experience (see Haney & Zimbardo, 1998; Zimbardo, et al., in press). We also felt that it was incumbent on us to go beyond simply publishing our research in academic journals to optimizing its impact on opinion leaders and by “giving it away to the public” in various ways (see Miller, 1980). The problem with the usual cost-benefit analysis of ethical decisions regarding research is that the costs to the subjects are tangible, real, and upfront. The alleged benefits to science and society are probable and delayed, if any. Many experiments don’t work, don’t even get published, don’t
DISCONTINUITY THEORY
467
get noticed, don’t have action or policy implications, don’t change anything. Therefore, those that do add special demands on the researchers to use all their resources to disseminate their message widely, to become advocates of their viewpoint, and even willing to act as social-change agents for their recommendations (see Zimbardo, 1975). Perhaps an appropriate way to conclude this section is to return to the critique of deception research by science writer Morton Hunt (1982) in his provocative New York Times Magazine essay. Hunt used as the journalistic device for his piece an in-depth analysis of one student-subject, Steve K., in the induced paranoia study, who was distressed during the study by feeling paranoid, but upset after the study because he had expected to be called back to be in further research using hypnosis and was not. Two years after his research participation, Steve is described by Hunt as looking back “on the period of his participation in Zimbardo’s research as one of the high points of his life thus far” (p. 145). That his experience with induced mood changes and deception made this young man think deeply about the ethics of research, is evident from his final statement to the investigative reporter: “I agree with people who say it’s not right to deceive human beings; it’s not right to treat people as if they were mice. But I agree with Professor Zimbardo that he couldn’t do his work on paranoia and deafness without deceiving his subjects, because if they knew what was going on, they wouldn’t react the same as if they didn’t. I can see both sides. That’s my dilemma, and I don’t think there’s any simple answer to it, only complicated ones” (p. 145).
VI. Conclusions and Reflections So there you have it-a simple conceptual model elaborated to incorporate and integrate social, cognitive, and clinical concepts, with a set of supporting experimental evidence. Enough of the predictions advanced from Discontinuity Theory have been validated to warrant its serious consideration, refinement, and extension by other scholars, experimenters, and clinicians. After 30 years of working on this program of research, I feel comfortable in drawing some tentative conclusions, and in proposing some reflections about future research and applications. I will start with a few grandiose ones and work down to the more mundane. The seeds of madness can be planted in anyone’s backyard, given transient perturbations in the life cycle of ordinary experiences. I think the general approach proposed here can help rescue the study of psychopathology from the confines of psychiatry and clinical psychology since it suggests
468
PHILIP G. ZIMBARDO
that fundamental cognitive, social, and cultural processes are involved in its development. Bringing the knowledge we have in those realms to bear on a fuller understanding of the mechanisms involved in the transformation of normal behavior into dysfunctional, symptomatic behavior can yield important insights regarding prevention, diagnosis, and treatment. Rather than seeking to find evidence of “premorbid” personality factors in those with “mental disorders,” we need to switch out of this medical model to a more socially oriented, public health model that seeks to find situational vectors of individual and societal disturbance. We have seen that the basic motivational trilogy of needs t o understand, to belong, and to sustain self-esteem, which sometimes lead to scientific discoveries, friendships, and heroic deeds, can at times also fuel the machine of madness, when misused. Violations of expectations about matters central to one’s self-image can trigger a cascade of cognitive, emotional, and motivational processes that, taken together, may start that person down the path to madness. While acknowledging the role of genetic and biological factors in some forms of psychopathology, I believe we need to chart their limitations in our understanding of how the human mind and spirit may be radically transformed by certain personal and social experiences. This research, and the model giving rise to it, has shown that specific types of symptoms, such as paranoia, phobia, or hypochondria, can be predicted by knowing the nature of a person’s cognitive search biases that are activated when trying to make sense of personal discontinuities. It should also be evident that hypnosis is a powerful methodological tool for manipulating motivational, cognitive and affective states. As such, it deserves wider utilization by social psychologists and other researchers for its many values that go beyond the realm of the relatively small number of psychologists who primarily study hypnosis itself. In a similar vein, I think that this report also makes clear the possibility of studying dynamic aspects of human functioning in rigorously focused laboratory research. The once active area of experimental psychopathology needs to be revived, and can, with the current research program serving as an exemplar. It is possible to conduct “ethically challenging” research, like my own, by working closely and sensitively with IRBs. Without prospective studies of the first stages in the development of symptomatic thinking, feeling, and acting, how can we ever really know about the origins of psychopathology? Surely, not from the traditional historical reconstruction of what must have been predisposing and precipitating factors derived from the long delayed recall of clients and patients. Beyond the statistically significant effects reported in my research are the dramatic, qualitative transformations in behavior that I observed among intelligent, normally functioning colleges students
DISCONTINUITY THEORY
469
that I knew reasonably well (from intensive contacts in our training sessions). Suddenly, they became inarticulate, confused, hyperactive, angrily banging on the desk, in near tears, frightened, picking away at a scab, anxious, or developing an uncontrollable muscle tic. That is the stuff of madness in the “real world,” the origins of which we must understand in order to act against it more effectively. However, it is equally important to note that within a minute of lifting the amnesia suggestion, and restoring memory for the source of the discontinuity, we witnessed dramatic, sudden reversals of the delusions that our subjects had been maintaining with conviction and vigor. Typically, there was a moment of confusion, followed by smiling, laughter, or shows of amazement by the debriefed subjects about the fact that they had believed so strongly in their false interpretation of what they were experiencing. The formation of delusions, along with other symptoms of psychopathology, revealed in our laboratory model using hypnosis to create subjectively compelling anomalies in personal experience, presupposes attributional explanations at work. One test of that assumption is offered by Kihlstrom and Hoyt (1988) who propose, If the attributional explanation is correct, delusions should form. as they did in the Zimbardo experiment, when subjects are amnesic for the hypnotic suggestion that is the true source of their experience; but they should drop the delusion as soon as the amnesia suggestion is canceled, and memory restored. (p. 99)
That is precisely what we found in every one of the hundreds of “discontinuity subjects” we have studied. Let’s consider next some limitations of this research. Obviously, the generalizability of the findings I have reported are constrained by reliance on a subject population that was highly hypnotizable, able to experience amnesia, and to follow posthypnotic suggestions. Because this trait is not shared by the majority of adults, caution must be exercised in how far one is willing to go beyond the evidence presented. However, since there are no reported data relating hypnotizability to virtually any other trait or personal attribute, it is unlikely that our most important findings were due to the operation of some pathological thought process characteristic of the special subject population we have utilized. Also the many predicted differential outcomes in conditions to which high hypnotizables were assigned randomly argue against a main effect of hypnotizability. Nevertheless, we need to develop other methods of inducing discontinuities that do not rely on hypnosis so that our conceptual assertions are not limited by any of its properties. Also, it is well to explore situations where naturally
470
PHILIP G . ZIMBARDO
occurring discontinuities might be expected, so that prospective studies can be done with pre-post-assessments. Another limitation comes from the use of relatively small size samples drawn from college student populations. We can assume they are generally biased toward being “explanatory-prone’’ when faced with discontinuities, rather than not thinking about them. We might seek out other convenient populations for future controlled laboratory research, which better represent the general population. More serious limitations on my theory itself come from its culture-bound nature. It is clearly based on an individualistic cultural orientation, in which self-image is vital, and people seek explanations for discontinuities through personal cognitive searches, rather than collective ones. Only cultural extensions of this theory will inform us of what may prove to be fascinating differences across societies in how people deal with their significant violations of expectations. A further limitation to be faced are the severe constraints imposed by IRBs on conducting future research to replicate these studies, or to move the theory in new directions. Just as I anticipate getting some negative feedback from readers of this chapter not satisfied with the ethical resolutions reported in the previous section, I wonder how many other social psychologists would be willing to undergo the high costs incurred in doing such experimental research by satisfying the demands of their IRBs? There are a number of extensions and refinements of the ideas presented here that we and other researchers might consider in our future research. *
*
-
-
To assess and study the operation of naturally existing explanatory search biases To compare reactions to discontinuity among experimental subjects for whom the induced search bias is either congruent or incongruent with their own preferred search bias, starting with a standard 2 X 2 design To study the social dimension of Discontinuity Theory, its normalizing aspect, its role in joining or starting nontraditional groups, such as cults and antisocial movements To study the behavioral search process that may, for example, lead to addictive and destructive behaviors as one outcome of drinking, taking drugs, or other actions designed to reduce the distress of unexplained arousal To study cultural variations in processing discontinuities To create conditions that facilitate self-disclosure of discontinuities so as to better study the strategies and tactics used in this special form of interpersonal communication To more fully explore the social pathology side of the predictions advanced in Table 11.
DISCONTINUITY THEORY *
47 1
To study hemispheric differences in brain functioning with EEG recordings and fMRI effects of unexplained arousal.
There are some practical applications and extensions of the line of thinking advanced in this chapter that I would like to highlight briefly. Although costly, patients should receive medical checkups prior to any final, serious psychiatric diagnosis, treatment, and commitment to mental health facilities. For example, hearing tests, and providing hearing aids to the partially deaf, could prevent and treat early stages of paranoia better than psychotherapy and medications. Therapists might be encouraged to be more sensitive to establishing timelines for causal sequences related to the possible origins of mental disorders in their clientdpatients. They should also be more aware of the explanatory search biases of those they treat, as well as recognizing their own biases toward overemphasizing inner, dispositional determinants, while minimizing situational influences on mental and behavioral functioning. They also need to combat the biased conceptual frameworks their training may have imposed on their attributional thinking. At a more general level, greater efforts need to be made in education to extend the basic kinds of critical thinking skills of scientists to the general population, so as to reduce the faulty thinking often associated with trying to understand personal and natural discontinuities. The canons of the scientific method should be integrated into school curricula so that, for example, students learn why correlation is not causation, why it is wrong to seek only confirmatory evidence for theories, why it is essential to consider the fullest possible range of alternative explanations, and not to become a biased theorist prematurely. We should also develop educational modules with a variety of violation-of-expectation exercises, demonstrations, and experiments to promote training in the process of searching for explanations within the context of discovery of how our physical, social, and psychological worlds work. The dynamic role that intellectual discontinuities play in scientific discoveries is clearly illustrated in Albert Einstein’s reply to a student’s inquiry about the nature of productive thinking. She had asked many noted intellectual figures of the time, “When d o productive thinking processes arise, and what occurs in such a sharp, lively process?” That student later became the renown psychologist Erika Fromm, who recently found that lost letter, and published it in the American Psychologist (Fromm, 1998). Einstein noted that while inertial frames are equivalent in mechanics and optics, it should also have been so in electrodynamics, but such equivalence appeared to be “unachievable” within the theoretical framework of electrodynamics. Einstein goes on to frame his self-analysis of the motivating effect of perceiv-
472
PHILIP G. ZIMBARDO
ing this anomaly in terms that support the underlying assumptions of my Discontinuity Theory: The desire to discover and remove it [the defect in the traditional theory] led to state of psychic tension in me, which after seven years of fruitless search was released through the relativization of the concept of time and distance. It was similar for the general theory of relativity. . . . It was always the search for a logically simple meaning of empirically established relationships, propelled by the conviction that there existed a simple logical meaning. (Fromm, 1998, p. 1198; Einstein’s letter was written in 1932).
Finally, this theory and research has a broader realm of applicability for all of us. Because well-adjusted individuals, our research participants, could so readily move across the line between sanity and madness for a while, it should increase our compassion for the mentally ill and reduce our tolerance for stigmatizing them as deviants in our society. We need to recognize the fluidity of the boundaries between normal and abnormal, between wellness and illness. Doing so may enable us to begin to see the “righteous Us” and the “wrong-headed Them” as kindred spirits trying to make sense of the puzzles and challenges of human existence. Perhaps the following assertion by the poet, Ralph Waldo Emerson, from his Essays and Lectures (1983) may serve as a fitting thematic conclusion to this chapter, and the appropriate book-end to its opening literary quotations. The sun shines and warms and lights us, yet we have no curiosity to know why this is so but we ask the reason of all evil, of pain, and hunger, and mosquitoes, and silly people . . . [and of our personally significant discontinuities].
Acknowledgments This chapter is dedicated to Dr. Carlo Piccione, a collaborator on this research program and a dear friend, who died too soon. Since none of this research was awarded foundation grant support, its funding by Philip G. Zimbardo, Inc., is gratefully acknowledged. Recognition goes out to my primary collaborators (alphabetically noted): Susan Andersen; Lisa Butler; Stephen LaBerge; Gary Marshall; Christina Maslach, and Carlo Piccione. Superb research support also came from Ariel Lang, Loren Kabat, Hope Landrine, Carolyn Weisz, and Wes Williams, among many others. Manuscript preparation benefitted considerably from the perceptive and detailed editorial feedback I got from my three favorite critics: Lisa Butler, Rose McDerrnott, and Christina Maslach (Zimbardo).
