Consciousness and Cognition EDITOR William P. Banks Pomona College, Claremont, California
Bruce Bridgeman University of California Santa Cruz
ASSOCIATE EDITORS Axel N. Cleeremans
James T. Enns
Universite´ Libre de Bruxelles
University of British Columbia
Antti Revonsuo University of Turku
EDITORIAL BOARD Jackie Andrade
Steven A. Hillyard
Keith Oatley
University of Plymouth
University of California San Diego
Ontario Institute for Studies in Education
J. Allan Hobson
Steven Palmer
Bernard J. Baars The Neurosciences Institute San Diego
Talis Bachmann
Massachusetts Mental Health Center
University of Tartu
Larry L. Jacoby
Alan Baddeley
New York University
MRC Applied Psychology Unit
E. Roy John
John Bargh
New York University Medical Center
New York University
Arthur L. Blumenthal
John F. Kihlstrom
The New School University
University of California Berkeley
Gordon H. Bower
Christof Koch
Stanford University
Deborah Burke Pomona College
Wallace Chafe University of California Santa Barbara
David Chalmers Antonio Damasio
Stephen LaBerge
University of Iowa
Stanford University
Meredyth Daneman
Donald G. MacKay
University of Wisconsin
Daniel C. Dennett
University of Louisville
Ernst Pçppel Ludwig-MaximiliansUniversitt, Mnchen
William Prinzmetal University of California Berkeley
Arthur Reber
David Rosenthal
David LaBerge
University of Arizona Tucson
Richard Davidson
John Pani
California Institute of Technology Brooklyn College of CUNY Stephen M. Kosslyn Harvard University Eyal Reingold University of Toronto Alfred B. Kristofferson Ontario, Canada University of California Irvine
University of Toronto
University of California Berkeley
University of California Los Angeles
George Mandler University of California San Diego
Tufts University
Bruce Mangan
Andreas K. Engel Hamburg University
University of California Berkeley
Matthew Erdelyi
Graduate School of CUNY
Daniel Schacter Harvard University
Arnold Scheibel University of California Los Angeles
Jonathan W. Schooler University of Pittsburgh
Tim Shallice University College London
Jerome L. Singer
Anthony Marcel
Yale University
Brooklyn College of CUNY
MRC Applied Psychology Unit
David Spiegel
Owen Flanagan
Hazel R. Markus
Duke University
University of Michigan
Stanford University School of Medicine
David Galin
Philip M. Merikle
Petra Stoerig
Langley Porter Psychiatric Institute, San Francisco
University of Waterloo
Heinrich-Heine-Universitat
Thomas Metzinger
Giulio Tononi
Michael S. Gazzaniga
The Neurosciences Institute
Dartmouth College
Johannes GutenbergUniversitt Mainz
Anthony G. Greenwald
Jeff Miller
University of Nottingham
University of Washington
University of Otago
Henk J. Haarman
Michael C. Mozer
University of Maryland
University of Colorado
Stevan Harnad
W. Trammell Neill
Princeton University
University at Albany
Geoffrey Underwood Daniel M. Wegner Harvard University
Charles Yingling University of California San Francisco
Consciousness and Cognition 19 (2010) 687–689
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Editorial
Self, other and memory: A preface q
The key topics of this special issue are self-consciousness, understanding others and the role of memory, while the latter is especially investigated with respect to its relation to self-consciousness. It is furthermore a characteristic feature of this special issue that most of the papers are dedicated to an interdisciplinary research program combining the perspectives of philosophy, psychology, neurobiology and cognitive neuroscience. The issue of self-consciousness is firstly investigated concerning the feature of perspective taking (Kockler et al.). This empirical study is complemented by two studies that focus on embodiment: A basic constituent of bodily self-consciousness is surely the anchoring of self-related experiences in one’s own body image (Blanke et al.) and an interesting case is being made of language as being related to embodiment as well (Binkofski et al.). Then three articles are dealing with self-deception including some interconnected discussion (Greve & Wentura; Michel & Newen, Mele). Furthermore, a theoretical investigation of self-consciousness refers to the compositionality of our thoughts (Werning). Investigating the self is contrasted by an empirical study on the perception of others under specific influence of different cultural backgrounds (Bente et al.). With the article of Staniloiu et al. the volume widens the discussion to account for memory as well. Three empirical investigations help to improve our understanding of the role of memory while one theoretical account discusses a core thesis about memory: One important focus is the question what is necessary to establish individual self-consciousness over time (Staniloui et al.): this leads to a discussing of diachronic self-consciousness and autobiographic episodic memory. The development of episodic memory is the focus of the behavioral study of Perner et al. arguing that the ability of mental rotation of figures is initiating a switch from knowledge retrieval to episodic remembering. In addition to the distinction of episodic and semantic memory, Sauvage discusses convincing evidence to support the distinction between familiarity and recollection. The paper of Volz et al. is investigating our subjective experience of intuition as addressed by a fluency heuristic paradigm from the perspective of cognitive neuroscience. Finally, in a philosophical paper Vosgerau argues that, on the basis of a functionalist approach of mental representation, representations only have a specific content if they are actually in use, therefore, stored mental representations do not have a specific content. We shortly characterize the main claims of the papers: One feature of self-consciousness is the capacity to take different perspectives: closely related to an intact body image is our capacity to change perspectives in space. This already ‘‘classical” empirical approach to study spatial perspective taking has been investigated by Kockler et al. for the first time in dynamic environments. In their study ‘‘Visuospatial perspective taking in a dynamic environment” participants were asked to perform left–right-decisions in animated and static virtual environments, both from a first- and third-person-perspective. The results showed a significantly increased activation in the right posterior intraparietal sulcus in conditions in which a dynamic stimulus had to be observed and judged upon from a first-person-perspective. This special case of oneself being involved in object-directed action preparation is interpreted and discussed as the neural mechanism of ‘‘readiness for (re)action”. A very important approach to any cognitive function are disturbances of self-consciousness, this important and sometimes underestimated approach is employed by Blanke et al. in their contribution concerning bodily self-consciousness. The basic key assumption here is that the integration of multisensory bodily signals from the entire body is crucial for bodily self-consciousness. Not only empirically grounded, but also philosophically arguable they refer to a newly developed taxonomy that comprises the three different features of self-location, first person perspective, and self-identification and that allows to categorize body schema disturbances. The authors are focusing on illusory body perceptions of one’s own body as a particularly interesting disturbance of the body schema or body representation. Based on detailed clinical and neuroanatomical data the paper presents two interesting case reports with a disturbance of self-identification and shows their relevance for the neural mechanisms underlying bodily self-consciousness.
q
This article is part of a special issue of this journal on Self, Other and Memory.
1053-8100/$ - see front matter Ó 2010 Published by Elsevier Ltd. doi:10.1016/j.concog.2010.07.002
688
Editorial / Consciousness and Cognition 19 (2010) 687–689
On the background of such empirical evidence, the concept of embodied cognition has become one the leading frameworks in the study of cognitive processes in general. A very interesting and stimulating approach refers to the hypothesis that also language can be interpreted as embodied cognition. In the study ‘‘Grasping language – a short story on Embodiment.” the authors Binkofski et al. employ a new meta-analytic tool that allows to study neural mechanisms of cognitive functions across different studies. The key finding of cumulative evidence for the recruitment of the somatotopic activation of motor areas clearly supports this hypothesis of language as embodied cognitive faculty and emphasizes the role of sensorimotor areas during language processing. The discussion of self and embodiment is followed by a group of three papers on self-deception. There is few disagreement in the recent philosophical and psychological literature on self-deception about the fact that garden-variety selfdeception differs from stereotypical other-deception. What we call self-deception in everyday contexts can be explained within a framework of a unified self and does not require any partitioning of the self into different sub-centers of agency in order to enable mutual deception. But there is an intense debate about the role of motivational processes in self-deception which is the main focus of the contributions in this volume. Greve & Wentura assess two explanatory virtues that cognitive processes of self-immunization have with regard to selfdeception. The authors approach self-deception from the perspective of cognitive psychology. They draw on empirical evidence for unconscious processes of self-immunization they have provided in a series of studies. Via semantic priming tasks they have tracked automatic and sub-personal processes of self-immunization. On the basis of this evidence Greve & Wentura demonstrate how we balance stability and changes in our self-images in a self-serving way when the pleasure- and the reality-principle have clashed and need to be reconciled. As a sub-personal process, the authors show that self-immunization does not fall victim to the paradoxes of self-deception that arise on a personal level. Michel & Newen show in detail to which degree self-deception remains rational by providing a general model of self-deception that describes self-deception in a proper sense as pseudo-rational regulation of belief-systems. They distinguish self-deception from alternative forms of motivational influence on acceptance (in opposition to the position of Mele) which have been proposed in the recent philosophical literature such as motivational bias and pretense. Self-deceivers violate their own standards of belief-regulation by employing a dual standard of reasoning that remains opaque to them. Drawing on evidence from cognitive psychology on processes of self-immunization and dual rationality the authors demonstrate how self-deception against strong counter-evidence can be reconciled with the subject’s attaining belief-status for the target proposition. The question of belief-status is in the focus of Alfred Mele’s contribution. Mele has put forward an influential analysis which assumes that self-deceivers actually develop a belief of the target proposition. In contrast Robert Audi’s analysis claims that self-deceptive avowals are lacking belief-status. Mele provides evidence for the belief-view which has folk-intuition on its side. He has carried out two surveys that do not support Audi’s claim that a pre-theoretical understanding of the concept of ‘‘self-deception” includes the absence of belief. The results favor the opposite view. Concerning self-consciousness, Werning discusses the question whether it makes sense to presuppose a Cartesian inner self as the bearer of mental content. He argues that we need such an inner self without presupposing Descartes’ dualism. It is argued that the idea of an inner self is unavoidable if we accept two principles: the phenomenal transparency of experience and the semantic compositionality of conceptual content. It is assumed that self-awareness is a second-order state either in the domain of experience or in the domain of thought. In the former case self-awareness turns out empty if experience is transparent. In the latter, it can best be conceived of as a form of mental quotation. Since the only theory of quotation that is compositional is a phonological theory, we have to presuppose an inner self as the bearer of inner speech guaranteeing the compositionality of thought. An innovative transcultural study on the perception of nonverbal behavior that allows to study processes related to person perception and impression formation of others is presented by Bente et al. in their study ‘‘The Others. Universals and Cultural Specificities in the Perception of Status and Dominance from Nonverbal Behavior.” The authors employed a novel methodology that allowed to study nonverbal behavior from different countries in a culture-fair manner by masking to ethnicity and culture, the results of which show that the processing of dominance cues is universal whereas evaluative responses to nonverbal behavior are culture-dependent. This study is an important contribution to the literature on person perception in general and on its cultural influences in particular. In a bundle of four papers memory dimensions are investigated in detail (Staniloiu et al., Perner et al., Sauvage; Vosgerau). From a clinical neuropsychological point of view, Staniloiu et al. study the interesting case ‘‘Psychogenic Amnesia – A Malady of the Constricted Self.” referring to autobiographical–episodic memory as key component of self-consciousness. Autobiographical–episodic memory is understood as the conjunction of subjective time, autonoetic consciousness and the experiencing self. Neurobiologically, the prefrontal cortex and the limbic system are assumed to play key roles. These brain regions appear to be particularly vulnerable to stress-induced brain and are therefore of relevance for disorders of the autobiographical memory like psychogenic amnesia that are linked to environmental influences. The aim of Perner et al. is to investigate the role of the ability of mental rotation for episodic memory. To do that they investigate the common development of children’s ability to ‘‘look back in time” (retrospection, episodic remembering) and to ‘‘look into the future” (prospection). They register free recall, as a measure of episodic remembering (i) with recall of visually experienced events and (ii) with recall of indirectly conveyed events. Quite unexpectedly, ‘‘mental rotators” were markedly worse on indirect items than ‘‘non-rotators”. They speculate that with the ability to rotate children switch from the ability of knowledge retrieval to episodic remembering.