DISCONTINUITY THEORY
473
References ABC-TV News. (1990, August 3). "20/20" segment on Bob Richardson. American Psychiatric Association. Author. (1980). Diagnostic and statistical manual of mental disorders, (3rd ed.). Washington, DC: Author Allport, G. W., & Postman, L. J. (1947). The psychology of rumor. New York: Holt Rinehart & Winston. Anderson, C. A. (1983). Abstract and concrete data in the perseverance of social theories: When weak data lead to unshakable beliefs. Joiirnal of Experimental Social Psychology, 19, 93-108. Anderson, C. A,, Lepper, M. R., & Ross, L. (1980). The perseverance of social theories: The role of explanation in the persistence of discredited information. Journal of Personality and Social Psychology, 39, 1037-1049. Antaki, C. (Ed.) (1981). The psychology of ordinary explanations of social behavior. London: Academic Press. Arkin, R. M., & Baumgardner, A. H. (1985). Self-handicapping. In. J. H. Harvey & G. Weary (Eds.), Basic issues in attribution theory and research (pp. 169-202). New York: Academic Press. Aronson. E. (1968). Dissonance theory: Progress and problems. In R. P. Abelson, E. Aronson, W. J. McGuire. T. M. Newcomb, M. J. Rosenberg, & P. H. Tannenbaum (Eds.). Theories of cognitive consistency: A sourcebook (pp. 5-27). Skokie: Rand McNally. Artiss, K. L.. & Bullard, D. M. (1966). Paranoid thinking in everyday life. Archives of General Psychiatry, 14 (January). 89-93. Bandura, A. (1997). S&f-e@cacy: The exercise of control. New York: Freeman. Bargh, J. A. (1982). Attention and automaticity in the processing of self-relevant information. Journal of Personality and Social Psychology. 43, 425-436. Baroja, J. D. (1973). The world of witches. Chicago: University of Chicago Press. Baumrind, D. (1985). Research using intentional deception. American Psychologist, 40, 165-174. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 479-529. Beck, A. (1974). Cognition, affect, and psychopathology. In H. London & R. E. Nisbett (Eds.). Thoughr and feeling: Cognitive alteration of feeling slates ( p p . 127-140). Chicago: Aldine. Bell. L. V . (1980). Trearing the mentally ill: From Colonial times to the present. New York: Praeger. Bem. D. J. (1967). Self-perception: An alternative interpretation of cognitive dissonance phenomena: Psychological Bulletin, 74, 183-200. Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 6, p p . 1-63). New York: Academic Press. Billig, M. (1987). Arguing and rhinking:A rhtoricol approach to socialpsychofogy. New York: Cambridge University Press. Blum, G. S. (1979). Hypnotic programming techniques in psychological experiments. In E. Fromm & R. Shor (Eds.), Hypnosis: Research developments and perspectives, 2nd ed. (pp. 457-481). Chicago: Aldine-Atherton. Bower, G . H. (1981). Mood and Memory. American Psychologist, 36, 129-148. Breuer, J., & Freud, S. (1955). Studies o n hysteria. In L. Strachey (Ed.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 2). London: Hogarth Press. (Original work published in 1895).
474
PHILIP G. ZIMBARDO
Brodt, S. E., & Zimbardo, P. G. (1981). Modifying shyness-related social behavior through symptom misattribution. Journal of Personality and Social Psychology, 41, 437-449. Bromberg, P. B. (Doctoral dissertation, N. Y. U., 1967). (1968). The effects of fear and two modes of anxiety reduction on social affiliation and phobic ideation. [Vol. 28 (11-B, 4753-5754)]: Dissertation Abstracts. Brown, J. D. (1998). The selj Boston: McGraw-Hill. Bruner, J. S., & Goodman, C. C. (1947). Value and need as organizing factors in perception. Journal of Abnormal and Social Psychology, 42, 33-44. Burgoon, J. K. (1993). Interpersonal expectations, expectancy violations, and emotional communication. Journal of Language and Social Psychology, 12 (I & 2), 30-48. Burton, R. (1977). The anatomy of melancholy: What it is, with all the kinds, causes, symptoms, prognostickes & several cures of it. New York: Vintage Books. (Original work published in 1621). Butler, L., Koopman, C., & Zimbardo, P. G. (1995). The psychological impact of viewing the film, “JFK”: Emotions, beliefs, and political behavioral intentions. Political Psychology, 16, 237-251. Butler, L. D. (1993). I feel, therefore I think: The role of anger in the development of paranoid symptomafology. Unpublished Unpublished doctoral dissertation, Stanford University. Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and Social Psychology, 42, 116-131. Cameron, N. (1943). The development of paranoic thinking. Psychological Review, 50,219-233. Caporeal, L. R. (1976). Ergotosm: The Satan loosed in Salem. Science, 192, 21-26. Carducci, B. J., & Zimbardo, P. G. (1995). Are you shy? Psychology Today, 28, 34ff. Carey, S. (1985). Conceptual change in childhood. Cambridge, MA: MIT Press. Chaikin, A. L., & Derlega, V. J. (1974). Variables affecting the appropriateness of selfdisclosure. Journal of Consulting and Clinical Psychology, 42, 588-593. Chapman, L., & Chapman, J. (1967). The genesis of popular but erroneous psychodiagnostic observations. Journal of Abnormal Psychology, 72, 193-204. Chapman, L., & Chapman, J. (1969). Illusory correlation as an obstacle to the use of valid psychodiagnostic signs. Journal of Abnormal Psychology, 74, 271-280. Chua-Eoan, H. (1997, April 7). Imprisoned by his own passions. Time, 40-42. Ciotti, P. (1980). Murder in the math department. New West (January 28), 39-54. Claridge, G. (1985/1995). Origins of mental illness: Temperament, deviance and disorder. Cambridge, MA: Malor Books. Cleveland, W. S. (1985). The elements of graphing data. Monterey, CA: Wadsworth. Coe, W. C., & Ryken, K. (1979). Hypnosis and risks to human subjects. American Psychologist, 34, 673-681. Cohen, S., & McKay, G. (1983). Social support, stress, and the buffering hypotheses: A theoretical analysis. In A. Baum, S. E. Taylor, & J. Singer (Eds.), Handbook ofpsychology and health (Vol. 4). Hillsdale, NJ: Erlbaum. Cooper, A. F. (1976). Deafness and psychiatric illness. British Journal of Psychiatry, 129, 2 16-226. Cooper, A. F., & Curry, A. R. (1976). The pathology of deafness in the paranoid and affective psychosis of later life. Journal of Psychosomatic Research, 20, 97-105. Cooper, A. F., Kay, D. W. K., Garside, R. F., & Roth, M. (1974). Hearing loss in the paranoid and affective psychoses of the elderly. Lancet, ii, 851-854. Cooper, J., & Fazio, R. H. (1984). A new look at dissonance theory. In L. Berkowitz (Ed.), Advances in experimental social psychology, Vol. 17 (pp. 229-266). New York: Academic Press.
DISCONTINUITY THEORY
475
Dahlstrom, W. G., Welsh. G. S., & Dahlstrom. L. F. (1975). A n MMPI handbook, Vol 11: Research applications. Minneapolis: University of Minnesota Press. Darley, J. M., & Batson, C. D. (1973). From Jerusalem to Jericho: A study of situational and dispositional variables in helping behavior. Journal of Personality and Social Psychology, 27, 100-108. Darley, J . M.. & Fazio, R. H. (1980). Expectancy confirmation processes arising in the social interaction sequence. American Psychologist, 35, 867-881. Davison, G . C., & Valins, S. (1969). Maintenance of self-attributed and drug-attributed behavior change. Journal of Personality and Social Psychology, 11, 25-33. Deckers, L., & Devine, J . (1981). Humor by violating an existing expectancy. Journal of Psychology, 208, 107-1 10. Dedman, B. (1998, August 8). Secret service report looks at patterns among assassins. San Francisco E.raminer, pp. A6. Derlega, V. J., Margulis, S. T. & Winstead, B. A. (1987). A social-psychological analysis of self-disclosure in psychotherapy. Journal of Social and Clinical Psychology, 5, 205-215. Derogatis. L. R. (1982). The Brie,f Symptom Inventory (BSI): Administration, scoring and procedures manual: 1. Baltimore, MD: Author. Derogatis, L. R. (1983). SCL-90- R administration, scoring, and procedures manual-ll for the R (revised) version. Towson, MD: Clinical Psychometric Research. Dodge, K. A,, & Frame, C. L. (1982). Social cognitive biases and deficits in aggressive boys. Child Development, 53, 620-635. Donchin, E. (1981). Surprise! . . . Surprise? Psychophysiology, 18, 493-513. Donchin, E., & Coles, M. G. H. (1988). Is the P300 component a manifestation of context updating? Behaviorai and Brain Sciences, 11, 3.57-374. Eastwood. M. R. (1975). The Relation Between Physicaland Mental Illness. Toronto: University of Toronto Press. Eaton, W. W. (1980). The sociology of mental disorders. New York: Praeger. Emerson, R. W. (1983). Essu-ysand lectures. New York: Literary classics of the United States. Endler, N. S. (1983). Generality of the interaction model of anxiety with respect to two social evaluation field studies. Canadian Journal of Behavioural Science, 15, 60-69. Esterson, A. (1998). Jeffrey Masson and Freud's seduction theory: A new fable based on old myths. History of the Human Sciences, 11, 1-21. Evans, F. J. (1979). Contextual forgetting: Post-hypnotic source amnesia. Journal ofAbnorma1 Psychology, 88, 556-563. Evans, F. J., Orne. E. C., & Orne, M. T. (1980). Attempting to breach posthypnotic amnesia. Journal of Abnormal Psychology, 89, 603-616. Fabing. H. D. (1956). On going berserk: A neurochemical inquiry. The Scientific Monthly, 83 (November), 232-237. Fadiman, J., & Kewman, D. (Eds.). (1979). Exploring madness: Experience, theory, and research. Monterey, CA: Brooks/Cole. Fein. R. A,, & Vossekuil, B. (1997). Preventing assassination: Secret Service exceptional case study project (NCJ167224). Washington, D.C.: National Institute of Justice. Fenigstein, A,, & Vanable, P. A. (1992). Paranoia and self-consciousness. Journal of Personality and Social Psychology, 62, 129-138. Festinger, L. (1954). A theory of social comparison processes. Hunzan Relations, 7, 117-140. Festinger, L. (1957). A theory ofcognitive dissonance. Stanford, CA: Stanford University Press. Fcstinger, L.. Riecken, H. W., & Schachter, S. (1956). When prophecy fails. Minneapolis, MN: University of Minnesota Press. Fleming. J.. & Arrowood, A. J. (1978). Information processing and the perseverance of discredited self-perceptions. Personality and Social Psychology Bulletin, 5, 201-205.
476
PHILIP G. ZIMBARDO
Fodor, J. (1983). The modularity of mind. Cambridge: MIT Press. Foucault, M. (1967). Madness and civilization. New York: Mentor. Frankel, C. (1973). The nature and sources of irrationalism. Science, 180 (June), 927-931. Frankl, V. (1970). ManS search for meaning: A n introduction to logotherapy. New York: Simon & Schuster. Franzini, L. R., & Grossberg, J. M. (1994). Eccentric and bizarre behaviors. New York: Wiley. Freud, S. (1911). Psychoanalytic notes upon an autobiographical account of a case of paranoia (dementia paranoides). In J. Strachey (Ed.), The complete psychological works (Vol. 12,). New York: Norton, 1976. Freud, S. (1946). The ego and the mechanisms of defense. New York: International University Press. Freyd, J. J. (1996). Betrayal trauma: The logic of forgetting childhood abuse. Cambridge, MA: Harvard University Press. Gazzaniga, M. S. (1998). The mind’s past. Berkeley, CA: University of California Press. Geer, J. H., Davison, G. C., & Gatchel, R. (1970). Reduction of stress in humans through nonveridical perceived control of aversive stimulation. Journal of Personality and Social P,Tychology, 16, 731-738. Freyhan, F. A,, Giannelli, S., O’Connell, R. A., & Mayo, J. A. (1971). Psychiatric complications following open heart surgery. Comprehensive Psychiatry, 12, 181-195. Fromm, E. (1998). Lost and found half a century later: Letters by Freud and Einstein. American Psychologist, 53, 1195-1198. Fromm, E., & Shor, R. (Eds.), (1979). Hypnosis: Research developments and perspecrives. Chicago, II: Aldine-Atherton. Glasner, S. (1966). Benign paralogical thinking. Archives of General Psychiatry, 14, 94-99. Gleick, E. (1997, April 7). With an exit sign from heaven. Time, 28-36. Goldfried, M. R., & Robins, C. (1983). Self-schema, cognitive bias, and the processing of therapeutic experiences. In P. C. Kendall (Ed.), Advances in cognitive-behavioral research and therapy (Vol. 2, pp. 33-80). New York: Academic Press. Coleman, D. (1985). Vital lies, simple truths: Psychology of self deception. New York: Simon & Schuster. Gottleib, B. H. (1981). Social networks and social support. Beverly Hills, CA: Sage. Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision of personal history. American Psychologist, 35, 603-618. Gruder, C. L., Gurwitz, S., & Reiss, S. (1977). Abnormal behavior in “normal” people. In S. Reiss & R. A. Peterson (Eds.), Abnormality: Experimental and clinical approaches (pp. 548-575). New York: Macmillan Publishing Co., Inc. Cur, R. C., & Sackheim, H. A. (1979). Self-deception: A concept in search of a phenomenon. Journal of Personality and Social Psychology, 37, 147-169. Hamilton, D. L., Dugan, P. M., & Trolier, T. K. (1985). The formation of stereotypic beliefs: Further evidence for distinctiveness-based illusory correlations. Journal of Personality and Social Psychology, 48, 5-17. Haney, C. (Unpublished doctoral dissertation, Stanford University, 1978). (1978). Consensus information and attitude change: Modifying the effects of counteratfifudinalbehavior with information about rhe behavior of others. (Vol. DDJ78-22512). Ann Arbor, MI: International Dissertation Abstracts. Haney, C., & Zimbardo, P. G . (1998). The past and future of U S . prison policy: Twenty-five years after the Stanford Prison Experiment. American Psychologist, 53,709-727. Hansen, C. (1969). Witchcraft at Salem. New York Mentor. Harrower, M. R., & Steiner, M. E. (1951). Large scale Rorschach techniques (Znd ed.). Springfield, IL: C. C. Thomas.
DISCONTINUITY THEORY
477
Hassan, S. (1988). Combatting cult mind control. Rochester: Park Street Press. Hastie, R. (1985). Causes and effects of causal attribution. Journal of Personality and Social Psychology, 46, 44-56. Hays, W. L. (1965). Statistics for psychologists. New York: Holt, Rinehart & Winston. Hechler. D. (1994, April 18). Who’s telling the truth? When memories tear families apart. First, 54-57. Heider, F. (1958). The psychology of interpersonal relationships. New York: Wiley. Henry, J. (1971). Pathways to madness. New York: Random House. Hilgard, E. R. (1965). Hypnotic susceptibility. New York: Harcourt, Brace & World. Hilgard, E. R. (1982). Hypnotic susceptibility and implications for measurement. International Journal of Clinical and Experimental Hypnosis, 30, 394-403. Hodgins. H. S., & Zuckerman, M. (1993). Beyond selecting information: Biases in spontaneous questions and resultant conclusions. Journal of Experimental Social Psychology, 29, 387-407. Holman. A., & Zimbardo, P. G . (1999). The times of your life: Time perspective and adult social relationships. Unpublished manuscript. Horowitz, L. M., Weckler, D. A,, & Doren, R. (1983). Interpersonal problems and symptoms: A cognitive approach. In P. C. Kendall (Ed.), Advances in cognitive-behavioral research and therapy (Vol. 2, pp. 81-125). New York: Academic Press. House, J. S., Landis, K. R., & Umberson, D. (1988). Social relationships and health. Science, 241, 540-545. Huggins, M., Haritos-Fatouros, M., & Zimbardo, P. G. (1999). Torture, murder, and secrecy: State sanctioned during Brazil’s dictatorship (1964-1985). In preparation. Hunt. M. (1982, Sept. 12). Research through deception. The New York Times Magazine, p. 66 ff. Jacobs, D. M. (1993). Secret life: Firsthand, documented accounts of UFO abductions. London: Fourth Estate. Jang, S. J., & Thornberry, T. P. (1998). Self-esteem, delinquent peers, and delinquency: A test of the self-enhancement thesis. American sociological review, 63, 586-598. Jaspers, K. (1962). General psychopathology. Chicago: University of Chicago Press. (Original work published in 1923). Jennings, D. L., Lepper, M. R., & Ross, L. (1981). Persistence of impressions of personal persuasiveness: Perseverance of erroneous self-assessments outside the debriefing paradigm. Personality and Social Psychology Bulletin, 7, 257-263. Johnson, J. K.. & Raye, C. L. (1981). Reality monitoring. Psychological Review, 88, 67-85. Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in social psychology. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 219-266). New York: Academic Press. Jones, J. (1997). Prejudice and racism. (2nd ed.). New York: McGraw Hill. Jussim, L. (1990a). Expectancies and social issues: Introduction. Journal of Social Issues, 46, 1-8. Jussim, L. (1990b). Social reality and social problems: The role of expectancies. Journal of Social Issues, 46, 9-34. Kanzer, M., & Glenn, J. (Eds.). (1980). Freud and hispatients. (Vol. 11). New York: Jason Aronson. Kay, D. W.. Cooper, A. F., Garside, R. F.. & Roth, M. (1976). The differentiation of paranoid from affective psychoses by patients’ premorbid characteristics. British Journal of Psychiatry, 129, 207-215. Kelley. H . H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska symposium on motivation (Vol. 15, pp. 192-246). Lincoln, NB: University of Nebraska Press.
478
PHILIP G. ZIMBARDO
Kelley, H. H. (1971). Attribution in social interaction. In E. E. Jones, D. E. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.), Attribution: Perceiving the causes of behavior (pp. 151-174). Morristown, NJ: General Learning Press. Kietzman, M. L., Sutton, S., & Zubin, J. (Eds.), (1975). Experimental approaches to psychopathology. New York: Academic Press. Kihlstrom, J. F. (1979). Hypnosis and psychopathology: Retrospect and prospect. Journal of Abnormal Psychology, 88, 459-473. Kihlstrom, J. F. (1984). Conscious, subconscious, unconscious: A cognitive view. In K. S. Bowers & D. Meichenbaum (Eds.), The unconscious: Reconsidered (pp. 149-21 1). New York: Wiley. Kihlstrom, J. F. (1985). The cognitive unconscious. Science, 237, 1445-1452. Kihlstrom, J. F., & Cantor, N. (1984). Mental representations of the self. In L. Berkowitz (Ed.), Advances in experinrentalsocialpsychology, Vol17 (pp. 1-48). New York: Academic Press. Kihlstrom, J. F., Evans, F. J., Orne, E. C., & Orne, M. T. (1980). Attempting to breach posthypnotic amnesia. Journal of Abnormal Psychology, 89, 603-616. Kihlstrom, J. F., & Hoyt, I. P. (1988). Hypnosis and the psychology of delusions. In T. F. Oltmans & B. A. Maher (Eds.), Delusional beliefs (pp. 66-109). New York: Wiley. Klauder, J. R. (1978). Continuous and discontinuous perturbations. Science, 199, 735-740. Klayman, J., & Ha, Y. (1987). Confirmation, disconfirmation, and information in hypothesis testing. Psychological Review, 94, 211-228. Kohler, W. (1958). The obsessions of normal people (A paper read at the inauguration of the graduate school of arts and sciences of Brandeis University). Waltham, Massachusetts: Brandeis University. Kolditz, T., & Arkin, R. M. (1982). An impression management interpretation of the selfhandicapping phenomenon. Journal of Personality and Social Psychology, 43, 492-502. Koran, L. M., Sox, H. C . , Marton, K. I., Moltzen, S., Sox, C. H., Kraemer. H. C., Imai, K., Kelsey, T. G., Rose, T. G., Levin, L. C., & Chandra, S. (1989). Medical evaluation of psychiatric patients. Archives of General Psychiatry, 46, 733-740. Kramer, R. M. (1998). Paranoid cognition in social systems: Thinking and acting in the shadow of doubt. Personality and social psychology review, 2, 251-275. Kramer, H., & Sprenger, J. (1971). The Malleus Maleficarium (“Witches Hammer”) (M. Summers, Trans.). New York: Dover. (Original work published in 1486). Kulik, J. A., & Mahler, H. I. M. (1989). Social support and recovery from surgery. Health Psychology, 8, 221-238. Laberge. S., & Zimbardo, P. G. (1999). Event-related potential correlates of suggested hypnotic amnesia. Unpublished manuscript. Langer. E. J. (1975). The illusion of control. Journal of Personality and Social Psychology, 32, 311-328. Latanb, B., & Darley, J. M. (1970). The unresponsive bystander. New York: AppletonCentury-Crofts. Lattin, D. (1988, October 9). The end of the world: Doomsday beliefs flourish as the year 2000 nears. San Francisco Chronicle, pp. 9-10. Lawrence, P. (1964). Road belong Cargo: A study of the cargo movement in the Southern Mandang district, New Guinea. Manchester: Manchester University Press. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer. Leary, M., & Miller, R. S. (1986). Social psychology and dysfunctional behavior: Origins, diagnosis, and treatment. London: Springer Verlag. LeDoux, J. E. (1989). Cognitive-emotional interactions in the brain. Cognition and Emotion, 3. 267-289.
DISCONTINUITY THEORY
479
LeDoux, J. E. (1995). Emotion: Clues from the brain. Annual Review of Psychology, 46, 209-235. Lefcourt. H. M., & Martin, R. A. (1986). Humor and life stress: Antidote to adversity. New York: Springer-Verlag. Lemert. E. M. (1962). Paranoia and the dynamics of exclusion. Sociometry, 25, 2-20. Levi-Straws, C. (1966). The savage mind. Chicago: University of Chicago Press. Lewis. M. (1981). The ethics of inducing paranoia in an experimental setting. IRB, 3, p. 9. Liu, J. H., Karasawa, K., & Weiner, B. (1992). Inferences about the causes of positive and negative emotions. Personality and Social Psychology Bulletin, 18, 603-615. Loftus, E., & Ketcham, K. (1994). The myth of repressed memory: False memories and allegations of sexual abuse. New York: St. Martins. Lord, C., Ross, L., & Lepper, M. R. (1979). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personality and Social Psychology, 37, 2098-2109. Luborsky, L. (1970). New directions in research on neurotic and psychosomatic symptoms. American Scientist, 58 (November-December), 661-668. Lynn, S. J., & Payne, D. G. (Eds.), (1997). Memory as the theater of the past: The psychology of false memories. Current directions in psychological science, 6 ( 3 ) ,Entire issue. Mack, J. E. (1995). Abduction: Human encounters wirh aliens. London: Simon & Schuster. Maher, B. (1974a). Delusional thinking and cognitive disorder. In H. London & R. Nisbett (Eds.), Thought and feeling: Cognitive alteration of feeling states (pp. 85-103). Chicago, IL: Aldine. Maher, B. A. (1974b). Delusional thinking and perceptual disorder. Journal of Individual Psychology, 30,98-113. Maher, B. A,, & Ross, J. S. (1984). Delusions. In H. E. A d a m & P. B. Sutker (Eds.), Comprehensive Handbook of Psychopathology. (pp. 343-409). New York: Plenum. Malle, B. F. (in press). How people explain behavior: A new theoretical framework. Personality and Social Psychology Review, 3. Marcel, A. J. (1983). Conscious and unconscious perception: An approach to the relation between phenomenal experience and perceptual processes. Cognirive Psychology, 15, 238-300. Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63-78. Markus, H., Cross, S., & Wurf, E. (1990). The role of the self-system in competence. In R. J. Sternberg & J. Lollgian (Eds.). Competence considered (pp. 205-225). New Haven: Yale University Press. Marsella, A. J. (1979). Cross-cultural studies of mental disorders. In A. J. Marsella, R. G. Sharp, & T. J. Giborowski (Eds.), Perspectives on cross-culturalpsychology (pp. 233-262). New York: Academic Press. Marshall, G.. & Zimbardo, P. G. (1979). The affective consequences of inadequately explained physiological arousal. Journal of Personality and Social Psychology, 37, 970-988. Maser, J. D., & Seligman, M. E. P. (Eds.). (1977). Psychopathology: Experimental models. San Francisco: Freeman. Maslach, C. (1979). Negative emotional biasing of unexplained arousal. Journal of Personality and Social Psychology, 37, 953-969. Maslach, C . , Marshall, G., & Zimbardo, P. G. (1972). Hypnotic control of peripheral skin temperature: A case report. Psychophysiology, 9, 600-605. Maslach, C., Zimbardo, P. G.,& Marshall, G. (1979). Hypnosis as a means of studyingcognitive and behavioral control. In E. Fromm & R. Shor (Eds.), Hypnosis: Research developments and perspectives, (2nd Ed.) (pp. 539-571). Chicago, IL: Aldine-Atherton.
480
PHILIP G. ZIMBARDO
Maslow, A. H . (1970). Motivation and personality. (Rev. ed.). New York: Harper & Row. Massad, C. M., Hubbard, M., & Newtson, D. (1979). Selective perception of events. Journal of Experimental Social Psychology, 15, 513-532. Matossian, M. K. (1989). Poisons ofthepast: Molds, epidemics, and history. New Haven, CT: Yale University Press. McNair, D. M., Lorr, M., & Doppleman, L. F. (1971). E l T S Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service. Medvedev, Z., & Medvedev, R. (1971). A question of madness: Repression by psychiatry in the Soviet Union. New York Vintage. Meyer, G. J. (1995, July). Dancing with headhunters: Scenes from downsized life. Harper’s Magazine, 37-56. Meyer, M. M., & Ekstein, R. (1970). The psychotic pursuit of reality. Journal of Contemporary Psychotherapy, 3, 3-12. Meyer, N. (1974). The seven-per-cent solution; being a reprint f r o m the reminiscences of John H, Watson M. D., as edited by Nicholas Meyer. New York Dutton. Mezzich, J . E., & Berganza, C. E. (1984). Culture & psychopathology. New York: Columbia University Press. Middleton, J. (Ed.). (1967). Magic, witchcrafl, and curing. New York: The Natural History Press. Milgram, S . (1974). Obedience to authority. New York: Harper & Row. Mill, J. S . (1966). John Stuart Mill. New York: Odyssey Press. (Original autobiography published in 1873). Miller, G. (1980). Giving psychology away in the ‘80s. Psychology Today, 13, 38ff. Miller, M. (1997, April 14). Secrets of the cult. Newsweek, 29-37. Miller, N., & Zimbardo, P. G . (1966). Motives for fear-induced affiliation: Emotional companson or interpersonal similarity. Journal of Personality, 34, 481-503. Mineka, S., & Kihlstrom, J. F., (1978). Unpredictable and uncontrollable events: A new perspective on experimental neurosis. Journal of Abnormal Psychology, 87, 256-271. Montiero, K., & Zimbardo, P. G. (1987). The path from classroom seating to hypnotizability-a dead e n d A brief communication. International Journal of Clinical and Experimental Hypnosis, 35,83-86. Natale, M., & Hantas, M. (1982). Effect of temporary mood states on selective memory about the self. Journal of Personality and Social Psychology, 42, 927-934. National Public Radio. Author. (1985, October 10). Chemicals in the workplace. Newsweek (1997, April 7). Web of death: “The next level.” 26-43. Nisbett, R. E., & Ross, L. (1980). Human inference: Strategies and shortcomings of social judgment. Englewood Cliffs: Prentice Hall. Nisbett, R. E., & Schachter, S. (1966). Cognitive manipulation of pain. Journal o f Experimental Social Psychology, 2, 227-236. Nisbett, R. E., & Wilson, T. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231-259. Nolen-Hoeksema, S., Parker, L. E., & Larson, J. (1994). Ruminative coping with depressed mood following loss. Journal of Personality and Social Psychology, 67, 92-104. Ofshe, R., & Watters, E. (1994). Making monsters: False memories, psychotherapy, and sexual hysteria. New York: Scribners. Olson, J. M., Roese, N. J., & Zanna, M. P. (1996). Expectancies. In E. T. Higgins & A. W. Kruglanski (Eds.), Social Psychology: Handbook of Basic Principles (pp. 211-238). New York Guilford. Orford, J. (1976). The social psychology of mental disorder. New York: Penguin.