Editorial / Consciousness and Cognition 19 (2010) 687–689
689
In her contribution ‘‘ROC in animals: uncovering the neural substrates of recollection and familiarity in episodic recognition memory.” Magdalena Sauvage focuses to the basic neurobiological mechanisms of episodic memory and explores the question whether familiarity and recollection are qualitatively distinct processes that in turn recruit different brain regions or whether they refer to the same process which would let us to assume that shared brain mechanisms are at work. Based on an animal model and innovative research methodologies she favours the view that familiarity and recollection are different processes, from a neurobiological point of view she provides data that lead to the conclusion that the hippocampus supports recollection, but not familiarity. Also related to recognition memory is the employment of simple heuristics that allows decision making on the ground of only sparse information. Volz et al. study the so-called fluency heuristic based on the relative speed with which two different objects are recognized in their study. ‘‘It just felt right: The neural correlates of the fluency heuristic.” Searching for the neural mechanisms of the fluency heuristic showed that the claustrum was recruited during fluency heuristic decisions suggesting that this brain region is responsible for the integration perceptual and memory elements into a conscious ‘‘Gestalt” and, hence, for the subjective experience of fluency. Vosgerau argues that any theory of content has to adopt a ‘‘functionalistic core” according to which representations are defined as substitutes in functions that describe the flexible behavior to be explained by the representation. The content of a representation can thus only be determined if the representation is ‘‘in use”. The stored entities in memory are not in use while they are stored, and hence cannot be assigned a specific content. The term ‘‘template” is introduced to describe stored entities in memory while activated memory is essentially in need of a construction process on the basis of the templates. Early versions of most of the papers were first presented at an international and interdisciplinary conference at the Hanse-Institute of Advanced Study in Delmenhorst (Germany) organized by the editors in summer 2009. This conference was a highlight of a series of ongoing interdisciplinary cooperations including the edition of this volume. We want to thank the VolkswagenStiftung and the Hanse-Institute of Advanced Study, especially its rector Reto Weiler who all supported the interdisciplinary endeavor, not only financially, but also intellectually. The peer-review process was managed at the University of Bochum with important organizational support by Robert Schütze. Finally, we would like to thank the management of the journal, Ann Barajas, and the main editor of the journal, William Banks, for helpful comments and encouraging support. Albert Newen Universität Bochum, Institut für Philosophie, Universitätsstr. 150, 44892 Bochum, Germany E-mail address:
[email protected] Kai Vogeley University Hospital Cologne, Department of Psychiatry, Kerpener Str. 62, 50924 Cologne, Germany Christoph Michel Universität Bochum, Institut für Philosophie, Universitätsstr. 150, 44892 Bochum, Germany Available online 8 August 2010
Consciousness and Cognition 19 (2010) 690–701
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Visuospatial perspective taking in a dynamic environment: Perceiving moving objects from a first-person-perspective induces a disposition to act q H. Kockler a,*, L. Scheef b, R. Tepest a, N. David c, B.H. Bewernick d, A. Newen e, H.H. Schild b, M. May f, K. Vogeley a a
Department of Psychiatry, University of Cologne, Germany Department of Radiology, University of Bonn, Germany c Department of Neurophysiology and Pathophysiology, University Medical Center Hamburg-Eppendorf, Germany d Department of Psychiatry, University of Bonn, Germany e Department of Philosophy, Ruhr University Bochum, Germany f Faculty of Mind and Social Sciences, Helmut-Schmidt-University, Hamburg, Germany b
a r t i c l e
i n f o
Article history: Received 21 December 2009 Available online 2 April 2010 Keywords: Perspective taking First-person-perspective Virtual reality Functional magnetic resonance imaging (fMRI) Time-to-collision (TTC) Embodiment Enactment Intraparietal sulcus (IPS) Motion Perception Action
a b s t r a c t Spatial perspective taking is an everyday cognitive process that is involved in predicting the outcome of goal directed behavior. We used dynamic virtual stimuli and fMRI to investigate at the neural level whether motion perception interacts with spatial perspective taking in a life-like design. Subjects were asked to perform right-left-decisions about the position of either a motionless, hovering (STATic) or a flying ball (DYNamic), either from their own (1PP) or from the perspective of a virtual character (avatar, 3PP). Our results showed a significant interaction of STIMULUS TYPE and PERSPECTIVE with significantly increased activation in right posterior intraparietal sulcus (IPS) for 1PPDYN condition. As the IPS is critically involved in the computation of object-directed action preparation, we suppose that the simple perception of potentially action-relevant dynamic objects induces a ‘readiness for (re)action’, restricted to the 1PP. Results are discussed against the background of current theories on embodiment and enactive perception. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Integrating internal models of the outer world and of one’s own organism for the purpose of orientation and survival in the world is the central function of human self-consciousness and self-understanding. In the framework of Damasio this corresponds to the ‘core consciousness’ the scope of which is the ‘here and now’ (Damasio, 1999, p. 16) and the so-called ‘core self’ as a ‘transient entity, ceaselessly re-created for each and every object with which the brain interacts’ (Damasio, 1999, p. 17). One essential component of self-consciousness is a first-person-perspective, which refers to the experience of one’s own subjective multimodal experiential space centered around one’s own body. The ability of perspective taking enables us not only to live in a coherent self-centered world, but also to simulate how other people experience their environment and to predict their behavior by taking their perspective (Vogeley & Fink, 2003). Neurological and psychiatric syndromes demonstrate that perspective taking as body-centered representation of the self and the world is indispensable. Neglect as disturbance of the self-centered spatial representation of the world and/or the own person leads to a distortion of the q
This article is part of a special issue of this journal on Self, Other and Memory. * Corresponding author at: Department of Psychiatry, University of Cologne, Kerpener St. 62, 50924 Cologne, Germany. E-mail address:
[email protected] (H. Kockler).
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.03.003
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
691
egocentric reference frame so that – in cases of right-sided lesions – the left side of the body and/or the surrounding world is ignored (Fink et al., 2003; Halligan, Fink, Marshall, & Vallar, 2003). Deficits in taking a mental third-person-perspective (theory of mind) are assumed to be associated with autism (Baron-Cohen, 1995; Frith, 2001) and schizophrenia (Doody, Gotz, Johnstone, Frith, & Owens, 1998; Frith & Corcoran, 1996; Vogeley, Kurthen, Falkai, & Maier, 1999). However, to what extent spatial and mental perspective taking rely on similar or different neural mechanisms is currently still a matter of debate (see e.g. Aichhorn, Perner, Kronbichler, Staffen, & Ladurner, 2006; Uddin, Iacoboni, Lange, & Keenan, 2007). Several functional neuroimaging studies have focused on perspective taking in space and were recently reviewed by Zacks and Michelon (2005). In one of their own studies, Zacks and colleagues asked healthy subjects to make right-leftjudgements about line drawings of human bodies (front- or back-facing, upright or inverted) either from their own, or from the figure’s perspective. They found differential activation of the medial prefrontal cortex and cingulate gyrus during 1PP and differential activation of the left parietal–temporal–occipital junction during 3PP (Zacks, Rypma, Gabrieli, Tversky, & Glover, 1999). Vogeley et al. (2004) investigated the neural correlates of spatial perspective taking in a 3D-visuospatial task in which a virtual scene consisting of a virtual character surrounded by red balls was presented. Subjects were asked to assess the number of objects as seen either from their own or from the avatar’s perspective. Results showed a predominantly right sided superior parietal activation during 3PP as opposed to 1PP, while bilateral anterior medial prefrontal, posterior cingulate and superior temporal cortices were associated with 1PP as compared to 3PP. With the exception of a recent study of our own group (David et al., 2006), all fMRI-studies with animated stimulus material examining spatial perspective taking so far employed static stimuli. For the purpose of a systematic study of the influence of real life-like dynamic stimuli on spatial perspective taking, we conducted a study with a two-factorial design in which we systematically varied the factors STIMULUS TYPE (DYN (dynamic stimuli) versus STAT (static stimuli)) and PERSPECTIVE (1PP (first-person-perspective) versus 3PP (third-person-perspective)). Stimulus material showed a virtual character (avatar) in a virtual 3D-scenery. Subjects were asked to decide whether a static, hovering ball was located to the left or right side of their own (1PPSTAT) or the avatar’s visual field (3PPSTAT) or – in the case of dynamic stimuli – whether a flying ball passed one´s own (1PPDYN) or the avatar´s left or right side (3PPDYN). This design allowed us to investigate the influence of a dynamic environment on perspective taking as interaction of the factors STIMULUS TYPE and PERSPECTIVE. 2. Methods 2.1. Subjects Eighteen right-handed, healthy male volunteers (age 24.4 ± 2.1 years) without neurological or psychiatric illness, difficulties in handedness-judgements or disturbed vision participated in the study. Informed consent was obtained before participation. Participants were naïve with respect to the experimental task as well as the purpose of the study. Before taking part in the experiment, subjects were instructed and trained via a standardized slide show. The study was approved by the ethics committee of the Medical Faculty of the University of Bonn. 2.2. Stimulus material Fully controlled three-dimensional virtual scenes were generated using the software packages 3D Max (Version 4.0, Discreet, Division of Autodesk, Montreal, Canada) and Poser (Poser5, Curious Labs Inc., Santa Cruz, CA). The basic scene consisted of a quadrangular room with a male avatar positioned in the center of the room. A blue ball was either hovering motionless in the room or flying through the scenery. There were no additional visual cues in the scene like windows, doors, or furniture. The camera view was directed from the frontal aspect of the room onto the opposite wall. This basic scenery was systematically varied with respect to: (1) the orientation of the avatar, (2) the position of the ball in STAT and (3) the ball’s trajectory in DYN. (1) The orientation of the avatar was varied in a clockwise fashion around its vertical axis at angular distances of 30° so that 12 different positions were created. Avatar positions were counterbalanced throughout the whole experiment. Head and body orientation and gaze direction were always congruent. (2) In static stimuli (pictures) the ball was posted in the room at 24 equidistant positions on a circle around the avatar’s position. Only those ball positions inside the avatar’s field of view (covering 150°) were allowed. To avoid ambiguous scenes with respect to 3PP conditions, midline positions and positions at 15° besides midline were excluded. These rules gave rise to 96 different static stimuli. (3) In dynamic stimuli (video animations), ball trajectories were created by defining 12 different points of departure and arrival with a spacing of 30° equidistantly on a circle around the avatar’s position. The ball always flew directly from one of these points to another with a spacing of at least 120° between the point of departure and the point of arrival. A spacing of 180° was excluded to avoid collisions of ball and avatar. From all possible combinations of avatar orientations and ball trajectories, we chose 96, which allowed unambiguous solutions of the left–right task from both perspectives as pilot data showed. Each trajectory was used in both directions with respect to the subject’s position (towards and away from the observer). Both static and dynamic stimuli systematically varied presentation in both hemifields. The height of the flying ball was on the level of the avatar’s head, like in static stimuli, and its velocity was approximately constant within a given video sequence. In order to standardize the length of video sequences, the velocity varied between ±30% between videos due to length differences of the trajectories. To minimize interstimulus differences in velocity the longest and the shortest possible trajectories were not included. Stimuli