DISCONTINUITY THEORY
481
Ornstein, R. (1986). Multimind: A new way of looking at human behavior. Boston: Houghton-Mifflin. Pears, D. (1982). Motivated irrationality: Freudian theory and cognitive dissonance. Cambridge, England: Cambridge University Press. Pennebaker, J. W., & Epstein. D. (1983). Implicit psychophysiology: Effects of common beliefs and idiosyncratic physiological responses on symptom reporting. Journal of Personality, 51, 468-496. Peterson, D. (Ed.). (1982). A mad people's history of madness. Pittsburgh: University of Pittsburgh Press. Piaget, J. (1954). The construction of reality in the child. New York: Basic Books. Piaget, J. (1977). The development of thought: Equilibrium of cognitive structures. New York: Viking Press. Piccione, C., Hilgard, E. R., & Zimbardo, P. G. (1989). On the degree of stability of measured hypnotizability over a 25-year period. Journal of Personality and Social Psychology, 56, 289-295. Pickering, G. (1974). Creative malady. New York: Delta. Pilusuk, M., & Parks, S. H . (1986). The healing web: Social networks and human survival. Hanover, NH: University Press of New England. Pirandello, L. (1923). Three plays. New York: Dutton. PIOUS,S., & Zimbardo, P. G. (1986). Attributional biases among clinicians: A comparison of psychoanalysts and behavior therapists. Journal of Consulting and Clinical Psychology, 54, 568-570. Plutchik, R. (1980). Emotion: A psychoevolutionary synthesis. New York: Harper & Row. Plutchik, R., & Conte, H. R. (Eds.). (1997). Circurnplex models of personality and emotions. American Psychological Association: Washington, DC. Politzer. L. M. (1862). Daniel Gottlieb Moritz Schreber. Jahrbrcch fur Kinderheilkunde und Physische Erziehung, 5(2). 1-7. Polivy, J. (1981). On the induction of emotion in the laboratory: Discrete moods or multiple affect states? Journal of Personality and Social Psychology, 41, 803-817. Pope, K. S., & Brown, L. S. (1996). Recovered memories of abuse: Assessment, therapy, forensics. Washington, DC: American Psychological Association Press. Porter, R. (1989). A social hisrory of madness: The world through the eyes of the insane. New York: Dutton. Post, F. (1966). Persistent persecutory states of the elderly. London: Pergamon. Pratto, F. (1999). The puzzle of continuing group inequality: Piecing together psychological, social, and cultural forces in social dominance theory. In M. Zanna (Ed.), Advances in experimental social psychology, (Vol. 31, pp. 191-263). San Diego, CA: Academic Press. Prentice, D. A,, & Miller, D. T. (1996). Pluralistic ignorance and the perpetuation of social norms by unwitting actors. In M. Zanna (Ed.), Advances in Experimental Social Psychology, (Vol. 28, pp. 161-209). San Diego, CA: Academic Press. Proudfoot, W., & Shaver, P. (1975). Attribution theory and religion. Journal for the Scientific Study of Religion, 14, 371-390. Pyszczynski, T. A., & Greenberg, J. (1981). Role of disconfirmed expectancies in the instigation of attributional processing. Journal of Personality and Social Psychology, 40, 31-38. Quattrone, G. A,, & Tversky, A. (1984). Causal versus diagnostic contingencies: On selfdeception and on the voter's illusion. Journal of Personality and Social Psychology, 46, 237-248. Rachman, S. (1981). The primacy of affect: Some theoretical implications. Behavior Research and Therapy, 19, 279-290.
482
PHILIP G. ZIMBARDO
Rachman, S., & Hodgson, R. (1980). Obsessions and compulsions. Englewood Cliffs, N.J.: Prentice-Hall. Reed, G. (1974). The psychology of anomalous experience: A cognitive approach. Boston: Houghton Mifflin. Reiman, E. M., Lane, R. D., Ahern, G. L., Schwartz, G. E., Davidson, R. J., Friston, K. J., Yun, L. S., & Chen, K. (1997). Neuroanatomical correlates of externally and internally generated human emotion. American Journal of Psychiatry, 154, 918-925. Reuters News Service. Author. (1998, December 12). Student confesses in professor’s death, police say. S.F. Chronicle, P. A2. Reynolds, M. (1998, November 22). Economic ills give rise to anti-Semitism in Russia. Los Angeles Times, pp. A 20, 22. Roberts, L. W., Hollifield, M., & McCarty, T. (1998). Psychiatric evaluation of a “monk” requesting castration: A patient’s fable, with morals. American Journal of Psychiatry, 155, 415-420. Rosen, G . (1968). Madness in society: Chapters in historical sociology of mental illness. London: Routledge and Kegan Paul Limited. Rosenbaum, C. P. (1970). The meaning of madness. New York: Science House. Rosenbaum, M., & Muroff, M. (Eds.). (1984). Fourteen contemporary reinterpretations. New York: Free Press. Rosenberg, M. J. (1960). Cognitive reorganization in response to the hypnotic reversal of attitudinal affect. Journal of Personality, 28, 39-63. Rosenhan, D. L. (1973). On being sane in insane places. Science, 179, 250-258. Rosnow, R. L., & Fine, G. A. (1976). Rumor and Gossip. New York: Elsevier. Ross, L. (1977). The intuitive scientist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 10, pp. 174-221). New York: Academic Press. Ross, L. (1978). Some afterthoughts on the intuitive psychologist. In L. Berkowitz (Ed.), Cognitive theories in psychology (pp. 385-400). New York: Academic Press. Ross, L., & Lepper, M. R. (1980). The perseverance of beliefs: Empirical and normative considerations. In R. A. Shweder & D. Fiske (Eds.), New directions for methodology of behavioral science: Fallible judgment in behavioral research (pp. 17-36). San Francisco: Jossey-Bass. Ross, L., Rodin, J., & Zimbardo, P. G. (1969). Attribution therapy: The reduction of fear through induced cognitive-emotional misattribution. Journal of Personality and Social P.yychology, 12, 279-288. Ross, M., & Sicoly, P. (1979). Egocentric biases in availability and attribution. Journal of Personality and Social Psychology, 37, 322-336. Rothbart, M. K. (1976). Incongruity, problem solving, and laughter. In A. J. Chapman & H. C. Foot (Eds.), Humour and laughter: Theory, research, and applications (pp. 37-54). London: Wiley. Sarbin, T. R, (1981). On Self-deception. Annals of the New York Academy of Scietices, 364, 220-235. Sarnoff, I. (1971). Testing Freudian concepts. New York: Springer. Sarnoff, I., & Zimbardo, P. G. (1961). Anxiety, fear, and social affiliation. Journal ofAbnormal and Social Psychology, 62, 356-363. Savin, H. B. (1973). Professors and psychological researchers: Conflicting values in conflicting roles. Cognition, 2, 147-149. Schachter, S. (1959). The psychology of aflliation. Stanford, CA: Stanford University Press. Schachter. S., & Singer, J. (1962). Cognitive, social and physiological determinants of emotional state. Psychological Review, 69, 379-399.
DISCONTINUITY THEORY
483
Schachter, S.,& Wheeler, A. L. (1962). Epinephrine, chlorpromazine, and amusement. Journal of Abnormal and Social Psychology. 63, 121-128. Schatzman, M. (1973). Soul murder: Persecution in the family. New York: Random House. Scheff, T. J. (Ed.). (1975). Labeling madness. Englewood Cliffs: Prentice Hall. Schlenker, B. R. (1980). Impression Management: The Self-Concept, Social Identity, and Interpersonal Relations. Belmont, CA: Wadsworth. Schooler. J. W., Bendiksen, M., & Ambadar, Z. (1998). Taking the middle line: Can we accommodate both fabricated and recovered memories of sexual abuse? In M. Conway (Ed.), False and recovered memories (pp. 251-292). Oxford, England: Oxford University Press. Schreber, D. P. (1955). Memoirs o f m y nervous illness. London: W. Dawson. (Original work published in 1903). Schwarz, N. (1990). Feelings as information: Informational and motivational functions of affective states. In E. T. Higgins & R. M. Sorrentin (Eds.), Handbook of Motivation and Cognition: Foundations of Social Behavior (Vol. 2, pp. 527-561). New York: Guilford. Sebald, H. (1978). Witchcraft The herilage o f a heresy. New York: Elsevier. Seligman. M. E. P. (1975). Helplessness: O n depression, development, and death. San Francisco: W. H. Freeman. Shafir, E., Simonson, I., & Tversky. A. (1993). Reasoned-based choice. Cognition, 49, 11-36. Shapiro, D. (1965). Neurotic styles. New York: Basic Books. Shapiro, D. (1981). Autonomy and rigid character. New York: Basic Books. Sholle, S. R. (1992). A controlled study of sensation awareness and verbal disclosure for regulation of arousal and anxiety. Perceptual and Motor Skills, 74, 307-320. Shor, R. E., & Orne, E. C. (1962). Harvard group scale of hypnotic susceptibility, Form A . Palo Alto. CA: Consulting Psychologists Press. Shotter, J. (1984). Social accountability ond selfhood. Oxford, England: Blackwell. Singer, B., & Benassi, V. A. (1981). Occult beliefs. American Scientist, 69, 49-55. Skov, R. B., & Sherman, S. J. (1986). Information-gathering processes: Diagnosticity, hypothesis confirmatory strategies, and perceived hypothesis confirmation. Journal of Experimental Social Psychology, 22, 93-121. Small, G. W., & Nicholi, A. M. (1982). Mass hysteria among schoolchildren: Early loss as a predisposing factor. Archives of General Psychiatry, 39, 721-724. Snell, W. E., Miller, R. S., & Belk, S. S. (1988). Development of the emotional self-disclosure scale. Sex Roles, 18, 467-486. Snyder. M., Campbell, B., & Preston. E. (1982). Testing hypotheses about human nature: Assessing the accuracy of social stereotypes. Social Cognition, I , 256-272. Silva. C. E., & Kirsch, I. (1987). Breaching hypnotic amnesia by manipulating expectancy. Journal of Abnormal Psychology, 96, 325-329. Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimental socialpsychology (Vol. 21, pp. 261-302). New York: Academic Press. Stephen, W. G., Wenzel, B., & Cornelius, J. (1991). Intergroup interaction and self-disclosure Journal of Applied Social Psychology, 21, 1370-1378. Stiles, W. B., Shuster, P. L., & Harrigan, J. A. (1992). Disclosure and anxiety: A test of' the fever model. Journal of Personality and Social Psychology, 63, 780-788. Strelan, J. G. (1977). Search for salvation: Studies in the history and theology of Cargo cults. Adelaide: Lutheran Publishing House. Suls, J. M. (1983). Cognitive processes in humor appreciation. In P. E. McGhee & J. H. Goldstein (Eds.), Handbook of humor research (pp. 39-58). New York: Springer-Verlag.
484
PHILIP G. ZIMBARDO
Suls, J. M. (1972). A two-stage model for the appreciation of jokes: An information-processing analysis. In J. H. Goldstein & P. E. McGhee (Eds.), Thepsychologyof humor: Theoretical perspectives and empirical issues (pp. 43-60). New York: Academic Press. Szasz, T. (Ed.). (1973). The age of madness. New York: Doubleday. Taylor, M. (1997). Cult survivor thinks there’s a reason. S. F. Chronicle, pp. A 10. Taylor, S. E. (1986). Health psychology. New York Random House. Taylor, S. E. (1991). Asymmetrical effects of positive and negative events: The mobilizationminimization hypothesis. Psychological Bulletin, 110, 67-85. Taylor, S. E., Kahneman, D., Slovic, P., & Tversky, A. (1982). The availability bias in social perception and interaction, Judgment under uncertainty: Heuristics and biases (pp. 190200). Cambridge, England: Cambridge University Press. Tetlock, P. E., & Manstead, A. S. R. (1985). Impression management versus intrapsychic explanations in social psychology: A useful dichotomy? Psychological Review, 92, 59-77. Thornton, E. M. (1984). The Freudian fallacy: An alternative view of Freudian theory. New York Dial Press. Tice, D. M., Butler, J. L., Muraven, M. B., & Stillwell, A. M. (1995). When modesty prevails: Differential favorability of self-presentation to friends and strangers. Journal of Personality and Social Psychology, 69, 1120-1138. Triandis, H. C. (1994). Culture and social behavior. Boston: McGraw-Hill. Trope, Y.,& Bassok, M. (1982). Confirmatory and diagnosing strategies in social information gathering. Journal of Personality and Social Psychology, 43, 22-34. Trope, Y., & Bassok, M. (1983). Information-gathering strategies in hypothesis-testing. Journal of Experimen fa1 Social Psychology, 19, 560-576. Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.), Organization of memory. (pp. 381-403). New York Academic Press. Tulving, E. (1983). Elements of episodic memory. Oxford, England: Clarendon Press. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5, 207-232. VanGennep, A. (1960). The rites of passage. Chicago: University of Chicago Press. Wegner, D. M. (1989). White bears and other unwanted thoughts: Suppression, obsession, and the psychology of mental control. New York: Viking. Weinberger, D. A,, & Schwartz, G. E. (1990). Distress and restraint as superordinate dimensions of adjustment: A typological perspective. Journal of Personality, 58, 381-417. Weiner, B. (1985). An attributional theory of motivation and emotion. Psychological Review, 92, 548-573. Weitzenhofer, A. M., & Hilgard, E. R. (1962). Stanford hypnotic susceptibility scale, Form C. Palo Alto, CA: Consulting Psychologists Press. Wellman, H. M., & Gelman, S. A. (1992). Cognitive development: Foundational theories of core domains. Annual Review of Psychology, 43, 337-375. Welsh, G. S. (1948). An extension of Hathaway’s MMPI profile coding system. Journal of Consulting Psychology, 12, 343-344. Wenegrat, B. (1995). Illness and power: Women’s mental disorders and the battle between the sexes. New York: NYU Press. Wetzel, C. G., & Walton, M. D. (1985). Developing biased social judgments: The falseconsensus effect. Journal of Personality and Social Psychology, 49, 1352-1359. Wicklund, R. A., & Brehm, J. W. (1976). Perspectives on cognitive dissonance. Hillsdale, NJ: Erlbaum Associates. Wiener, D. N. (1947). Differences between the individual and group forms of the MMPI. Journal of Consulting Psychology, 11, 104-106.