692
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
with trajectories that passed the camera too close were excluded as well to avoid a startle response or emotional arousal. In total, 192 stimuli (half of them static, half dynamic) were created. Each stimulus was presented twice per subject. All stimuli had a spatial resolution of 200 400 pixels. The video sequences were shown with a rate of 50 frames per second. Stimulus examples are shown in Fig. 1. 2.3. Tasks and study design We constructed the study in a two-factorial design with the factors PERSPECTIVE and STIMULUS TYPE with four different conditions, namely 1PPSTAT, 1PPDYN, 3PPSTAT, 3PPDYN. Stimuli were presented by Presentation (Version 0.76, Neurobehavioral Systems, Albany, CA) in a block design fashion. For this purpose, a series of 12 stimuli per block was shown on a black background and projected into the scanner environment via a mirror. Before each block of pictures subjects either read the instruction ‘Is the ball on the right or on the left of the AVATAR?’ (3PPSTAT) or ‘Is the ball on the right or on the left side of YOU?’ (1PPSTAT). Before each block of videos either ‘Does the ball pass the AVATAR on his right or on his left side?’ (3PPDYN) or ‘Does the ball pass YOU on your right or on your left side?’ (1PPDYN) was shown. Four experimental runs were presented consisting of 8 blocks with 12 stimuli each in randomized order, systematically varying the four conditions. Each block corresponded to approximately 11 scans per block at a TR of 3.0 s. In advance of each block, a black screen (18 s) and the instruction for the following block (6 s) was shown as a low-level-baseline condition (rest). The first seven scans were discarded later to allow for T1 saturation effects. Each stimulus was shown for approximately 2.7 s. The duration of the stimulus pictures was timed exactly, while the video sequences were presented with some duration uncertainty because of technical limits of the Presentation software. Because of these video timing uncertainties, which accumulated over time, runs differed in their length and lasted between 155 and 159 scans. To obtain an equal length for analyses, the last baseline of each run was cut off after TR 154. The overall duration of the scanning session was approximately 34 min. As dependent variables, reaction times and errors (correctness scores) were recorded using an fMRI compatible response device (Lumitouch, Lightwave Medical Industries, CST Coldswitch Technologies, Richmond, CA). Answers were given by index and ring fingers of the right or left hand in alternating order to avoid differential hemispheric activation due to finger movement. Independent of the answering hand, the left of both fingers indicated the answer ‘left’ and the right finger indicated the answer ‘right’ for the convenience of the test person. 2.4. Statistical analyses of behavioral data Statistical analyses of the behavioral data were performed using SPSS for Windows (Version 11.0). Dependent variables were reaction times (RT) and correctness scores (CS) as percentage of correct answers given within a maximum timeframe (stimulus length). Median values were calculated over each block per subject and then subsequently mean-averaged for each condition across subjects. A repeated-measures ANOVA with the main factors PERSPECTIVE and STIMULUS TYPE as withinsubject factors was performed. The significance level for all analyses was set at p < .05 (two-tailed). 2.5. Functional magnetic resonance imaging Functional MR imaging was performed employing a 3.0 T whole-body imager (Intera; Philips Medical Systems, Best, Netherlands), equipped with a quadrature head coil. Functional images were obtained using a single-shot gradient echo,
Fig. 1. Experimental design. Two-factorial design with PERSPECTIVE (third-person-perspective, 3PP, versus first-person-perspective, 1PP) and STIMULUS TYPE (static stimuli, STAT, versus dynamic stimuli, DYN) as main factors. Instructions and stimuli examples are shown. The static stimulus example is shown as used in the experiment, the dynamic stimulus example is a photomontage of single frames extracted from one video sequence used in the study. Stimuli originally had been colored.
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
693
echoplanar imaging (EPI) sequence using the following imaging parameters: TR = 3000 ms, TE = 35 ms, flip angle = 90°, FOV 230 230 mm2, voxel size 3.6 3.6 3.6 mm3. The optimal echo time was chosen to match the T2* of gray matter to ensure optimal echo time, compensating for the inherently shorter T2* relaxation times at 3.0 T (Clare, Francis, Morris, & Bowtell, 2001). Each session contained 170 functional images. To aid in localization of activation, a high-resolution T1-weighted magnetization-prepared rapid gradient-echo imaging (MP-RAGE) as 3D MRI image was acquired (TR: 7.7 ms, TE: 3.93 ms, flip angle = 15°, TFE-shot interval 2200 ms, FOV 256 256 mm2, matrix 256 256, voxel size 1 1 1 mm3). 2.6. Image processing and analysis Functional images were preprocessed and statistically analyzed using SPM2 (Statistical Parametric Mapping Software, Wellcome Department of Imaging Neuroscience, London, UK) implemented in MATLAB 6.5 (Mathworks Inc., Sherborn, MA, USA). The preprocessing comprised eye masking in order to avoid artefacts, reorientation according to AC–PC-line, realignment and unwarping to correct for head movement, co-registration of functional and anatomical images, normalization to the MNI stereotactic space, and smoothing with a 10 10 10 mm3 Gaussian kernel. For statistical analyses, the general linear model and a boxcar waveform convolved with the hemodynamic response function were used. Subject-specific, low frequency drifts in signal changes were removed by a high pass filter. Global signal changes were treated as a covariate of no interest. Specific effects were tested by applying appropriate linear contrasts to the parameter estimates for each condition resulting in a t-statistic for each voxel. These t-statistics constitute a statistical parametric map (SPM). The threshold was set at p < .05 throughout the whole study (family-wise error (FWE) corrected) and an extent threshold of 20 contiguous voxels. The following contrast images were calculated for every subject: main effect of 1PP [(1PPDYN plus 1PPSTAT) relative to (3PPDYN plus 3PPSTAT)], main effect of 3PP [(3PPDYN plus 3PPSTAT) relative to (1PPDYN plus 1PPSTAT)], main effect of DYN [(1PPDYN plus 3PPDYN) relative to (1PPSTAT plus 3PPSTAT)], main effect of STAT [(1PPSTAT plus 3PPSTAT) relative to (1PPDYN plus 3PPDYN)]. Additionally, data were analyzed for interactions by calculating the contrasts for the interactions of PERSPECTIVE and STIMULUS TYPE as interaction 1 (IA1) employing the formula [(1PPDYN relative to 1PPSTAT) relative to (3PPDYN relative to 3PPSTAT)] and interaction 2 (IA2) employing the formula [(3PPDYN relative to 3PPSTAT) relative to (1PPDYN relative to 1PPSTAT)]. To determine the main effects of the factors PERSPECTIVE and STIMULUS TYPE and their interactions on the group level, corresponding contrast images were calculated and analyzed as described above applying a random effects model (Friston et al., 1995). Stereotactic MNI coordinates of the voxels of local maximal activation were determined within regions of significant relative signal change associated with the different conditions. The anatomic localization of local maxima and other activated voxels was transformed to Talairach coordinates and assessed by reference to the standard stereotactic atlas (Talairach & Tournoux, 1988). Furthermore, approximations of the corresponding Brodmann areas (BA) are provided on the basis of this atlas. However, it should be kept in mind that Talairach’s and Tournoux’ version of the Brodmann map does not reflect the complete and up-to-date cytoarchitectonic organization of the human brain. Nevertheless Brodmann areas are given for the reader’s convenience and for comparability to other studies. Figs. 2 and 3 provide the SPM {t}-maps of the significant main effects on the group level as overlay images onto one normalized 3D data set (p < .05, FWE corrected, extent threshold = 20 voxels).
Fig. 2. Neural correlate of IA1. Significantly activated clusters are superimposed on a single, normalized T1 data (p < .05, FWE corrected, extent threshold = 20 voxels). The statistical interaction IA1 of the main factors PERSPECTIVE and STIMULUS TYPE, mainly reflecting the differential effect of 1PPDYN > 1PPSTAT, is associated with increased neural activity in the right posterior intraparietal sulcus (IPS). Data in Figs. 2 and 3 correspond to those in Table 2. Fig. 2 is shown in neurological orientation (right = right, left = left in coronal sections).
694
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
Fig. 3. Neural correlates of 3PP > 1PP (a), and DYN > STAT (b). Significantly activated clusters are superimposed on a single, normalized T1 data (p < .05, FWE corrected, extent threshold = 20 voxels). Data in Figs. 2 and 3 correspond to those in Table 2. (a) Taking a third-person-perspective (3PP > 1PP) is associated with increased neural activity in the pre-SMA bilaterally, the posterior parietal cortex bilaterally with predominance of the right hemisphere, inferior occipital cortex (EBA), cerebellum, subregions of the dorsolateral prefrontal cortex and the anterior insula bilaterally. (b) During the dynamic condition (DYN > STAT), differential activations in occipitotemporal (V5/MT+) cortex, superior frontal areas and superior and medial parietal (precuneus) cortex bilaterally can be observed.
In order to characterize the differential influences of the different conditions in IA1 more precisely, the areas activated in IA1 were defined as regions-of-interest (ROI) and additional analyses were performed in these ROIs: 1PPDYN > 1PPSTAT; 1PPSTAT > 1PPDYN; 3PPDYN > 3PPSTAT; 3PPSTAT > 3PPDYN. For ROI-analyses, the WKU PickAtlas (www.fmri.wfubmc.edu/download.htm) and Marsbar (www.marsbar.sourceforge.net) were used. Furthermore, the BOLD signal changes at the principally activated voxels of the significant key contrasts were taken from raw data of every participant and their means were calculated for each condition as well as their standard errors (Fig. 4). 3. Results 3.1. Behavioral data Overall means and standard deviations (SD) of the dependent variables are reported in Table 1. Both the ANOVA on reaction times (RT) and the ANOVA on correctness scores (CS) revealed a significant effect of PERSPECTIVE (for RT:
Fig. 4. Mean percent BOLD signal changes compared to baseline for IA1 (a), 3PP (b) and DYN (c) at the respective principally activated voxel. The signal changes at these voxels were taken from raw data and plotted over time. Means were calculated for each condition and show the differential increase in signal over the four conditions. The mean standard errors are given as error bars. Voxel coordinates are provided according to Talairach and Tournoux (1988). (a) Principally activated voxel in IA1: x = 14, y = 74, z = 56 (MNI coordinates). As the figure shows, 3PPDYN (entering negatively into the contrast) and 3PPSTAT (entering positively into IA1) approximately neutralize their influence in contrast IA1 ((3PPSTAT – 3PPDYN) plus (1PPDYN – 1PPSTAT)). Therefore, IA1 is mainly due to the difference of 1PPDYN versus 1PPSTAT. Please note that the relatively high activation of the IPS in 3PP conditions reflects the mental rotation operation that is necessary to solve the task (see Section 4 of main effects). (b) Principally activated voxel in 3PP: x = 4, y = 20, z = 46. (c) Principally activated voxel in DYN: x = 46, y = 64, z = 4.