DISCONTINUITY THEORY
485
Wiener, D. N., & Harmon, L. R. (1946). Subtle and obvious keys for the MMPI: Their development. Advisement Bulletin No. 16, Regional Veterans Administration Office, Minneapolis, MN. Wiggins, J. S. (1966). Substantive dimensions of self-report in the MMPI item pool. Psychological Monographs, 80 (whole no. 630). 1-22. Wiggins, J. S. (1979). A psychological taxonomy of trait-descriptive terms: The interpersonal domain. Journal of Personality and Social Psychology, 33, 409-420. Wiggins, J. S., & Broughton. R. (1984). The interpersonal circle: A structural model for the integration of personality research. In R. Hogan & W. H. Jones (Eds.), Perspectives in personality: Theory, measurement and interpersonal dynamics. (Vol 1, pp. 1-47). Greenwich, CT: JAL Press. Wilson, M. (1951/1970). Good company: A study of Nyakyusa Age-Villages. Oxford, England, Boston: Oxford University Press, Beacon. Wilson, M. (1962). Communal rituals of the Nyakyusa. Oxford, England: Oxford University Press. Wilson, M. (1970). Rituals of kinship among the Nyakyusa. London, England: Oxford University Press. Wilson, T. D., Lisle, D. J., Kraft, D., & Wetzel, C. G. (1989). Preferences as expectationdriven inferences: Effects of affective expectations on affective experience. Journal of Personality and Social Psychology, 56, 5 19-530. Wing, J. K. (1978). Reasoning about madness. London: Oxford University Press. Wong, P. T. P., & Weiner, B. (1981). When people ask “why” questions, and the heuristics of attributional search. Journal of Personality and Social Psychology, 40, 650-663. Wood, J. V. (1989). Theory and research of social comparisons of personal attributes. Psychological Bulletin, 106, 231-248. Worsley, P. (1968). The trumpet shallsound: A study of “cargo” cults in Melanesia, (2nd ed.). New York: Schocken Books. Young, K. (1927). Mental Epidemics, Source book for social psychology (pp. 693-721). New York: Alfred A. Knopf. Zajonc, R. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35, 151-175. Zimbardo, P. G . (1969). The cognitive control of motivation. Glenview, IL: Scott, Foreman. Zimbardo, P. G . (1970). The human choice: Individuation, reason, and order versus deindividuation. impulse, and chaos. In W. J. Arnold & D. Levine (Eds.), 1969Nebraska symposium on motivation (pp. 237-307). Lincoln, NB: University of Nebraska Press. Zimbardo, P. G. (1972). Pathology of imprisonment. Society, 6, 4-8. Zimbardo, P. G . (1973). On the ethics of intervention in human psychological research: With special reference to the Stanford Prison Experiment. Cognition, 2, 243-256. Zimbardo, P. G. (1975). On transforming experimental research into advocacy for social change. In M. Deutsch & H. Hornstein (Eds.), Applying social psychology: Implications for research, practice, and training (pp. 33-66). Hillsdale: Lawrence Erlbaum. Zimbardo, P. G. (1976). A social-psychological analysis of vandalism: Making sense of senseless violence. In E. P. Hollander & R. G . Hunt (Eds.), Current perspectives in socialpsychology (4th ed., pp. 129-134). New York: Oxford University Press. Zimbardo, P. G. (1978). Psychology of evil: On the perversion of human potential. In L. Krames, P. Pliner, & T. Alloway (Eds.), Advances in study of communication and affect (Vol. 4, pp. 115-169). New York: Plenum Press. Zimbardo, P. G. (1981). The ethics of inducing paranoia in an experimental setting: A reply to M. Lewis, M.D. IRE: A Review of Human Subjects Research, 3, 1, 9, 10-11.
486
PHILIP G. ZIMBARDO
Zimbardo, P. G. (1982). Shyness and the stresses of the human connection. In L. Goldberger & S. Breznitz (Eds.), Handbook of Stress: Theoretical and Clinical Aspects (pp. 466-481). New York: Free Press. Zimbardo, P. G. (1984a). Mind control in 1984. Psychology Today (January), 68-72. Zimbardo, P. G. (1984b). Mind control: Political fiction and psychological reality. In P. Stansky (Ed.), On nineteen eighty-four (pp. 197-215). New York: Freeman Press. Zimbardo, P. G. (1997, May). What messages are behind today’s cults? A P A Monitor, p. 14. Zimbardo, P. G., & Andersen, S. M. (1999). Alternative paths to induced paranoia among normal people. Unpublished manuscript. Zimbardo, P. G., Andersen, S. M., & Kabat, L. G. (1981). Induced hearing deficit generates experimental paranoia. Science, 212, 1529-1531. Zimbardo, P. G., & Boyd, S. N. (1999). Putting time inperspective:A new individual differences metric. Unpublished manuscript. Zimbardo, P. G., & Formica, R. (1963). Emotional comparison and self-esteem as determinants of affiliation. Journal of Personality, 31, 141-162. Zimbardo, P. G., LaBerge, S., & Butler, L. (1993). Psychophysiological consequences of unexplained arousal: A posthypnotic suggestion paradigm. Journal of Abnormal Psychology, 102, 466-473. Zimbardo, P. G., Haney, C., Banks, W. C., & Saffe, D. (1973). The mind is a formidable jailer: A Pirandellian prison. The New York Times Magazine, 6, 38-60. Zirnbardo, P. G., Marshall, G., & Maslach, C. (1971). Liberating behavior from time-bound control: Expanding the present through hypnosis. Journal of Applied Social Psychology, I, 305-323. Zimbardo, P. G., Marshall, G., White, G., & Maslach, C. (1973). Objective assessment of hypnotically induced time distortion. Science, 181, 282-284. Zimbardo, P. G., Maslach, C., & Haney, C. (in press). Reflections on the Stanford Prison Experiment: Genesis, transformations, consequences. In T. Blass (Ed.), Obedience to authority: Currentperspectives on the Milgram paradigm. Mahwah, NS: Erlbaum Publisher. Zimbardo, P. G., Maslach, C., & Marshall, G. (1972). Hypnosis and the psychology of cognitive and behavioral control. In E. Fromm & R. Shor (Eds.), Hypnosis: Research developments and perspectives (pp. 539-571). Chicago, I L Aldine-Atherton. Zimbardo, P. G., & Piccione, C. (1999). Pathologicalconsequences of searching for explanations for personal discontinuites by normal individuals. Unpublished manuscript. Zimbardo, P. G., & Williams, W. (1999). Failures of interpersonal disclosure when experiencing unexplained arousal. Unpublished manuscript. Zuckerman, M., & Lublin, B. (1965). Manual for the Multiple Affect Adjective Check List. San Diego: Educational and Industrial Testing Service.
A
role, 8-9 self-evaluation general issues, 61-62 life satisfaction, 62, 64 likelihood estimates, 66 self-worth, 64-65 trait judgments, 65-66 judgments bias ability, 34-36 corrections, 33-34 motivation. 34-36 heuristic basis, 36-37 influences empirical studies, 55-58 theory, 54-55 relevancy, 31-33 use, 29-31 knowledge declarative, 4-5 ambiguous information, 12-22 episodic information, 22-25 overview, 15-18 selective encoding, 19-21 semantics, 18-19 procedural, 5, 22-25 moods, 5-7 motivation consequence, 39-40 theory, 38-39 negative description, 2 expectations, 352-353 positive, 2 response evaluation accuracy, 52 communication, 50-52 creativity, 45-46
Ability, 34-36 Abstractions language in broadcasting, 90--91 components, 80 confirmability, 107-1 12 development, 105-106 patterns, 100-101 Abuse, 394 Accessibility, chronic, 156 Accuracy, feed-back model, 52 Additive models, 163 Aesthetic values, 352 Affect attitudes, 7 cognitive capacity empirical ambiguities, 40-41 theory, 38 cognitive consequences, 14-15 creativity. 1 definition, 3-4 determinants, 9, 11 effectiveness, 2 elicitation cognitively mediated, 13-14 individual differences, 14 nonconscious, 11-13 processes, 16-17 eliciting, 3 emotions, 5-7 evaluations, 7 facial, 140-142 information processing communications attitude-focused, 59-60 belief-focused, 59 persuasion, 58-59 487
488 Affect (continued) impression formation, 46-49 nonlaboratory situations, 52-53 overview, 41-45 persuasion, 50-52 stereotyping, 49-50 stimulus, 3-4 Affective reactions attitudes, 7 creativity, 1 definition, 3-4 determinants, 9, 11 effectiveness, 2 emotions, 5-7 evaluations, 7 knowledge declarative, 4-5 procedural, 5 moods, 5-7 stimulus, 3-4 Affiliations, see Belonging Aggression, mindless, 354 AIDS suffers, 290-291 Ambiguities, empirical, 40-41 Amnesia hypnotic, 411, 414 unexplained arousal, 415-425 Anti-Drug Abuse Act, 203-204 Anti-Semitism, see also Jews cultural frame, 216-217 discontinuity association, 360 group status, 230-231 language abstraction, 108-1 09 in mass media, 88-91 motivation, 192 Anxiety, coping, 354-355 Applicability definition, 127 transient sources, 156-158 APT, see Authoritarian personality theory Arousal experience, 356 measures, 446-448 ART, see Attitude representation theory Assumptions perceptions and, 323-326 types, 320-323 Asthma, nighttime, 380 Attachment styles, 178 theory, 171-174
INDEX Attitude representation theory assumptions perceptions and, 323-326 types, 320-323 attitude-relevant situations, 318-320 change resistance, 281-284 contact hypothesis vs. overview, 288-290 representations, 291-294 specific representations, 290-291 experience effects, 326-329 matching targets characteristics, 301-304 context, 308-311 description, 269-270 identities, 299-301 irrelevant, 311-313 relevant research, 316-318 responses, 304-308 predictions, 277-278 processes, 270-274 representation postulate, 269 research summary attitude-behavior consistency, 294-295 consistent exemplar activation, 296-297 overview, 294-295 response generality, 296-297 response consistency, 281-284 Attitudes ART, see Attitude representation theory -behavior association, 266-267 consistency, 294-295 characterization, 7 concept, 265-266 concept maps, 311-313 consistent activation, 281-287 created, 303 expert, 325-326 expression, 327 focused communication, 58-59 inconsistency, 267-268 manipulating, 303-304 novice, 325-326 predictive powers, 309-310 -relevant situations, 318-320 response stability, 281-284 Attributes, negative personal, 388
INDEX Attributions discontinuity, 373-374 case studies, 395-399 errors, 373-374 madness, 381 measures, 45 1-452 social context, 196-197 Authoritarianism, research, 192-193 Authoritarian personality theory, 193
B Behavior -attitudes association, 266-267 consistency, 294-295 individual, LIB model, 95-96 negative description. 95 nonlaboratory, 52-53 stereotype-induced, 198 Beliefs conspiracy, 358 cult, 359 focused communication, 59 personal, language use. 104 Belonging needs, 349-350 process, 367-368 Betti, Ugo, 383 Bias, see also Prejudice body, 440-441 explanatory, 385-389 external, 387 focus, 386 judgments ability, 34-36 corrections, 33-34 motivation, 34-36 LEB model application, 102 description, 96 developmental studies, 106 mechanism, 101 message recipient, 107 LIB model abstractions, 107-112 abstract language use, 102-105 concrete language use, 102-105
489
confirmability, 107-1 12 controls, 112-1 16 description, 79-82 developmental issues, 105-106 language, 100-101 mass media, 88-91 measurement, 85-87 message recipient, 106-107 paradigm, 82-83 replication, 85 sports reports, 87-88 underlying mechanisms determining, 91-92 differential expectancy, 92-95 individual behavior, 95-96 language bias, 100-101 motivational processes, 97-100 self-protection, 100-101 people, 441-444 Bipolar disorder, 402-404 Bonding demands, 349-350 Brains adaptability, 348 aggression and, 354 impairments, 382 Brown v. Board of Education, 202 Bush, George, 220-221
C Cage, John, 350 Cargo cults, 405 Categorization affect impact, 18-19 attitudes, 277-278 automatic, 12 coincidence, 376 default assumptions, 18-19 questions, 279-280 racial, 287-288 stereotyping, 164 transference, 163 Chance, see Coincidences Characterization, see Stereotypes Children abstract concepts, 105-106 transference process, 130 Chinese acceptance, 308-309 action against, 305
490
INDEX
Chinese (continued) characterization generalizations, 301-304 individualizing, 318-319 negative, 311 default values, 270, 272-273 stereotypes, 266-267 Churchill, Winston, 268, 275 Clinton, William Jefferson accessibility, 288-289 action study response, 307 political corruption, 277-279 as political exemplar, 281-282, 285, 297, 299 voting for, reasons, 305 Cognition affect capacity empirical ambiguities, 40-41 theory, 38 comparison, 3-8 encoding processes, 14-15 mediated, 13-14 memory, 14-15 determinants, 1 emotion-associated, 6 internal, 368-369 search process, 365-367 stereotyping and, 314-315 transference model, see Social-cognitive model Coincidences, 376, 400-402 Collectivity, shared, 362-364 Communications attitude-focused, 58-59 belief-focused, 59 feed-back model, 50-52 generalizability, 58-59 information processing, 60-61 mass, see Mass media Confucianism, 217 Conservatism Black, 221 correlates, 208-209 Consistency activation attitude change, 284-287 relevant research, 296-297 response stability and, 281-284 attitudes responses, 281-284 stereotypes, 267-268
Conspiracy theories case histories, 400-402 origins, 358 Contact hypothesis overview, 288-290 representations generalization, 291-294 specific, 288-290 Context ART targets, 308-311 discovery, 377 discrimination, 202-204 group, SDO, 233-235 ideology mediation, 241-244 justification, 377 psychological process attribution role, 196-197 linkage, 199-200 SDO expression, 247 Corruption, politics, 277-278 Creativity affect, 1 feed-back model, 45-46 Credit-worthiness, 244 Cued sentence completion alternatives, 85-87 description, 83 Cues proprioceptive, 16 stereotype, 197 transference, 161-162 Cults associated activities, 388 beliefs, 359 cargo, 363, 405 Culture based-studies, transference, 176 SDO, 213, 216-217, 246 D
DAV, see Descriptive action verbs Deafness, conductive, 363 Deception, research issue, 460-462 Declarative knowledge affective reactions ambiguous information, 21-22 episodic information, 22-25 overview, 4-5, 15-18
INDEX selective encoding. 19-21 semantics, 18-19 attitudes, 7 Default values. 270, 272-273 Defensiveness, 367 Delinquency, 370 Depression affect association, 13 discontinuity and, 387 Descriptive action verbs application, 86-87 description. 80 Development, abstract concepts, 105-106 Deviance, 352 Diderot, Denis, 358 Differential expectancy, 92-95 Disasters, natural, 352, 364 Disclosures, interpersonal, 452-458 Discontinuities benefits, 353 biased rationality, 371-378 categories, 364 culturally constructed, 359-360 cumulative, 361-362 definition, 351 definition, original, 408 episodic, 365 explanations external, 369 false, 374-376 functions, 355-356 internal, 368-369 ideological constructs, 363-364 locating, 356-359 memory and, 364-365 paranoid, 379, 383-385 perception, 353-354 personal, 378-379 procedural, 365 reason-based application. 376-377 short-circuiting, 382 searches associated pathologies, 385-389 case studies, 402-405 mass reactions, 405-406 types, 382-385 semantic, 364-365 source attributions alternative hypotheses, 399-400
inaccurate, 373-374.395-399 misidentification, 372, 389-395 randomness response, 400-402 triggers creation, 428-429 experimental, 424 unexplainable, 379-380 variance, 360-365 Discontinuity theory cases studies mass reactions, 405-406 searches, 402-405 source recognition alternative hypotheses, 399-400 misattribution, 395-399 misidentification, 389-395 randomness response, 400-402 experimental paradigm demand characteristics results, 446-448 Study, 444-446 features pilot-testing, 41 1-414 procedural, 410-41 1 findings, overview, 414-415 interpersonal disclosures results, 456-458 study, 452-456 overview, 407-408 paranoia pathways results, 432-444 Study, 425-426,432 participants, 408-409 search frames results, 437-444 study, 434-437 training, hypnosis, 409-41 0 unexplained arousal results, 418-425 Study, 415-417 knowledge needs, 349-350 motivation, 350-351 normalization processes, 365-371 overview. 345-347 paradox, 346 purpose, 347-348 repression model vs., 354-355 self-esteem needs, 349-350 social isolation, 349-350 Discovery, context, 377
491
492
INDEX
Discrimination celebrities, 287-288 motivators, 231-232 societal patterns, 202-204 stereotype, association, 195 Disharmony, 359 Distortions, parataxic, 130 Dole, Robert, 285 Dominance, see Group dominance
Explanations bias, predicting, 385-389 discontinuity functions, 355-356 search, 374-376 mystical, 412 searches, 383 validation. 378-379
F E Education, discrimination, 202 EEOC, see Equal Employment Opportunity Commission Ego, totalitarian, 367 Emotions, see also Moods affect, 5-7 characterization, 6 measures, 448-450 negative, development, 352 onset, 6-7 Employment discrimination, 203 Encoding, selective, 19-21 Episodic discontinuity, 365 Equal Employment Opportunity Commission, 220 Ethics, research constraints, 462-465 cost/benefits, 465-467 deception, 460-462 general observations, 468-470 overview, 458-460 Evaluations characterization, 7 representation-derived, 129 Evolutionary theory, 210-21 1 Expectations differential, 92-95 negative affective, 352-353 transference, 142-143 violations daily life, 361 effects, 351 types, 351-352 Experiences arousal, 356 attitudes and, 325-329 first-time, 361 social affects, 36-37
Feed-back model accuracy, 52 communication, 50-52 creativity, 45-46 impression formation, 46-49 life situations, 64 nonlaboratory situations, 52-53 overview, 41-45 persuasion, 50-52 stereotyping, 49-50 Free verbalizations, 86 Freud, Sigmund, 354 Fry, Christopher, 380 G
Gender differences, mate preference, 213 Generalization attitudes, 265 relevant research, 296-297 representativeness vs., 291-294 Gingrich, Newt, 278, 283, 285 Global cognitive appraisal, 16-17 Goals attainment, 42 processing-directed, 9 Good fortune, 351-352 Group dominance approval, 229-230 attention, 252 characterization, 197 creation, causes, 245 cultural variation, 246 hierarchies, 194-195 intergroup relations, 245-246 maintenance, 253-254 psychological orientations, 228-229 realization, 228-229 SDO relation, 209-210,233-235
INDEX social-cognitive theories, 196- 198 social contexts, 228-229 social ideologies, 195-202 social practices, 245-246 universality, 191
H Handbook of Social Psychology, 265 Heaven’s Gate, 404-405 Hierarchies enhancing essays experiment 1, 222-224 experiment 2, 224-226 experiment 3,227-228 enhancing ideologies, 217-218 group-based dominance APT, 194-195 universality, 191 occupations and, 238-241 roles, SDO, 237-241 Hill, Anita, 220-221 Horror, 352 Humor, 352 Hypochondria case studies, 395-396 onset, predicting, 386
I IAV, see Interpretive action verbs Identification discontinuity sources case studies, 389-395 process, 372-372 matching targets, 299-301 social, SDO, 246-247 Ideologies correlations, 208-209 definition, 199-200 discontinuity-generated, 363-364 features, 201 influences, 201-202 legitimization, 202 merit rewarding, 242-243 policy justification, 248 right-wing, 252-253 SDO usage, 216-217
social justification, 217-218, 220-225 social mediation, 241-244 IF-THEN relations, 165-168 Immigration laws, 305 SDO, 222-224 Impression formation, 46-49 Individualism, effects, 253 Individuals affect elicitation, 14 behavior, LIB model, 95-96 characteristics, 313-314 prejudice, 192-195 rights scale, 209 Inferences, representation-derived emergence, 129 evaluation, 135-139, 140 expectancies, 142-143 facial affect, 143-146 motivation, 140-142 self role, 151 triggers, 156-160 Information ambiguous, 12, 21 episodic, 22-25 feed-back model accuracy, 52 communication, 50-52 creativity, 45-46 impression formation, 46-49 nonlaboratory situations, 52-53 overview, 41-45 persuasion, 50-52 stereotyping, 49-50 processing affect-based communications, 58-59 overview, 8-9 communication affect role, 60-61 attitude-focused, 59-60 belief-focused, 59 persuasion, 58-59 self-evaluation general issues, 61-62 life satisfaction, 62, 64 likelihood estimates, 66 self-worth, 64-65 trait judgments, 65-66 processing model, 384-385 seeking strategies, 198
493
494
INDEX
In-group protection basis, 91-92 evidence, 92-95 individual role, 95-96 relevancy, 97-100 variation, 101-102 Institutional discrimination, 202 Intelligence, stereotyping, 96 Interpersonal disclosures, 452-458 Interpersonal measures, 450-451 Interpretive action verbs application, 86-87 description, 80 Isolation pathology, 349-350 Italians language abstraction, 108-109 stereotypes, 93-95 Ivins, Molly, 229-230
J
Jealousy, sexual, 212-213 Jews, see also Anti-Semitism anti-Semitism, reporting on, 88-91 AshkenadMizrachi distinction, 231-232 prejudice against, 192 status, 216 stereotyping, 108-109 Jordan, Michael, 287 Judgments accuracy, 52 affect-based bias ability, 34-36 corrections, 33-34 motivation, 34-36 heuristic basis, 36-37 influences empirical studies, 55-58 theory, 54-55 relevancy, 31-33 use, 29-31 attitude-relevant, 280-281 impression formation, 47-49 nonattitudinal, 275-277 self-esteem, 61-62 stereotyping, 49-50 trait, 65-66 Justice system, discrimination, 203-204
Justification context, 377 ideological, 217-218, 220-225
K Knowledge declarative affective reactions ambiguous information, 21-22 episodic information, 22-25 overview, 4-5, 15-18 semantics, 18-19 attitudes, 7 selective encoding, 19-21 desire, 365 episodic, 22-25 needs, 349-350 procedural, 5, 22-25 stored, transference, 162-163
L Language abstractions in broadcasting, 90-91 components, 80 confirmability, 107-112 consequences, 102-105 development, 105-106 patterns, 100-101 bias, 100-101 concrete, 102-105 DAV application, 86-87 description, 80 differential, 101-102 free verbalization, 86 IAV application, 86-87 description, 80 personal beliefs, 104 spontaneous, 87-91 LEB, see Linguistic expectancy bias LIB model, see Linguistic intergroup bias model Life satisfaction information processing, 62, 64 Lincoln, Abraham, 268
INDEX Linguistic expectancy bias application, 102 description, 96 developmental studies, 106 mechanism, 101 message recipient, 107 Linguistic intergroup bias model abstractions, 107-1 12 abstract language use, 102-105 concrete language use, 102-105 confirmability, 107-1 12 controls, 112- 116 description, 79-82 developmental issues, 105-106 mass media political reports, 88-91 sports reports, 87-88 measurement, 85-87 message recipient, 106-107 paradigm, 82-83 replication, 85 underlying mechanisms determining, 91 -92 differential expectancy, 92-95 individual behavior, 95-96 language bias, 100-101 self-protection, 100-101 underlying mechanisms, 97-100
M Madness aspects, 345 attribution, 381 definition, 380 normal side, 381 physical aspects, 382 seeds. 467-468 societal impact, 382 symptoms, 381-385 variation, 381 Magic, 352 Manipulation, threat, 99 Marriage environments, 212-213 gender differences, 213 Marshall, Thurgood, 220-221 Marxism. influence, 201-202 Mass hysteria, 388, 405-406
495
Mass media political reports, 88-91 sports reports, 87-88 Mate preferences, 212-213 Mediation, 104 Memory accessibility, 15 affect cognitive consequence, 14-15 constructs, 5 discontinuity and, 364-365 recovered, 394 search, 352 transference evaluation, 140 evidence, 135-139 self role, 151 Milliken v. Bradley, 202 Milton, John, 378 Miracles, 352 Misfortune, 351 Moods, see also Emotions affect, 5-7 attention and. 20 onset, 6-7 transient, 146-149 Motivation affect impact consequence, 39-40 theory, 38-39 bias adjustments, 34-36 importance, 97-98 triggers, 99 discontinuity, 350-351 discrimination, 231-232 ingroup protection, 92 language bias, 100-101 personal, 372 satisfaction, 131- 132 self-protection, 100-101 transference, 140-142 Multiple choice alternatives, 85-87 description, 83
N Natural disasters, 352, 364 Navajo, social harmony, 359
496
INDEX
Noblesse oblige, 218, 220-221 Nodes, person, 312 Normalcy discontinuity and, 365-371 madness, 381 Norms, subjective, 310 Nyakusa, social harmony, 360
0 Obedience, unconditional, 394 Occupations, hierarchical value, 238-241 Orwell, George, 394
P Paranoia case studies, 397-399 characterization, 383-385 development, 383 discontinuity and, 379 experimentally-induced results, 437-444 study, 425-426,432 prejudice link, 388 sources, 386 Parataxic distortion, 130 Parsimony, 250 Past-negative factor, 387-389 People-related measure, 451 Perceptions assumptions and, 323-326 discontinuity, 353-354 Personalities dimensions, 205-208 rigid, 379 SDO, 213, 216 Personifications, childhood, 130 Person nodes, 312 Persuasion feed-back model, 50-52 impact, 1 information processing, 58-59 Planned behavior theory, 310-311 Political reports, 88-91 Politics arguments, bias, 251-252 attitudes, 268 corruption, 277-278
individual representation, 281-285 subtypes, 299-301 symbolic, 199-200 Prejudice, see also Bias individual, 192-195 paranoia link, 388 Priming repetition, 276 sources, 127, 156-158 Procedural discontinuity, 365 Procedural knowledge affect, 22-25 characterization, 5 Processing goal-directed, 9 strategies heuristic, 36-37, 54 substantive, 54 Productions respone-specif, 26 situation-specific, 26 Promiscuity, 213 Proprioceptive cues, 16 Protection ingroup LIB model, 92-95 motivation basis, 91-92 evidence, 92-95 individual role, 95-96 relevancy, 97-100 variation, 101-102 self, 100-101 Psychological process distress, physical effects, 380 social context attribution role, 196-197 linkage, 199-200 social structure, 251 Psychosexual-conflict model, 130
R Racism celebrities and, 287 correlates, 208-209 cultural frame, 216-217 in justice system, 203-204