Table 1 Behavioral data. Mean values and standard deviations (SD) of the dependent variables (reaction times and correctness scores) are summarized for the main conditions. 3PPDYN
Reaction time (msec) Correctness score (%)
3PPSTAT
1PPDYN
1PPSTAT
Mean
SD
Mean
SD
Mean
SD
Mean
SD
1565.3 95.0
141.3 5.6
989.2 87.9
173.3 5.3
1230.2 98.3
128.5 1.7
552.7 99.7
118.6 0.6
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
695
F(1,17) = 205.19; p < .001; for CS: F(1,17) = 50.73; p < .001) and STIMULUS TYPE (for RT: F(1,17) = 828.04; p < .001; for CS: F(1,17) = 19.28; p < .001) as well as a significant interaction PERSPECTIVE STIMULUS TYPE (for RT: F(1,17) = 18.36; p 6 .001; for CS: F(1,17) = 32.61; p < .001). Whereas the difference of RT between DYN and STAT was due to the latency with which objects in DYN had to pass until an adequate decision could be made, the RT difference in PERSPECTIVE most likely reflected task difficulty. Correctness scores higher than 75% assured correct task solving and allowed adequate interpretation of fMRI data. In the 3PPDYN-condition, reaction times (means of individual medians ± standard deviation) show a dependence on the degree of rotation of the avatar (Fig. 5). 3.2. Neural correlates of perspective, stimulus type and their interaction Differential neural activity as correlate of IA1 [(1PPDYN relative to 1PPSTAT) relative to (3PPDYN relative to 3PPSTAT)] was located in the right posterior intraparietal sulcus (IPS) and related superior parietal areas (Table 2a and Fig. 2). The neural correlate of the main effect of 3PP was associated with increased neural activity in the pre-supplementary motor area (pre-SMA) bilaterally, the posterior parietal cortex bilaterally predominantly on the right side, cerebellum, subregions of the dorsolateral prefrontal cortex bilaterally and the anterior insula bilaterally (Table 2b and Fig. 3a). Areas of significantly increased neural activity associated with the main effect of DYN were located in the occipitotemporal (approximately V5/ MT+), superior frontal and superior/medial parietal cortex (precuneus) bilaterally (Table 2c and Fig. 3b). With regard to the second interaction IA2 as well as to the main effects for 1PP and STAT, no differential neural activity was found. These findings are further illustrated by plots of the mean BOLD signal changes (Fig. 4). It can be demonstrated that for the main effects 3PP and DYN, the mean BOLD signal changes at the principally activated voxel (3PP: 4, 20, 46; DYN: 46, 64, 4) correspond to the expected values, as each of the relevant conditions that contributes to a particular main effect shows an adequate and comparable increase in signal change (Fig. 4b and c). The significant BOLD increase in contrast IA1 [(1PPDYN relative to 1PPSTAT) relative to (3PPDYN relative 3PPSTAT)] at the principally activated voxel (14, 74, 56) cannot be related to any relevant difference between 3PPSTAT and 3PPDYN and is thus due to the difference between 1PPDYN and 1PPSTAT (Fig. 4a). This effect of the interaction IA1 was further corroborated by subsequent region-of-interest (ROI) analyses as the ROI corresponding to the clusters activated during IA1 were also significantly activated in 1PPDYN > 1PPSTAT, but neither in 3PPDYN > 3PPSTAT nor in 3PPSTAT > 3PPDYN (p < .05; FWE corrected; not illustrated). These confirmatory analyses allow to conclude that the neural correlate of IA1 is essentially constituted by the differential effect of 1PPDYN relative to 1PPSTAT although the overall activation of the IPS during 3PP-conditions compared to baseline is at least as high as during 1PPDYN. 4. Discussion The main experimental question of this study was to examine the influence of life-like dynamic stimuli on spatial perspective taking as a necessary prerequisite of human self-consciousness and the ‘core self’ (Damasio, 1999). Behavioral data
Fig. 5. Reaction times of 3PPDYN-condition, dependent on the position of the avatar. Reaction times show the typical pattern of a mental rotation operation.
696
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
Table 2 Significant clusters and their peak coordinates for the neural correlates of IA1, 3PP, and DYN. Volume statistics of the main effects and the interaction between the experimental factors (p < .05, FWE corrected, extent threshold = 20 voxels). Voxel coordinates are provided according to Talairach and Tournoux (1988). BA = Brodmann’s area, L = left hemisphere, R = right hemisphere. Region (a) IA1 Superior parietal R IPS R
a b c
Cluster size (voxels)
MNI-coordinates (x, y, z)
t-value
BA
28 39
14 34
74 82
56 30
8.53 8.20
7
(b) 3PP (3PP > 1PP) Medial frontal R/La Superior parietal R Cerebellum L Inferior occipital Rb Inferior/middle front R Superior frontal L IPS/superior parietal L Cerebellum L Midbrain Anterior insula R Superior frontal R Anterior insula L Inferior/middle front L Middle occipital L
271 1596 416 891 543 238 1080 587 452 150 312 97 94 57
4 40 36 44 50 26 36 12 2 34 28 32 42 46
20 52 64 66 12 4 38 78 34 28 2 22 4 68
46 62 28 10 30 58 40 28 4 6 48 2 26 4
10.44 10.43 10.16 9.72 9.38 9.28 9.27 8.76 8.41 8.29 8.15 8.09 7.64 7.37
8 40
(c) DYN (DYN > STAT) Middle temporal Lc Middle temporal Rc/superior occipital Rb Superior/middle front L Superior parietal L/R/precuneus L/R Middle frontal R Postcentral L
646 883 1033 2259 272 34
46 48 26 34 28 60
64 66 4 42 0 18
4 4 56 60 62 36
10.11 9.91 9.12 9.05 7.13 7.04
37/19 44 6 40
47 6 47 9 37
37 37 6 6 43
Corresponding to pre-SMA. Corresponding to extrastriate body area (EBA). Corresponding to V5/MT+.
as well as extensive statistical analyses of fMRI data revealed a significant interaction between the experimental factors PERSPECTIVE and STIMULUS TYPE with differential activation of the right posterior intraparietal sulcus (IPS) due to an increase in neural activation during 1PPDYN relative to 1PPSTAT. 4.1. The interaction effect While it is not surprising that the right IPS is highly activated in 3PP-states due to its well documented involvement in mental rotation tasks (see Section 4.2), the additional IPS-activation in 1PPDYN compared to 1PPSTAT is a new and unexpected finding. The IPS until now has neither been reported as a primary motion-processing area nor as a neural correlate of a first-person-perspective as such (Vogeley et al., 2004). Instead, the majority of studies on the IPS demonstrated a role in visually guided actions that are directed to objects (Goodale & Milner, 1992; Rizzolatti & Matelli, 2003). Anatomy and function of the IPS has been extensively studied in macaques (for reviews see Colby and Goldberg (1999) and Grefkes and Fink (2005)), these studies provide evidence for a central role in action organization, such as reaching and grasping and eye-armcoordination (Battaglia-Mayer, Caminiti, Lacquaniti, & Zago, 2003; Burnod et al., 1999; Gregoriou & Savaki, 2001). During the last years, a number of studies in humans demonstrated that the IPS of humans and macaques share a similar functional organization (Astafiev et al., 2003; Grefkes & Fink, 2005; Orban et al., 2006; Shikata et al., 2007). Furthermore, there is now converging evidence on the IPS’ function that can be inferred from clinical data, virtual lesion studies (transcranial magnetic stimulation, TMS) and functional neuroimaging. Lesions of the right IPS and adjacent superior parietal cortex result in a neurological syndrome called ‘optic ataxia’: patients show an impairment in the visual control of target-directed arm movements (Battaglia-Mayer & Caminiti, 2002; Perenin & Vighetto, 1988; Weiss et al., 2006). Functional MRI studies in humans underlined this clinical finding: The IPS shows increased BOLD signal during visually guided, objectdirected action, especially during reaching (Kertzman, Schwarz, Zeffiro, & Hallett, 1997; Levy, Schluppeck, Heeger, & Glimcher, 2007), grasping (Begliomini, Wall, Smith, & Castiello, 2007; Frey, Vinton, Norlund, & Grafton, 2005), attending and pointing to a (peripheral) visual location (Astafiev et al., 2003). Strongest evidence is provided by studies using transcranial magnetic stimulation (TMS). Transient lesions of the left posterior IPS impaired the adjustment of reaching movements to a moving environment (Della-Maggiore, Malfait, Ostry, & Paus, 2004), whereas TMS-induced transient lesions to the anterior part of the IPS caused a disturbance of the on-line control of reach-to-grasp movements (Tunik, Frey, & Grafton, 2005). Anatomically, extensive input from visual areas such as the temporo-parietal motion-responsive brain region V5/MT+ and projections from the IPS to ventral prefrontal and primary motor areas (Grefkes & Fink, 2005; Sakata, Taira, Kusunoki,
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
697
Murata, & Tanaka, 1997) further support that the IPS is crucial for the integration and coordination of external peripersonal visual information on the one hand and internal representations of action schemata and reference frames on the other hand which are both essential components of object-directed action. Andersen and Buneo (Andersen, Snyder, Bradley, & Xing, 1997; Buneo & Andersen, 2006) as well as Rizzolatti and Matelli (2003) proposed that one of the major functions of the IPS and IPS-frontal circuits is the representation of object positions in an effector-centered, thus egocentric reference frame in order to allow for motor descriptions of stimulus positions. In humans, it could be shown that the visuomotor coordinate transformation subserving goal-directed hand movements activates the superior parietal cortex with a maximum in the medial wall of the IPS (Andersen et al., 1997; Grefkes, Ritzl, Zilles, & Fink, 2004). In short, the function of the IPS (as part of the dorsal stream of vision) is concisely described by the term ‘vision for action’ coined by Goodale (1998). The present finding that the IPS is specifically involved during 1PPDYN importantly extends the previous evidence of its involvement in a variety of object-directed action-processing operations and enriches the concepts on its functions by suggesting that already the mere visual perception of dynamic objects from a first-person-perspective leads to its recruitment. We propose that this reflects an automatic, prereflexive disposition to act or – in our context – an early´ readiness to react´ that is independent of any conscious intention to act. This ‘readiness for (re)action’ is activated even if an action is neither explicitly required nor actively performed (as in our study) and best understood as an unconscious state of ‘being prepared’ to react to the object if it should become necessary. This is plausible from an evolutionary point of view: Moving objects are more action-relevant than static ones due to their potential live-threatening character and thus can require very fast modifications of behavior. Our own finding is underlined by direct evidence from monkey studies that showed that an IPS subregion, the ventral intraparietal area (VIP), contributes to collision avoidance and defensive behavior (Graziano & Cooke, 2006). In a similar vein, Cooke, Taylor, Moore, and Graziano (2003) evoked complex defensive facial expressions by microstimulation of the VIP in the macaque. In direct relevance to the present study, Field and Wann (2005) recently investigated the neural correlates of a so-called ‘time-to-collision (TTC)’ mechanism in humans. TTC refers to the estimation of the remaining time until an approaching object reaches a target (here: the participating subjects of the study). During a collisional condition compared with control conditions of similar visual quality that did not induce the impression of an object moving in depth, increased neural activity in sensorimotor areas were observed that are target of the dorsal visual system. Similar to our study, this activation could not be attributed to any actual movement or any intention to act, supporting the idea of an automatically activated disposition to react induced by the mere visual perception of a potentially threatening dynamic object. Thus, subjects are prepared for fast modifications of behavior that are required in dynamic situations, e.g. defensive responses in order to avoid collisions with objects. Interestingly, recent studies provide evidence that the ability to detect and predict a collision seems to be a function of perceptuomotor brain development and to depend on perceptuomotor experiences over lifetime (Lobjois, Benguigui, Bertsch, & Broderick, 2008; Van der Weel & van der Meer, 2009). Extending these previous results, our findings implicate that the ‘readiness for (re)action’ is stronger in 