INDEX social context, 308-311 discourse, 220-221 symbolic analysis, 199-200 Rationality, biased, 371-378 Reactions, see Affective reactions Reasoned action theory, 309-310 Reductionism, 249-250 Relationships close, 174-176 intergroup, 245-246 strained, 357-358 Repetition priming, 276 Representations accessibility, 287-288 attitude, see Attitude representation theory based assumptions, 320-323 generalizations, 291-294 inference-derived emergence, 129 evaluation, 135-139, 140 expectancies, 142-143 facial affect, 143-146 motivation, 140-142 self role, 151 triggers, 156-160 selective activation, 275-281 significant other, 130 closeness, activation, 141 insecurity, 131 principles, 155-156 resemblance, 134 self role, 150 theory-based content, 167 overview, 165-167 structure, 167-168 triggers, 168-170 types, 130 specific, 290-291 transient moods, 146-149 Repression model, 354-355 Reproduction, SDO, 210-213,245 Resemblance significant other, 134 to target, 136 Responses affect impact accuracy, 52
497
communication, 50-52 creativity, 45-46 impression formation, 46-49 nonlaboratory situations, 52-53 overview, 41-45 persuasion, 50-52 stereotyping, 49-50 attitude-relevant, 317 measures, 83, 85-87 Right-wing authoritarianism acceptance, 252-253 measuring, 192-193 SDO vs., 208-209 Rokeach vales, 209 Roosevelt, Franklin, 275 Roosevelt, Franklin D., 268 Rumination, 378 Russell, Bertrand. 365 RWA, see Right-wing authoritarianism S
Salem witchcraft, 399-400 San Antonio independent School District v Rodriguez, 202 Schreber, Dr. Daniel, 392-395 SDO, see Social dominance orientation Searches activities, 352 behavioral, 370-371 characterization, 367-370 cognitive, 365-367 confirmatory, 383 explanatory, 374-376, 383 external, 369 frames experimental, 434-444 theoretical, 374-376 internal, 368-369 memory, 352 normalcy, 402-405 predicting, 385-389 process, cognition, 365-367 Security, needs, 131 Segregation acceptance, 230 group, dominance, 197 Selective encoding, 19-21 Self-concepts close relations and, 174-176
498 Self-concepts (continued) transference content shifts, 151-153 evaluation shifts, 153-155 inference, 149-151 memory, 149-151 Self-esteem challenges, 378 collective, 99 delinquency and, 370 needs, 349-350 personal, 99 postexperimental, 99 Self-evaluation, affect-based general issues, 61-62 life satisfaction, 62, 64 likelihood estimates, 66 self-worth, 64-65 trait judgments, 65-66 Self-perception theory, 322-323 Self-protection, 100-101 Self-worth, 62 Semantics concept accessibility, 18-19 discontinuity. 364-365 Sexism correlates, 208-209 cultural frame, 216-217 Sexual jealousy, 212-213 Shapiro, Karl, 367 Significant others closeness, activation, 141 insecurity, 131 principles, 155-156 representations, 130 resemblance, 134 self role, 150 theory-based analogous data, 176-178 attachment, 171-174 content, 167 cross-cultural, 174-176 overview, 165-167 self/other roles, 174-176 structure, 167-168 triggers, 168-170 Sociability, stereotyping, 96 Social-cognitive models group dominance, 196-1 98 transference evidence, 135-139
INDEX idiographic methods, 132-135 manifestations, 128-129 nomothetic methods, 132-135 origins clinical, 130-132 constructs, 125- 126 principles, 155-156 self role construct, 149-151 content shifts, 151-153 evaluation shifts, 153-155 inference, 151 memory, 151 theory-based analogous data, 176-178 attachment, 171- 174 content, 167 cross-cultural, 176 overview, 165-167 self/other roles, 174-176 structure, 167-168 triggers, 168-170 triggers application, 158-165 constructs, 126- 128 examination, 156-158 Social comparison process, 350 theory, 371 Social dominance orientation cultural context, 213, 216-217, 246-247 definition, 205 group context, 233-237 discrimination, 233-237 dominance relation, 209-210 hierarchy roles, 237-241 ideological usage, 216-217 measuring, 205 personality, 213, 216 reproductive strategies, 210-213 social ideologies hierarchy role, 237-241 justification, 217-218, 220-225 uniqueness, 205-208 Social dominance theory discrimination, 204 group-based dominance, 195 Social ideologies, see Ideologies
INDEX Social-information-processingmodel. 384-385 Socioeconomic status, maintenance, 388 Sports reports, 87-88 Stability attitudes, 265 response, 281-284 Stanford prison experiment, 241 State verbs, derivation, 86 Stereotypes abstract, 107-112 behavior induction, 198 categorization, 164 Chinese, 266-267, 302 comparing, 111-112 concrete, 107-112 cues, 197 discrimination, association, 195 feed-back model, 49-50 in-group motives, 101-102 Italians, 93-95 Jews, 88-91 perpetuation, 80 reliance on, factors, 314-315 tokens, 197-198 transmission, 106 Stimulus configuration, 3-4 information processing, 8 priming, 12 Stress, conditions, 364 Structures, 251 Systematic group discrimination, 202-203
T TAT cards, see Thematic Apper Caption Test cards Tenderness theorem, 131 Thematic Apper Caption Test cards, 21 Thomas, Clarence, 220-221 Threat, manipulative, 99 Tokenism, 197-198 Totalitarian ego, 367 Totenberg, Nina, 220 Traits, judgment, 65-66 Transference client/therapist, 132
499
clinical, 124 inference, 135-140 interpersonal role, 146-149 manifestations, 128-139 memory, 135-139 origins clinical, 130-132 constructs, 125-126 phenomenon, 123-125 research methods idiographic, 132-135 nomot hetic, 132- 135 self role construct, 149-151 content shifts, 151-153 evaluation shifts, 153-155 inference, 149-151 memory, 149-151 theor y-based analogous data, 176-178 attachment. 171-174 content, 167 cross-cultural, 176 overview, 165-167 selflother roles, 174-176 structure, 167-168 triggers, 168-170 transient moods, 146-149 triggers application construct, 158-160 feature-based, 161- 165 theory-based, 161 constructs, 126-128 examinations, 156-158 ubiquitousness, 132 Two-value theory, 194
U Uncertainty, fear of, 195
V Values aesthetic, 352 default, 270, 272-273 Voltaire, FranGois Maire, 382
500
INDEX
W Washington, George, 268 Watts riots, 353-354 Welfare programs, 218
Winfrey, Oprah, 287 Women, jealousy, 212-213 r.
1
Zimbardo Time Perspective Inventory, 387
CONTENTS OF OTHER VOLUMES
Volume 1 Cultural Influences upon Cognitive Processes Harry C. Triandis The Interaction of Cognitive and Physiological Determinants of Emotional State Sianiey Schachter Experimental Studies of Coalition Formation William A. Gamson Communication Networks Marvin E. Shaw A Contingency Model of Leadership Effectiveness Fred E. Fiedler Inducing Resistance to Persuasion: Some Contemporary Approaches William J . McCuire Social Motivation, Dependency, and Susceptibility to Social Influence Richard H. Waiters and Ross D.Purke Sociability and Social Organization in Monkeys and Apes William A. Mason Author Index-Subject Index
Volume 2 Vicarious Processes: A Case of No-Trial Learning Albert Bandura Selective Exposure Jonathan L. Freedman and David 0. Sears Group Problem Solving L. Richard Hoffman Situational Factors in Conformity Vernon L. Allen Social Power John Schopler
From Acts to Dispositions: The Attribution Process in Person Perception Edward E. Jones and Keith E. Davis Inequality in Social Exchange J. Stacy Adams The Concept of Aggressive Drive: Some Additional Considerations Leonard Berkowitz Author Index-Subject Index
Volume 3 Mathematical Models in Social Psychology Robert P. A beison The Experimental Analysis of Social Performance Michael Argyie and Adam Kendon A Structural Balance Approach to the Analysis of Communication Effects N. T. Feather Effects of Fear Arousal on Attitude Change: Recent Developments in Theory and Experimental Research Irving L. Janis Communication Processes and the Properties of Language Serge Moscovici The Congruity Principle Revisited: Studies in the Reduction, Induction and Generalization of Persuasion Percy H. Tannenbaum Author Index-Subject Index Volume 4 The Theory of Cognitive Dissonance: A Current Perspective Elliot Aronson
501
502
CONTENTS O F OTHER VOLUMES
Attitudes and Attraction Donn Byrne Sociolinguistics Susan M. Ervin-Tripp Recognition of Emotion Nico H. Frijda Studies of Status Congruence Edward E. Sampson Exploratory Investigations of Empathy Ezra Stotland The Personal Reference Scale: An Approach to Social Judgment Harry S. Upshaw Author Index-Subject Index
Volume 5 Media Violence and Aggressive Behavior: A Review of Experimental Research Richard E. Goranson Studies in Leader Legitimacy, Influence, and Innovation Edwin P. Hollander and James W . Julian Experimental Studies of Negro-White Relationships Irwin Katz Findings and Theory in the Study of Fear Communications Howard Leventhal Perceived Freedom Ivan D. Steiner Experimental Studies of Families Nancy E. Waxler and Elliot G. Mishler Why Do Groups Make Riskier Decisions than Individuals? Kenneth L. Dion, Robert S. Baron, and Norman Miller Author Index-Subject Index
Volume 6 Self-perception Theory Daryl J. Bern Social Norms, Feelings, and Other Factors Affecting Helping and Altruism Leonard Berkowitz The Power of Liking: Consequence of Interpersonal Attitudes Derived from a Liberalized View of Secondary Reinforcement Albert J. Lott and Bernice E. Lott
Social Influence, Conformity Bias, and the Study of Active Minorities Serge Moscovici and Claude Faucheux A Critical Analysis of Research Utilizing the Prisoner’s Dilemma Paradigm for the Study of Bargaining Charlan Nemeth Structural Representations of Implicit Personality Theory Seymour Rosenberg and Andrea Sedlak Author Index-Subject Index
Volume 7 Cognitive Algebra: Integration Theory Applied to Social Attribution Norman A. Anderson On Conflicts and Bargaining Erika Apfelbaum Physical Attractiveness Ellen Bersheid and Elaine Walster Compliance, Justification, and Cognitive Change Harold B. Gerard, Edward S. Connolley, and Roland A. Wilhelmy Processes in Delay of Gratification Walter Mischel Helping a Distressed Person: Social, Personality, and Stimulus Determinants Ervin Staub Author Index-Subject Index
Volume 8 Social Support for Nonconformity Vernon L. Allen Group Tasks, Group Interaction Process, and Group Performance Effectiveness: A Review and Proposed Integration J. Richard Hackman and Charles G. Morris The Human Subject in the Psychology Experiment: Fact and Artifact Arie W . Kruglanski Emotional Arousal in the Facilitation of Aggression through Communication Percy H. Tannenbaum and Dolf Zillman The Reluctance to Transmit Bad News Abraham Tesser and Sidney Rosen Objective Self-Awareness Robert A. Wicklund
CONTENTS OF OTHER VOLUMES Responses to Uncontrollable Outcomes: An Integration of Reactance Theory and the Learned Helplessness Model Camille B. Wortman and Jack W . Brehm Subject Index Volume 9 New Directions in Equity Research Elaine Walster, Ellen Berscheid, and G. William Walster Equity Theory Revisited: Comments and Annotated Bibliography J. Stacy Adams and Sara Freedman The Distribution of Rewards and Resources in Groups and Organizations Gerald S. Levenihal Deserving and the Emergence of Forms of Justice Melvin J. Lerner, Dale T. Miller, and John G. Holmes Equity and the Law: The Effect of a Harmdoer’s “Suffering in the Act” on Liking and Assigned Punishment William Austin, Elaine Walster, and Mary Kristine Utne Incremental Exchange Theory: A Formal Model for Progression in Dyadic Social Interaction L. Lowell Huesmann and George Levinger Commentary George C. Homans Subject Index Volume 10 The Catharsis of Aggression: An Evaluation of a Hypothesis Russell G. Geen and Michael B. Quanty Mere Exposure Albert A. Harrison Moral Internalization: Current Theory and Research Martin L. Hoffman Some Effects of Violent and Nonviolent Movies on the Behavior of Juvenile Delinquents Ross D. Parke, Leonard Berkowitz, Jacques P. Leyens, Stephen G. West, and Richard Sebastian
503
The Intuitive Psychologist and His Shortcomings: Distortions in the Attribution Process Lee Ross Normative Influences on Altruism Shalom H. Schwartz A Discussion of the Domain and Methods of Social Psychology: Two Papers by Ron Harre and Barry R. Schlenker Leonard Berkowitz The Ethogenic Approach: Theory and Practice R. Harre On the Ethogenic Approach: Etiquette and Revolution Barry R. Schlenker Automatisms and Autonomies: In Reply to Professor Schlenker R. Harre Subject Index Volume 11 The Persistence of Experimentally Induced Attitude Change Thomas D. Cook and Brian F. Flay The Contingency Model and the Dynamics o f the Leadership Process Fred E. Fiedler An Attributional Theory of Choice Andy Uukla Group-Induced Polarization of Attitudes and Behavior Helmut Lamm and David G . Myers Crowding: Determinants and Effects Janet E. Stockdale Salience: Attention, and Attribution: Top of the Head Phenomena Shelley E. Taylor and Susan T. Fiske Self-Generated Attitude Change Abraham Tesser Subject Index Volume 12 Part 1. Studies in Social Cognition Prototypes in Person Perception Nancy Cantor and Walter Mischel A Cognitive-Attributional Analysis of Stereotyping David L. Hamilton
504
CONTENTS OF OTHER VOLUMES