1PP-states than in 3PP-states: If the ‘readiness for reaction’ would be independent of perspective, we should expect a higher IPS-activation in 3PPDYN than in 3PPSTAT corresponding to the higher activation in 1PPDYN than in 1PPSTAT. This is not the case. We can therefore presume that the disposition to act is restricted to 1PP-states which is intuitively plausible because one’s own disposition to act can have no functional role for someone else. An anticipatory coding of objects in terms of motor descriptions is only ecologically useful if it implies a behavioral advantage. This advantage of a readiness for actions does not exist in a third-person-perspective because we simply cannot act for another person. So, the restriction to one’s own perspective as well shows that the disposition to act depends on an action-relevant context – an inference that is further corroborated by studies of the neural computation of near versus far space. Processing of objects that are located in ‘near space’ defined as the peripersonal space within arm’s reach differentially activate the dorsal stream of vision including the IPS while processing of objects that are presented in far space show a stronger activation of the ventral stream of vision (Weiss, Marshall, Zilles, & Fink, 2003). Correspondingly, in monkeys, extrapersonal space within reaching distance is represented by two subpopulations of neurons within the IPS, the medial intraparietal area (MIP, ‘near space’) and the ventral intraparietal area (VIP, ‘ultra-near space’), respectively (Colby & Duhamel, 1991). These findings as well as our results may lead to the hypothesis that the visual perception of objects in events in general recruits action-relevant areas as soon as the perception of the object becomes action-relevant, in the sense of enabling a disposition to act in the immediate body-related environment. This is an observation of theoretical importance for current concepts on ‘embodiment’ and ‘enactive perception’: first of all, the IPSactivation in our experiment is indicative of what is meant by the ‘embodiment’ of the mind (Gallagher, 2005): in the perception of events certain features of objects can be strongly associated with motor activations in the brain reflecting a readiness to act on a behavioral level. Secondly, however, from these ‘enactive’ perceptual states, we can distinguish other perceptual states that are not – or at least significantly less – associated with this disposition to act. In our experiment, the question if the perception of an event leads to this neural response in cortical areas associated with action preparation or not critically depends on the subject’s perspective while the visual input is the same from both perspectives. Our finding thus leads us to the final assumption that the degree of enactment in the perception of events depends on the context, more precisely on the question if an event is action-relevant for a subject of not. Thus, the general claim that the perception of objects is equivalent to the preparation of object-directed actions (O’Regan & Noë, 2001; Noë, 2004) must be rejected. Reflecting critically our study, there are also alternative options to interpret our findings. Unfortunately, we could not obtain measurements of eye movements. As the posterior IPS is known to be involved in the generation of saccades we cannot completely rule out that the effect could be confounded by task dependent eye movements that could have been processed in the IPS. However, this appears unplausible to us based on the following considerations. First, eye movement related
698
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
parietal areas of each hemisphere topographically respond to targets for pro- and antisaccades in the contralateral hemifield (Medendorp, Goltz, & Vilis, 2005; Sereno, Pitzalis, & Martinez, 2001). In our study, each stimulus, both dynamic and static, was shown twice, the second instance right–left-mirrored compared to the first instance to avoid laterality effects. If the observed activity was due to saccadic eye movements, one had to assume a bilateral activation and the asymmetric, isolated activity in right parietal regions during 1PPDYN could not be explained. Second, if different eye movements during conditions would cause the intraparietal activation in our interaction, we would at least expect an additional activation of the frontal eye fields (FEF) because (at least in monkeys) the frontal eye fields as motor areas are even more strongly associated with the execution of eye movements than the parietal eye regions (Kusunoki et al., 2000). Third, in the experiment that served as basis for the current study and that employed similar stimulus material, Vogeley et al. (2004) showed that there were no differences in eye movements between 3PP and 1PP. Attentional effects can also be discussed as potential confounding mechanisms that recruit the IPS. Culham et al. (1998) showed in an fMRI study that attentive tracking of moving objects (versus passive viewing) without eye movement leads to increased activity in different regions including the IPS. In our opinion, such an effect of covert attention would not reject our interpretation of a ‘readiness for (re)action’ as attentive tracking of moving objects could also be understood as the constitutive cognitive mechanism of this disposition to act that is induced by moving objects: The higher the relevance of these objects is, the more attention will be allocated towards them. In this sense, recent electrophysiological studies in monkeys raised the idea that the role of the IPS in spatial attention consists in providing spatial ‘salience maps’ of objects by both topdown and bottom-up modulated coding of behaviorally relevant objects in terms of spatial coordinates of attention targets (see Goldberg, Bisley, Powell, and Gottlieb (2006), for a review). Recapitulating, Rizzolatti and Matelli (2003) precisely described the function that we suppose to be reflected by the neural correlate of the interaction in our experiment: ‘‘When a visual stimulus is presented in the peripersonal space, it evokes automatically a ‘potential motor action’, which, regardless of whether the action is subsequently executed or not, maps the spatial stimulus position in motor terms” (Rizzolatti & Matelli, 2003; similar: Culham & Valyear, 2006). For the fist time, our results provide evidence that potential motor actions are triggered by dynamic stimuli more than by static stimuli and probably are restricted to 1PP-states. These findings possibly reflect a more general mechanism that induces an early and prereflective disposition to act – neurally represented by the recruitment of action-relevant areas – in the presence of objects with an action-relevant character. 4.2. Main effect of perspective: 3PP The differential activation during the main effect of 3PP was mainly located in the predominantly right-sided posterior parietal cortex (PPC), the pre-supplementary motor area (pre-SMA) bilaterally and dorsolateral prefrontal areas. These findings in essence replicate the results of Vogeley and colleagues (2004) and show exceeding overlaps with the pattern that was found in a recent meta-analysis of 32 mental rotation studies (Zacks, 2008), suggesting that the process of taking the spatial perspective of another person is more closely related to mental rotation operations than to mental perspective taking tasks (theory of mind) on a neural and cognitive level. This conclusion is also supported by reaction time patterns that show a dependence of 3PP-reaction times on the degree of rotation of the avatar that is typical for mental rotation studies (Fig. 5). Accordingly, Harris et al. (2000) found that the right superior parietal cortex centered on the IPS and extending down into the superior posterior occipital cortex shows a correlation between increase in activation and increase in the amount of mental rotation during parametric modulation of the degree of rotation. The high percent signal change in the right IPS during 3PP-conditions thus reflects the mental rotation computations that are necessary to solve the task. Interestingly – although not based on mental rotation operations – various studies also found the right (superior, inferior or medial) parietal cortex being involved in distinguishing self from other (e.g. Decety & Sommerville, 2003; Feinberg & Keenan, 2005; Feinberg & Roane, 2005; Ruby & Decety, 2004; Seger, Stone, & Keenan, 2004; Stuss & Anderson, 2004; Uddin, Molnar-Szakacs, Zaidel, & Iacoboni, 2006). Additionally to the regions involved in mental rotation operations reported by Zacks, we found one region in the right inferior occipital region. Although we did not use an appropriate localizer, we assume that this region corresponds to the extrastriate body are (EBA) which is involved in the perception of human bodies and body parts (David et al., 2007; Downing, Jiang, Shuman, & Kanwisher, 2001). Right EBA especially computes bodies that are shown from an allocentric perspective – a finding that has been interpreted as the neural correlate of perceiving others (Saxe, Jamal, & Powell, 2005). 4.3. Main effect of stimulus type: DYN Activation patterns during the main effect of DYN were also in accordance with previous evidence. For example, the principally activated region during DYN was the visual motion area V5/MT+ (Born & Bradley, 2005; Watson et al., 1993; Wilms et al., 2005). The precuneus, also found here, has been associated with visual imagery, supporting our capacity to – metaphorically speaking – ‘inspection of internal images’ (Cavanna & Trimble, 2006; Ogiso, Kobayashi, & Sugishita, 2000), or ‘the mind’s eye’ (Fletcher et al., 1995), and the spatial localization of objects (Rao, Zhou, Zhuo, Fan, & Chen, 2003). According to these results, one could speculate that the precuneus possibly performs an imagined visual anticipation of the ball’s motion by generating an estimated ‘internal image’ of the ball’s trajectory. A similar function has been ascribed to the lateral premotor cortex that is also activated during the dynamic condition. Mental anticipation of the location of moving objects
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
699
increased activation of the dorsolateral premotor cortex as one part of a so-called ‘prehension network’ associated with object-related attention (Schubotz & von Cramon, 2001). 4.4. Other main effects: 1PP and STAT With respect to the other main effects, neither 1PP nor STAT reached the level of statistical significance applied in the present study for all other conditions. While we did not expect a certain differential activation for STAT, we had strong hypotheses for 1PP. Similar experiments of our group contrasting 1PP versus 3PP consistently showed a characteristic pattern of increased activation comprising medial prefrontal, medial parietal, inferior parietal and temporoparietal cortex bilaterally and for the present study we expected a similar activation pattern, accordingly. However, all these studies used exclusively static stimuli. As our results show, there is a major difference in 1PP if stimuli are dynamic. We therefore assume that his dissociation of 1PP explains the missing statistical power of a common 1PP [(1PPDYN plus 1PPSTAT) relative to (3PPDYN plus 3PPSTAT)]. 5. Conclusion Studying the influence of dynamic stimuli on spatial perspective taking revealed a significant increase in neural activation during motion perception from a first-person-perspective in the right posterior intraparietal sulcus (IPS). Our results indicate that the neural key process in the perceptual situation examined is related to the computation of object-directed action preparation. They suggest that potentially threatening and thus action-relevant, moving objects induce a state of ‘oneself in readiness for (re)action’. Acknowledgment We thank Jeffrey Zacks for very helpful comments on a previous version of the manuscript. M. M., A. N. and K. V. were supported by the German VolkswagenFoundation. References Aichhorn, M., Perner, J., Kronbichler, M., Staffen, W., & Ladurner, G. (2006). Do visual perspective tasks need theory of mind? NeuroImage, 30(3), 1059–1068. Andersen, R. A., Snyder, L. H., Bradley, D. C., & Xing, J. (1997). Multimodal representation of space in the posterior parietal cortex and its use in planning movements. Annual Review of Neuroscience, 20, 303–330. Astafiev, S. V., Shulman, G. L., Stanley, C. M., Snyder, A. Z., Van Essen, D. C., & Corbetta, M. (2003). Functional organization of human intraparietal and frontal cortex for attending, looking, and pointing. Journal of Neuroscience, 23(11), 4689–4699. Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Boston: MIT Press/Bradford Books. Battaglia-Mayer, A., & Caminiti, R. (2002). Optic ataxia as a result of the breakdown of the global tuning fields of parietal neurones. Brain, 125(2), 225–237. Battaglia-Mayer, A., Caminiti, R., Lacquaniti, F., & Zago, M. (2003). Multiple levels of representation of reaching in the parieto-frontal network. Cerebral Cortex, 13(10), 1009–1022. Begliomini, C., Wall, M. B., Smith, A. T., & Castiello, U. (2007). Differential cortical activity for precision and whole-hand visually guided grasping in humans. European Journal of Neuroscience, 25(4), 1245–1252. Born, R. T., & Bradley, D. C. (2005). Structure and function of visual area MT. Annual Review of Neuroscience, 28, 157–189. Buneo, C. A., & Andersen, R. A. (2006). The posterior parietal cortex: Sensimotor interface for the planning and online control of visually guided movements. Neuropsychologia, 44(13), 2594–2606. Burnod, Y., Baraduc, P., Battaglia-Mayer, A., Guigon, E., Koechlin, E., Ferraina, S., et al (1999). Parieto-frontal coding of reaching: An integrated framework. Experimental Brain Research, 129(3), 325–346. Cavanna, A. E., & Trimble, M. R. (2006). The precuneus: A review of its functional anatomy and behavioral correlates. Brain, 129(3), 564–583. Clare, S., Francis, S., Morris, P. G., & Bowtell, R. (2001). Single-shot T2* measurement to establish optimum echo time for fMRI: Studies of the visual, motor, and auditory cortices at 3.0 T. Magnetic Resonance in Medicine, 45, 930–933. Colby, C. L., & Duhamel, J. R. (1991). Heterogeneity of extrastriate visual areas and multiple parietal areas in the macaque monkey. Neuropsychologia, 29(6), 517–537. Colby, C. L., & Goldberg, M. E. (1999). Space and attention in parietal cortex. Annual Review of Neuroscience, 22, 319–349. Cooke, D. F., Taylor, C. S., Moore, T., & Graziano, M. S. (2003). Complex movements evoked by microstimulation of the ventral intraparietal area. Proceedings of the National Academy of Sciences, USA, 100(10), 6163–6168. Culham, J. C., Brandt, S. A., Cavanagh, P., Kanwisher, N. G., Dale, A. M., & Tootell, R. B. (1998). Cortical fMRI activation produced by attentive tracking of moving targets. Journal of Neurophysiologie, 80(5), 2657–2670. Culham, J. C., & Valyear, K. F. (2006). Human parietal cortex in action. Current Opinion in Neurobiologie, 16(2), 205–212. David, N., Bewernick, B. H., Cohen, M. X., Newen, A., Lux, S., Fink, G. R., et al (2006). Neural representations of self versus other: Visual-spatial perspective taking and agency in a virtual ball-tossing game. Journal of Cognitive Neuroscience, 18(6), 898–910. David, N., Cohen, M. X., Newen, A., Bewernick, B. H., Shah, N. J., Fink, G. R., et al (2007). The extrastriate cortex distinguishes between the consequences of one’s own and others’ behavior. NeuroImage, 36(3), 1004–1014. Damasio, A. (1999). The feeling of what happens. Body and emotion in the making of consciousness. Harcourt, Inc. Decety, J., & Sommerville, J. A. (2003). Shared representations between self and other: A social cognitive neuroscience view. Trends in Cognitive Sciences, 7(12), 527–533. Della-Maggiore, V., Malfait, N., Ostry, D. J., & Paus, T. (2004). Stimulation of the posterior parietal cortex interferes with arm trajectory adjustments during the learning of new dynamics. Journal of Neuroscience, 24(44), 9971–9976. Doody, G. A., Gotz, M., Johnstone, E. C., Frith, C. D., & Owens, D. G. (1998). Theory of mind and psychoses. Psychological Medicine, 28(2), 397–405. Downing, P. E., Jiang, Y., Shuman, M., & Kanwisher, N. (2001). A cortical area selective for visual processing of the human body. Science, 293(5539), 2470–2473. Feinberg, T. E., & Keenan, J. P. (2005). Where in the brain is the self? Consciousness and Cognition, 14(4), 661–678. Feinberg, T. E., & Roane, D. M. (2005). Delusional misidentification. The Psychiatric Clinics of North America, 28(3), 665–683 [678–679].
700
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
Field, D. T., & Wann, J. P. (2005). Perceiving time to collision activates the sensorimotor cortex. Current Biology, 15(5), 453–458. Fink, G. R., Marshall, J. C., Weiss, P. H., Stephan, T., Grefkes, C., Shah, N. J., et al (2003). Performing allocentric visuospatial judgments with induced distortion of the egocentric reference frame: An fMRI study with clinical implications. NeuroImage, 20, 1505–1517. Fletcher, P. C., Frith, C. D., Baker, S. C., Shallice, T., Frackowiak, R. S., & Dolan, R. J. (1995). The mind’s eye – Precuneus activation in memory-related imagery. NeuroImage, 2(3), 195–200. Frey, S. H., Vinton, D., Norlund, R., & Grafton, S. T. (2005). Cortical topography of human anterior intraparietal cortex active during visually guided grasping. Brain Research. Cognitive Brain Research, 23(2–3), 397–405. Friston, K. J., Homes, A. P., Worsley, K. J., Poline, J. P., Frith, C. D., & Frackowiak, R. S. (1995). Statistical parametric maps in functional imaging: A general linear approach. Human Brain Mapping, 2, 189–210. Frith, C. D., & Corcoran, R. (1996). Exploring ‘theory of mind’ in people with schizophrenia. Psychological Medicine, 26(3), 521–530. Frith, U. (2001). Mind blindness and the brain in autism. Neuron, 32(6), 969–979. Gallagher, S. (2005). How the body shapes the mind. Oxford University Press. Goldberg, M. E., Bisley, J. W., Powell, K. D., & Gottlieb, J. (2006). Saccades, salience and attention: The role of the lateral intraparietal area in visual behavior. Progress in Brain Research, 155, 157–175. Goodale, M. A., & Milner, A. D. (1992). Separate visual pathways for perception and action. Trends in Neurosciences, 15(1), 20–25. Goodale, M. A. (1998). Vision for perception and vision for action in the primate brain. Novartis Foundation Symposium, 218, 21–34. Graziano, M. S., & Cooke, D. F. (2006). Parieto-frontal interactions, personal space, and defensive behavior. Neuropsychologia, 44(6), 845–859. Grefkes, C., Ritzl, A., Zilles, K., & Fink, G. R. (2004). Human medial intraparietal cortex subserves visuomotor coordinate transformation. NeuroImage, 23(4), 1494–1506. Grefkes, C., & Fink, G. R. (2005). The functional organization of the intraparietal sulcus in humans and monkeys. Journal of Anatomy, 207(1), 3–17. Gregoriou, G. G., & Savaki, H. E. (2001). The intraparietal cortex: Subregions involved in fixation, saccades, and in the visual and somatosensory guidance of reaching. Journal of Cerebral Blood Flow and Metabolism, 21(6), 671–682. Halligan, P. W., Fink, G. R., Marshall, J. C., & Vallar, G. (2003). Spatial cognition: Evidence from visual neglect. Trends in Cognitive Sciences, 7(3), 125–133. Harris, I. M., Egan, G. F., Sonkkila, C., Tochon-Danguy, H. J., Paxinos, G., & Watson, J. D. (2000). Selective right parietal lobe activation during mental rotation: A parametric PET study. Brain, 123(1), 65–73. Kertzman, C., Schwarz, U., Zeffiro, T. A., & Hallett, M. (1997). The role of posterior parietal cortex in visually guided reaching movements in humans. Experimental Brain Research, 114(1), 170–183. Kusunoki, M., Gottlieb, J., & Goldberg, M. E. (2000). The lateral intraparietal area as a salience map: The representation of abrupt onset, stimulus motion, and task relevance. Vision Research, 40(10–12), 1459–1468. Levy, I., Schluppeck, D., Heeger, D. J., & Glimcher, P. W. (2007). Specificity of human cortical areas for reaches and saccades. Journal of Neuroscience, 27(17), 4687–4696. Lobjois, R., Benguigui, N., Bertsch, J., & Broderick, M. P. (2008). Collision avoidance behavior as a function of aging and tennis playing. Experimental Brain Research, 184, 457–468. Medendorp, W. P., Goltz, H. C., & Vilis, T. (2005). Remapping the remembered target location for anti-saccades in human posterior parietal cortex. Journal of Neurophysiology, 94(1), 734–740. Noë, A. (2004). Action in perception. Cambridge, MA: The MIT Press. Ogiso, T., Kobayashi, K., & Sugishita, M. (2000). The precuneus in motor imagery: A magnetoencephalographic study. NeuroReport, 11(6), 1345–1349. Orban, G. A., Claeys, K., Nelissen, K., Smans, R., Sunaert, S., Todd, J. T., et al (2006). Mapping the parietal cortex of human and non-human primates. Neuropsychologia, 44(13), 2647–2667. O’Regan, J. K., & Noë, A. (2001). A sensorimotor account of vision and visual consciousness. Behavioral and Brain Sciences, 24, 939–1031. Perenin, M. T., & Vighetto, A. (1988). Optic ataxia: A specific disruption in visuomotor mechanisms. I. Different aspects of the deficit in reaching for objects. Brain, 111(3), 643–674. Rao, H., Zhou, T., Zhuo, Y., Fan, S., & Chen, L. (2003). Spatiotemporal activation of the two visual pathways in form discrimination and spatial location: A brain mapping study. Human Brain Mapping, 18(2), 79–89. Rizzolatti, G., & Matelli, M. (2003). Two different streams form the dorsal visual system: Anatomy and functions. Experimental Brain Research, 153(2), 146–157. Ruby, P., & Decety, J. (2004). How would you feel versus how do you think she would feel? A neuroimaging study of perspective-taking with social emotions. Journal of Cognitive Neuroscience, 16(6), 988–999. Sakata, H., Taira, M., Kusunoki, M., Murata, A., & Tanaka, Y. (1997). The TINS lecture. The parietal association cortex in depth perception and visual control of hand action. Trends in Neuroscience, 20(8), 350–357. Saxe, R. I., Jamal, N., & Powell, L. (2005). My body or yours? The effect of visual perspective on cortical body representations. Cerebral Cortex, 16(2), 178–182. Schubotz, R. I., & von Cramon, D. Y. (2001). Functional organization of the lateral premotor cortex: fMRI reveals different regions activated by anticipation of object properties, location and speed. Brain Research. Cognitive Brain Research, 11(1), 97–112. Seger, C. A., Stone, M., & Keenan, J. P. (2004). Cortical activations during judgements about the self and an other person. Neuropsychologia, 42, 1168–1177. Sereno, M. I., Pitzalis, S., & Martinez, A. (2001). Mapping of contralateral space in retinotopic coordinates by a parietal cortical area in humans. Science, 294(5545), 1350–1354. Shikata, E., McNamara, A., Sprenger, A., Hamzei, F., Glauche, V., Büchel, C., et al (2007). Localization of human intraparietal areas AIP, CIP, and LIP using surface orientation and saccadic eye movement tasks. Human Brain Mapping, 29(4), 411–421. Stuss, D. T., & Anderson, V. (2004). The frontal lobes and theory of mind: Developmental concepts from adult focal lesion research. Brain and Cognition, 55(1), 69–83. Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of the human brain. Stuttgart: Georg Thieme Verlag. Tunik, E., Frey, S. H., & Grafton, S. T. (2005). Virtual lesions of the anterior intraparietal area disrupt goal-dependent on-line adjustments of grasp. Nature Neuroscience, 8(4), 505–511. Uddin, L. Q., Molnar-Szakacs, I., Zaidel, E., & Iacoboni, M. (2006). RTMS to the right inferior parietal lobule disrupts self-other discrimination. Social Cognitive and Affective Neuroscience, 1(1), 65–71. Uddin, L. Q., Iacoboni, M., Lange, C., & Keenan, J. P. (2007). The self and social cognition: The role of cortical midline structures and mirror neurons. Trends in Cognitive Neuroscience, 11(4), 153–157. Van der Weel, F. R., & van der Meer, A. L. H. (2009). Seeing it coming: Infants’ brain responses to looming danger. Naturwissenschaften, 96, 1385–1391. Vogeley, K., Kurthen, M., Falkai, P., & Maier, W. (1999). Essential functions of the human self model are implemented in the prefrontal cortex. Conscious and Cognition, 8(3), 343–363. Vogeley, K., & Fink, G. R. (2003). Neural correlates of the first-person-perspective. Trends in Cognitive Sciences, 7(1), 38–42. Vogeley, K., May, M., Ritzl, A., Falkai, P., Zilles, K., & Fink, G. R. (2004). Neural correlates of first-person perspective as one constituent of human selfconsciousness. Journal of Cognitive Neurosciences, 16(5), 817–827. Watson, J. D., Myers, R., Frackowiak, R. S., Hajnal, J. V., Woods, R. P., Mazziotta, J. C., et al (1993). Area V5 of the human brain: Evidence from a combined study using positron emission tomography and magnetic resonance imaging. Cerebral Cortex, 3(2), 79–94. Weiss, P. H., Marshall, J. C., Zilles, K., & Fink, G. R. (2003). Are action and perception in near and far space additive or interactive factors? NeuroImage, 18(4), 837–846. Weiss, P. H., Rahbari, N. N., Lux, S., Pietrzyk, U., Noth, J., & Fink, G. R. (2006). Processing the spatial configuration of complex actions involves right posterior parietal cortex: An fMRI study with clinical implications. Human Brain Mapping, 27(12), 1004–1014.