Self-Monitoring Processes Mark Snyder Part 11. Social Influences and Social Interaction Architectural Mediation of Residential Density and Control: Crowding and the Regulation of Social Contact Andrew Baum and Stuart Valins A Cultural Ecology of Social Behavior J. W . Berry Experiments on Deviance with Special Reference to Dishonesty David P. Farrington From the Early Window to the Late Night Show: International Trends in the Study of Television's Impact on Children and Adults John P. Murray and Susan Kippax Effects of Prosocial Television and Film Material on the Behavior of Viewers J. Phillipe Rushton Subject Index
Volume 13 People's Analyses of the Causes of AbilityLinked Performances John M. Darley and George R. Goethals The Empirical Exploration of Intrinsic Motivational Processes Edward I. Deci and Richard M. Ryan Attribution of Responsibility: From Man the Scientist to Man as Lawyer Frank D. Fincham and Joseph M. Jaspars Toward a Comprehensive Theory of Emotion Howard Leventhal Toward a Theory of Conversion Behavior Serge Moscovici The Role of Information Retrieval and Conditional Inference Processes in Belief Formation and Change Robert S. Wyer, Jr. and Jon Hartwick Index
Cognitive, Social, and Personality Processes in the Physiological Detection of Deception William M. Waid and Martin T. Orne Dialectic Conceptions in Social Psychology: An Application to Social Penetration and Privacy Regulation Irwin Altman, Anne Vinsel, and Barbara B. Brown Direct Experience and Attitude-Behavior Consistency Russell H. Fazio and Mark P. Zanna Predictability and Human Stress: Toward a Clarification of Evidence and Theory Suzanne M. Miller Perceptual and Judgmental Processes in Social Contexts Arnold Upmeyer Jury Trials: Psychology and Law Charlan Jeanne Nemeth Index
Volume 15 Balance, Agreement, and Positivity in the Cognition of Small Social Structures Walter H. Crockett Episode Cognition: Internal Representations of Interaction Routines Joseph P. Forgas The Effects of Aggressive-Pornographic Mass Media Stimuli Neil M. Malamuth and Ed Donnerstein Socialization in Small Groups: Temporal Changes in Individual-Group Relations Richard L. Moreland and John M. Levine Translating Actions into Attitudes: An Identity-Analytic Approach to the Explanation of Social Conduct Barry R. Schlenker Aversive Conditions as Stimuli to Aggression Leonard Berkowitz Index
Volume 14
Volume 16
Verbal and Nonverbal Communication of Deception Miron Zuckerman, Bella M. DePaulo, and Robert Rosenthal
A Contextualist Theory of Knowledge: Its Implications for Innovation and Reform in Psychological Research William J. McGuire
CONTENTS OF OTHER VOLUMES Social Cognition: Some Historical and Theoretical Perspectives Janet Landman and Melvin Manis Paradigmatic Behaviorism: Unified Theory for Social-Personality Psychology Arthur W. Staats Social Psychology from the Standpoint of a Structural Symbolic Interactionism: Toward an Interdisciplinary Social Psychology Sheldon Stryker Toward an Interdisciplinary Social Psychology Carl W. Backman Index
Volume 17 Mental Representations of Self John F. Kihlstrom and Nancy Cantor Theory of the Self: Impasse and Evolution Kenneth J. Gergen A Perceptual-Motor Theory of Emotion Howard Leventhal Equity and Social Change in Human Relationships Charles G. McClintock, Roderick M. Kramer, and Linda J. Keil A New Look at Dissonance Theory Joel Cooper and Russell H. Fazio Cognitive Theories of Persuasion Alice H. Eagly and Shelly Chaiken Helping Behavior and Altruism: An Empirical and Conceptual Overview John F. Dovidio Index
Volume 18 A Typological Approach to Marital Interac-
tion: Recent Theory and Research Mary Anne Fitzpatrick Groups in Exotic Environments Albert A. Harrison and Mary M. Connors Balance Theory, the Jordan Paradigm, and the Wiest Tetrahedon Chester A. Insko The Social Relations Model David A. Kenny and Lawrence La Voie Coalition Bargaining S. S. Komorita
505
When Belief Creates Reality Mark Snyder Index
Volume 19 Distraction-Conflict Theory: Progress and Problems Robert S. Baron Recent Research on Selective Exposure to Information Dieter Frey The Role of Threat to Self-Esteem and Perceived Control in Recipient Reaction to Help: Theory Development and Empirical Validation Arie Nadler and Jeffrey D. Fisher The Elaboration Likelihood Model of Persuasion Richard E. Petty and John T. Cacioppo Natural Experiments on the Effects of Mass Media Violence on Fatal Aggression: Strengths and Weaknesses of a New Approach David P. Phillips Paradigms and Groups Ivan D. Steiner Social Categorization: Implications for Creation and Reduction of Intergroup Bias David A. Wilder Index
Volume 20 Attitudes, Traits, and Actions: Dispositional Prediction of Behavior in Personality and Social Psychology Icek Ajzen Prosocial Motivation: Is It Ever Truly Altruistic? C. Daniel Batson Dimensions of Group Process: Amount and Structure of Vocal Interaction James M. Dabbs, Jr. and R. Barry Ruback The Dynamics of Opinion Formation Harold B. Gerard and Ruben Orive Positive Affect, Cognitive Processes, and Social Behavior Alice M. lsen
506
CONTENTS O F OTHER VOLUMES
Between Hope and Fear: The Psychology of Risk Lola L. Lopes Toward an Integration of Cognitive and Motivational Perspectives on Social Inference: A Biased Hypothesis-Testing Model Tom Pyszczynski and Jeff Creenberg Index
Volume 21 Introduction Leonard Berkowitz Part 1. The Self as Known Narrative and the Self as Relationship Kenneth J. Gergen and Mary M. Gergen Self and Others: Studies in Social Personality and Autobiography Seymour Rosenberg Content and Process in the Experience of Self William J. McCuire and Claire V. McCuire Information Processing and the Study of the Self John F. Kihlstrom, Nancy Cantor, Jeanne Sicmi Albright, Beverly R. Chew, Stanley B. Klein, and Paula M. Niedenthal Part 11. Self-Motives Toward a Self-Evaluation Maintenance Model of Social Behavior Abraham Tesser The Self A Dialectical Approach Carl W . Backman The Psychology of Self-Affirmation: Sustaining the Integrity of the Self Claude M. Steele A Model of Behavioral Self-Regulation: Translating Intention into Action Michael F. Scheier and Charles S. Carver Index
Volume 22 On the Construction of the Anger Experience: Aversive Events and Negative Prirning in the Formation of Feelings Leonard Berkowitz und Karen Heimer
Social Psychophysiology: A New Look John T. Cacioppo, Richard E. Petty, and Louis G. Tassinary Self-Discrepancy Theory: What Patterns of Self-Beliefs Cause People to Suffer? E. Tory Higgins Minding Matters: The Consequences of Mindlessness-Mindfulness Ellen J. Langer The Tradeoffs of Social Control and Innovation in Groups and Organizations Charlan Jeanne Nemeth and Barry M. Staw Confession, Inhibition, and Disease James W . Pennebaker A Sociocognitive Model of Attitude Structure and Function Anthony R. Pratkanis and Anthony C. Greenwald Introspection, Attitude Change, and Attitude-Behavior Consistency: The Disruptive Effects of Explaining Why We Feel the Way We Do Timothy D. Wilson, Dana S. Dunn, Dolores Krafl, and Douglas J. Lisle Index
Volume 23 A Continuum of Impression Formation, from Category-Based to Individuating Processes: Influences of Information and Motivation on Attention and Interpretation Susan T. Fiske and Steven L. Neuberg Multiple Processes by Which Attitudes Guide Behavior: The MODE Model as an Integrative Framework Russell H. Fazio PEAT: An Integrative Model of Attribution Processes John W . Medcof Reading People’s Minds: A Transforrnation Rule Model for Predicting Others’ Thoughts and Feelings Rachel Karniol Self-Attention and Behavior: A Review and Theoretical Update Frederick X . Gibbons Counterfactual Thinking and Social Perception: Thinking about What Might Have Been
CONTENTS OF OTHER VOLUMES
Dale T. Miller, William Turnbull, and Cathy McFarland Index Volume 24 The Role of Self-Interest in Social and Political Attitudes David 0. Sears and Carolyn L. Funk A Terror Management Theory of Social Behavior: The Psychological Functions of Self-Esteem and Cultural Worldviews Sheldon Solomon, Jeff Greenberg, and Tom Pyxzczynski Mood and Persuasion: Affective States Influence the Processing of Persuasive Communications Norbert Schwarz, Herbert Bless, and Gerd Bohrier A Focus Theory of Normative Conduct: A Theoretical Refinement and Reevaluation of the Role of Norms in Human Behavior Robert B. Cialdini, Carl A. Kallgren, and Raymond R. Reno The Effects of Interaction Goals on Person Perception James L. Hilton and John M. Darley Studying Social Interaction with the Rochester Interaction Record Harry T. Reis and Ladd Wheeler Subjective Construal, Social Inference, and Human Misunderstanding Dale W. Grifjin and Lee Ross Index Volume 25 Universals in the Content and Structure of Values: Theoretical Advances and Empirical Tests in 20 Countries Shalom H. Schwartz Motivational Foundations of Behavioral Confirmation Mark Snyder A Relational Model of Authority in Groups Tom R. Tyler and E. Allan Lind You Can’t Always Think What You Want: Problems in the Suppression of Unwanted Thoughts Daniel M. Wegner
507
Affect in Social Judgments and Decisions: A Multiprocess Model Joseph Paul Forgas The Social Psychology of Stanley Milgram Thomas Blass The Impact of Accountability on Judgment and Choice: Toward a Social Contingency Model Philip E. Tetlock Index
Volume 26 Attitudes toward High Achievers and Reactions to Their Fall: Theory and Research Concerning Tall Poppies N. T. Feather Evolutionary Social Psychology: From Sexual Selection to Social Cognition Douglas T. Kenrick Judgment in a Social Context: Biases, Shortcomings, and the Logic of Conversation Norbert Schwarz A Phase Model of Transitions: Cognitive and Motivational Consequences Diane N. Ruble Multiple-Audience Problems, Tactical Communication, and Social Interaction: A Relational-Regulation Perspective John H. Fleming From Social Inequality to Personal Entitlement: The Role of Social Comparisons, Legitimacy Appraisals, and Group Membership Brenda Major Mental Representations of Social Groups: Advances in Understanding Stereotypes and Stereotyping Charles Stangor and James E. Lange Index Volume 27 Inferences of Responsibility and Social Motivation Bernard Weiner Information Processing in Social Contexts: Implications for Social Memory and Judgment Robert S. Wyer, Jr., and Deborah H. Gruenfeld
508
CONTENTS O F OTHER VOLUMES
The Interactive Roles of Stability and Level of Self-Esteem: Research and Theory Michael H. Kernis and Stephanie B. Waschull Gender Differences in Perceiving Internal State: Toward a His-and-Hers Model of Perceptual Cue Use Tomi-Ann Roberts and James W . Pennebaker On the Role of Encoding Processes in Stereotype Maintenance William von Hippel, Denise Sekaquaptewa, and Patrick Vargas Psychological Barriers to Dispute Resolution Lee Ross and Andrew Ward Index
Volume 28 The Biopsychosocial Model of Arousal Regulation Jim Blascovich and Joe Tomaka Outcome Biases in Social Perception: Implications for Dispositional Inference, Attitude Change, Stereotyping, and Social Behavior Scott T. Allison, Diane M. Mackie, and David M. Messick Principles of Judging Valence: What Makes Events Positive or Negative? C. Miguel Brendl and E. Tory Higgins Pluralistic Ignorance and the Perpetuation of Social Norms by Unwitting Actors Deborah A. Prentice and Dale T. Miller People as Flexible Interpreters: Evidence and Issues from Spontaneous Trait Inference James S. Uleman, Leonard S. Newman, and Gordon B. Moskowitz Social Perception, Social Stereotypes, and Teacher Expectations: Accuracy and the Quest for the Powerful Self-Fulfilling Prophecy Lee Jussim, Jacquelynne Eccles, and Stephanie Madon Nonverbal Behavior and Nonverbal Communication: What Do Conversational Hand Gestures Tell Us?
Robert M. Krauss, Yihsiu Chen, and Purnima Chawla Index
Volume 29 Counterfactual Thinking: The Intersection of Affect and Function Neal J. Roese and James M. Olson Terror Management Theory of Self-Esteem and Cultural Worldviews: Empirical Assessments and Conceptual Refinements Jeff Greenberg, Sheldon Solomon, and Tom Pyszczynski The Flexible Correction Model: The Role of Naive Theories of Bias in Bias Correction Duane T. Wegener and Richard E. Petty Self-Evaluation: To Thine Own Self Be Good, to Thine Own Self Be Sure, to Thine Own Self Be True, and to Thine Own Self Be Better Constantine Sedikides and Michael J . Strube Toward a Hierarchical Model of Intrinsic and Extrinsic Motivation Robert J. Vaiierand
Volume 30 Promotion and Prevention: Regulatory Focus as a Motivational Principle E. Tory Higgins The Other “Authoritarian Personality” Bob Altemeyer Person Preception Comes of Age: The Salience and Significanceof Age in Social Adjustments Joann M. Montepare and Leslie A . Zebrowitz On the Perception of Social Consensus Joachim Krueger Prejudice and Stereotyping in Everyday Communication Janet B. Ruscher Situated Optimism: Specific Outcome Expectancies and Self-Regulation David A. Armor and Shelley E. Taylor