H. Kockler et al. / Consciousness and Cognition 19 (2010) 690–701
701
Wilms, M., Eickhoff, S. B., Specht, K., Amunts, K., Shah, N. J., Malikovic, A., et al (2005). Anatomy and Embryology, 210(5–6), 485–495. Zacks, J. M., Rypma, B., Gabrieli, J. D., Tversky, B., & Glover, G. H. (1999). Imagined transformations of bodies: An fMRI investigation. Neuropsychologia, 37(9), 1029–1040. Zacks, J. M., & Michelon, P. (2005). Transformations of visuospatial images. Behavioral and Cognitive Neuroscience Reviews, 4(2), 96–118. Zacks, J. M. (2008). Neuroimaging studies of mental rotation: A meta-analysis and review. Journal of Cognitive Neuroscience, 20(1), 1–19.
Consciousness and Cognition 19 (2010) 702–710
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Illusory own body perceptions: Case reports and relevance for bodily self-consciousness q Lukas Heydrich a,b, Sebastian Dieguez a, Thomas Grunwald c, Margitta Seeck b, Olaf Blanke a,b,* a
Laboratory of Cognitive Neuroscience, Brain Mind Institute, Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland Department of Neurology, University Hospital Geneva, Switzerland c Swiss Epilepsy Center, Zurich, Switzerland b
a r t i c l e
i n f o
Keywords: Epilepsy Self Body representation Depersonalization Multisensory Parietal cortex Medial prefrontal cortex
a b s t r a c t Neurological disorders of body representation have for a long time suggested the importance of multisensory processing of bodily signals for self-consciousness. One such group of disorders – illusory own body perceptions affecting the entire body – has been proposed to be especially relevant in this respect, based on neurological data as well as philosophical considerations. This has recently been tested experimentally in healthy subjects showing that integration of multisensory bodily signals from the entire body with respect to the three aspects: self-location, first-person perspective, and self-location, is crucial for bodily self-consciousness. Here we present clinical and neuroanatomical data of two neurological patients with paroxysmal disorders of full body representation in whom only one of these aspects, self-identification, was abnormal. We distinguish such disorders of global body representation from related but distinct disorders and discuss their relevance for the neurobiology of bodily self-consciousness. Ó 2010 Elsevier Inc. All rights reserved.
‘‘I am what I seem to be, yet do not seem to be what I am; even to myself I am an insoluble riddle, for my personality has been torn apart!”1 E.T.A. Hoffmann, ‘‘The Devil’s Elixirs”
1. Introduction Unraveling the neural basis of self-consciousness is a major research topic in the cognitive neurosciences and science at large. One line of research has focused on the contributions of bodily processing and body representation to selfconsciousness (Blanke & Metzinger, 2009; Craig, 2002; Damasio, 1999; Jeannerod, 2007). Behavioral work in healthy subjects has studied multisensory (Botvinick & Cohen, 1998; Ehrsson, 2007) and sensorimotor (Blakemore & Frith, 2003; Blakemore, Frith, & Wolpert, 1999; Fourneret & Jeannerod, 1998) aspects of self-consciousness by revealing some of the mechanisms of how bodily processing influences self-consciousness. Neuroimaging studies in healthy subjects have q
This article is part of a special issue of this journal on Self, Other and Memory. * Corresponding author. Address: Laboratory of Cognitive Neuroscience, Brain Mind Institute, Ecole Polytechnique Fédérale de Lausanne (EPFL), Swiss Federal Institute of Technology, Station 19, 1015 Lausanne, Switzerland. Fax: +41 (0) 21 693 96 25. E-mail address: olaf.blanke@epfl.ch (O. Blanke). 1 ‘‘Ich bin das, was ich scheine, und scheine das nicht, was ich bin, mir selbst ein unerklärlich Rätsel, bin ich entzweit mit meinem Ich!!”. E.T.A. Hoffmann, ‘‘Die Elixiere des Teufels” 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.04.010
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
703
complemented this work on multisensory and sensorimotor mechanisms by measuring some of the associated neural mechanisms – including interoceptive and cognitive aspects of self-consciousness temporo-parietal junction, insula precuneus, and medial prefrontal cortex (for reviews see Blakemore & Frith, 2003; Craig, 2002; Northoff et al., 2006; Vogeley & Fink, 2003). Historically, however, it was the detailed clinical reports of neurological and psychiatric patients suffering from disorders of body representation, and illusory own body perceptions (Critchley, 1953; Hécaen & Ajuriaguerra, 1952; Ionasescu, 1960; Lhermitte, 1939; Sierra, Lopera, Lambert, Phillips, & David, 2002), which have suggested close links between bodily processing and self-consciousness for a long time. This clinical work highlighted the importance of multisensory bodily processing in patients suffering from migraine (Lippman, 1952; Podoll & Robinson, 1999; Todd, 1955), vascular stroke (Critchley, 1953), tumors (Hécaen & Ajuriaguerra, 1952; Sierra et al., 2002), epilepsy (Blanke, Landis, Spinelli, & Seeck, 2004; Blanke, Ortigue, Coeytaux, Martory, & Landis, 2003), and psychiatric disorders (Lukianowicz, 1963). Such disorders of body representation2 have been classified as a variety of alterations in perceptual bodily experience such as the experiences of the absence of a body part (Critchley, 1953; Frederiks, 1969; Hécaen & Ajuriaguerra, 1952), body part transformations (Critchley, 1953; Hécaen & Ajuriaguerra, 1952; Ionasescu, 1960; Lippman, 1952), body part displacement (Lippman, 1952; Nightingale, 1982), disconnection of one body part from the body (Blanke et al., 2003; Lippman, 1952), the delusional misidentification of one’s own body part (i.e. somatoparaphrenia) (Gerstmann, 1942; Vallar & Ronchi, 2009), as well as phantom limbs (Hécaen & Ajuriaguerra, 1952; Lhermitte, 1939), and supernumerary phantom limbs (Khateb et al., 2009; Vuilleumier, Reverdin & Landis, 1997) (for reviews see (Blanke, Arzy, & Landis, 2008; Brugger, 2006; Dieguez, Staub, & Bogousslavsky, 2007; Frederiks, 1969; Hécaen & Ajuriaguerra, 1952; Menninger-Lerchenthal, 1935). The brain damage leading to these disorders of body representation suggested a predominant involvement of the right posterior parietal cortex and ipsilesional subcortical regions, although other areas were also found to be implicated, i.e. the lateral prefrontal and premotor cortex (Arzy, Overney, Landis, & Blanke, 2006; Berti et al., 2005; Critchley, 1953; Dieguez et al., 2007; Hécaen & Ajuriaguerra, 1952) and the right posterior insular cortex (Baier & Karnath, 2008). Extending earlier accounts (Brugger, 2002, 2006; Devinsky, Feldmann, Burrowes, & Bromfield, 1989; Grüsser & Landis, 1991; Hécaen & Ajuriaguerra, 1952; Menninger-Lerchenthal, 1935; Mizumoto & Ishikawa, 2005), we have recently proposed that a well-defined group of disorders of body representation – illusory own body perceptions affecting the entire body (or primarily the head and trunk region) – are especially relevant for the study of bodily self-consciousness (Blanke, 2004; Blanke & Metzinger, 2009). We highlighted this importance by opposing such illusory own body perceptions with those affecting an isolated extremity or body part and have based this on neurological and neurophysiological data as well as philosophical arguments. Based on clinical data, several research groups have recently developed methods to study the mechanisms of full-body processing and its relationship to self-consciousness experimentally in healthy subjects (Altschuler & Ramachandran, 2007; Ehrsson, 2007; Lenggenhager, Tadi, Metzinger, & Blanke, 2007; Mizumoto & Ishikawa, 2005). These experimental and clinical data jointly suggest that the integration of visual and multisensory bodily signals from the entire body is important for three major aspects of bodily self-consciousness: self-location (SL; the volume in space where humans experience the self to be located [‘‘where I experience to be‘‘]), first-person perspective (1PP; the directedness of conscious experience [‘‘where I experience to perceive the world from”]) and self-identification (SI; the degree to which humans identify with their body [‘‘what I experience as my body”]). SL, 1PP, and SI are abnormal in patients with global illusory own body perceptions (Blanke & Metzinger, 2009). Here we present clinical and anatomical data from two epileptic patients suffering from rare illusory own body perceptions associated with abnormal SI and involving predominantly the trunk and head, due to damage to right posterior dorso-medial parietal cortex (patient 1) and the right dorso-medial prefrontal cortex (patient 2). We discuss the phenomenology, etiology, and lesion location in these two cases in regard to the neurological literature and the larger field of the cognitive neuroscience of bodily self-consciousness. 2. Methods Both patients were recruited at the University Hospital of Geneva, where they underwent full diagnostic workup, including a standardized neurological, psychiatric and neuropsychological examination (Pegna, Qayoom, Gericke, Landis, & Seeck, 1998), electroencephalography (EEG, including source imaging of interictal epileptic spikes, Grave de Peralta Menendez, Gonzalez Andino, Lantz, Michel, & Landis, 2001; Lantz, Grave de Peralta Menendez, Gonzalez Andino, & Michel, 2001; Michel et al., 2004), magnetic resonance imaging (MRI) and positron emission tomography (PET). Patient 2 in addition underwent invasive presurgical evaluations at the Swiss Epilepsy Center in Zurich. Epilepsy surgery of patient 2 was performed at the Department of Neurosurgery at the University Hospital of Zurich. In both patients we conducted a semi-structured interview focusing on aspects of bodily self-consciousness, such as 1PP, SL, SI, as well as somatosensory, visual, auditory, and vestibular symptoms and emotions (following Blanke & Mohr, 2005). Additionally patient 2 was asked to complete the Cambridge Depersonalization Scale (CDS) (Sierra & Berrios, 2000). The CDS comprises 29-items inquiring about subjective experiences classically associated with the depersonalization syndrome.
2 Due to the poor definition of the term body schema (as well as the related term body image) and even more importantly its inconsistent use in the neurological literature, we decided to use the more neutral term body representation (see de Vignemont, 2010).
704
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
Each item is rated on two Likert scales that quantify frequency (range 0–4) and duration (range 1–6) (cutoff score for depersonalization disorder is 70). 3. Results 3.1. Patient 1 Patient 1 is a 55 year old, left-handed male patient suffering from epilepsy since the age of 14 years. His simple partial sensorimotor seizures affected his left hand had been well controlled under anti-epileptic medication until the onset of paroxysmal episodes of vertigo 9 years before the current hospitalization. At that time he additionally started to experience the following, highly stereotypical pattern of symptoms: without any prior warning he would first have the impression of an increasing pressure in the entire left hemi-body. This sensation increased progressively in strength leading eventually to the sensation that he was invaded by a stranger in his left hemi-body. At this time he also sensed that the left half of his head, the upper part of his left trunk, his left arm and his left leg were no longer belonging to him (no misattribution), that these parts were disconnected from the rest of his body, and that his body was divided into two parts (Fig. 1A). Sometimes this was followed by the impression that the left arm was moving unintentionally and would disappear behind the patient’s back. During these episodes he never experienced any deformation or other changes of his body or the environment. Furthermore, no autoscopic hallucinations, no sensation of floating or disembodiment, no change in visuo-spatial or first-person perspective, no disturbance of language or vision and no loss of contact or consciousness were noted. During these sensations the patient localized the self as within the right side of his body (shown in grey in Fig. 1). He managed to remain calm and was able to continue standing, walking, and even give oral presentations while in front of audiences at work (surrounding persons usually did not notice his seizure manifestations). These simple partial seizures occurred on a daily basis and lasted 1 min. Neurological, psychiatric, and neuropsychological examinations were normal. Surface EEG at the time of hospitalization did not reveal any pathological correlates. MRI showed a hypo-dense lesion in the right posterior intraparietal sulcus predominantly in the medial wall in the T1 weighted sequence (Fig. 2), which was confirmed by fluid-attenuated inverse recovery (FLAIR) and PET compatible with arteriovenous malformation. The diagnosis of epileptic seizures was retained,
Fig. 1. Phenomenology patient 1. Simple partial seizures were characterized by the impression of increasing pressure in the entire left hemi-body. This sensation increased progressively in strength leading eventually to the sensation that he was invaded by a stranger in his left hemi-body. At this time he also sensed that the left half of his head, the upper part of his left trunk, his left arm and his left leg were no longer belonging to him, that these parts were disconnected from the rest of his body, and that his body was divided into two parts.
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
705
Fig. 2. MRI patient 1. T1-weighted magnetic resonance imaging reveals a hypointense cortical and subcortical lesion in the right posterior intraparietal sulcus (arrows), consistent with an arteriovenous malformation. (A) Axial view, (B) 3-D reconstruction. The hypointense lesion is indicated by an arrow.
even though no epileptic discharges were recorded on repeated EEGs. This was based on the history of the patient (left-lateralized seizures as a child), the current symptomatology, and the corresponding neocortical lesion location. 3.2. Patient 2 Patient 2 is a 30 year old, left-handed male patient suffering from pharmaco-resistant epilepsy. He was referred to the presurgical epilepsy unit at the University Hospital of Geneva and the Epilepsy Clinic in Zurich for further evaluation.
Fig. 3. Phenomenology patient 2. Complex partial seizures were characterized by altered bodily awareness including a total loss or a strongly diminished awareness of bodily signals, which predominated at the lower trunk and legs and also included – to a lesser degree – his upper trunk and neck (black; the patient could not indicate whether the upper extremities were affected or not, illustrated by dashed black and grey lines). This was accompanied by the impression that everything below his neck was ‘‘numb”, ‘‘useless” and somewhat inaccessible to conscious awareness while the experience of his head region was experienced as light and as detached from the rest of his lower body (light grey).
706
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
Seizures started at the age of 11 years, with episodes characterized by the following intense feelings of altered bodily awareness. During these episodes he felt a total loss or a strongly diminished awareness of his body (that he compared to numbness). This sensation was localized and predominated at the legs and lower trunk and also included – to a lesser degree – his upper trunk and neck (the patient could not indicate whether the upper extremities were affected or not (Fig. 3). This altered bodily awareness progressively increased in strength and eventually led to the impression that everything below his neck was ‘‘numb”, ‘‘useless” and somewhat inaccessible to conscious awareness whereas the perception of his head remained unaffected and was even experienced as light and as detached from the rest of his lower body. Following this, he described that he felt ‘‘as an observer of his body and as the one being observed” at the same time. However, he never experienced floating, elevation, or vestibular sensations, nor a change in visuo-spatial or first-person perspective, and he never felt physically disembodied (as is typically mentioned in out-of-body experiences; (Blanke et al., 2004; Brugger, Regard, & Landis, 1997; Devinsky et al., 1989) nor saw an image of himself in external space (autoscopy; i.e. (Blanke et al., 2004; Brugger et al., 1997). The patient also noted a feeling of disconnection from his own thoughts and past, and that he was no longer in control of his actions and speech, like being a robot. In addition the patient often felt as if being detached from the environment, as if sounds and voices were being transformed and distant, perceived as if ‘‘through a veil”. These sensations were mostly observed as isolated auras but also could indicate the imminent onset of a complex partial seizure with involuntary movements of the left arm, head deviation to the left, loss of consciousness, and secondary tonic-clonic generalization. The frequency of these auras was variable, ranging from one per day to one per hour and most often occurred during phases of relaxation. The duration was estimated as 30 s. Different anti-epileptic drugs were tried without any success. Family history and personal history were negative for epilepsy or psychiatric disease. Neurological and psychiatric examinations were unremarkable. The interictal neuropsychological evaluation showed executive and attentional deficits: high distractibility and intrusive thoughts, as well as mild impairments in mental flexibility. The CDS yielded a score of 46. Ictal and interictal high resolution EEG recordings (128 channels) revealed right frontal and fronto-temporal spikes, spike-waves, and sharp waves and subsequent source imaging of epileptic spikes (Grave de Peralta Menendez et al., 2001; Lantz et al., 2001; Michel et al., 2004) suggested an epileptogenic focus in the right middle and superior frontal gyrus (Fig. 4A). Single photon emission tomography (SPECT) demonstrated a hyper-perfusion of the right middle frontal gyrus, the right anterior cingulate cortex, and the bilateral operculum. Repeated MRI did not show any lesion, but intracranial EEG recordings and functional mapping using 52 surgically implanted subdural grid electrodes revealed an epileptogenic focus in the right supplementary motor area (SMA) and the right superior frontal gyrus (Fig. 4B). Subsequent complete resection of the right SMA and partial resection of the right superior frontal gyrus resulted in partial seizure control. There were no more seizures during the day and significantly fewer seizures during sleep. No more auras have been described since this first operation (15 months follow-up). Meanwhile, this first topectomy has been extended during a second epilepsy surgical procedure, after which no seizures have recurred yet (4 month follow-up).
Fig. 4. A. Interictal Spike Mapping. Interictal Spike Mapping with 256-channel EEG in patient 2 and source imaging of epileptic spikes suggested an epileptogenic focus in the right middle and superior frontal gyrus. The significant voxels at a p < 0.01 threshold (corrected for multiple comparisons, e.g. for the number of electrodes using Bonferroni correction) are indicated in green. The maximum of the estimated source of the averaged interictal spike is indicated in red (Lantz et al., 2001). B. Intracranial EEG and functional mapping. Intracranial EEG recordings and functional mapping using 52 surgically implanted subdural grid electrodes revealed an epileptogenic focus in the right supplementary motor area (SMA) and the right superior frontal gyrus (these electrodes are indicated in red). Subsequent partial resection of the right superior frontal gyrus and the SMA resulted in partial seizure control (postoperative lesion indicated in red). After a second operation extending the topectomy no seizures have recurred yet. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
707
4. Discussion We present two patients suffering from non-visual illusory own body perceptions due to epilepsy that affected the head-trunk region of the body and were accompanied by a disturbance of SI with the body. Patient 1 suffered from abnormal own body perceptions restricted to his left contralesional hemi-body including the head that was experienced as strange, as if invaded by a stranger, and as if disconnected from his right ipsilesional hemi-face and -body. Patient 2 reported abnormal own body perceptions for the trunk, neck, and the legs bilaterally (but not the head), which were experienced as weaker and numb and as if disconnected from the head. In both patients these illusory own body perceptions were caused by neocortical epilepsy involving two dorso-medial brain regions that have been linked to self-related processing by recent cognitive neuroscience work in healthy subjects. Here, we (1) distinguish such rare illusory own body perceptions (those that affect the head-trunk region) from the more commonly reported illusory own body perceptions that affect the contralesional upper extremity and discuss their underlying neural correlates, (2) propose that they share aspects with certain cases of somatoparaphrenia and visual illusory own body perceptions, (3) discuss their relevance for the study of bodily self-consciousness, and (4) highlight contributions of the dorso-medial posterior parietal cortex and prefrontal cortex for the neurobiology of bodily self-consciousness. 4.1. Illusory own body perceptions: head and trunk versus upper extremity Compared to most patients with disorders of body representation, the present two patients differ concerning the affected body part, the size of the affected body surface and their underlying neural correlates. Disorders of body representation are usually characterized by an alteration in bodily experience that mostly concerns the contralesional upper extremity and are mostly caused by damage to the right ventro-lateral prefrontal cortex, right inferior parietal cortex, or right temporo-parietal cortex. Classically, these are the experience of the absence of a body part (Arzy et al., 2006; Critchley, 1953; Frederiks, 1969; Hécaen & Ajuriaguerra, 1952), of body part transformations (Critchley, 1953; Hécaen & Ajuriaguerra, 1952; Ionasescu, 1960; Lippman, 1952), of body part displacement (Lippman, 1952; Nightingale, 1982), of a disconnection of one body part from the body (Blanke et al., 2003; Lippman, 1952), as well as somatoparaphrenia (Gerstmann, 1942; Vallar & Ronchi, 2009) and supernumerary phantom limbs (Khateb et al., 2009). A discussion of the present two patients with respect to somatoparaphrenia seems particularly relevant and we will focus here on this comparison. Somatoparaphrenia is associated with somatosensory and hemiplegic motor deficits of the contralesional (mostly upper) extremity as well as neglect and results mostly from damage to right temporo-parietal cortex and/or the right insula (Baier & Karnath, 2008). Somatoparaphrenic patients typically misidentify the affected hand as belonging to somebody else or misidentify another person’s hand as belonging to themselves (Baier & Karnath, 2008; Gerstmann, 1942; Vallar & Ronchi, 2009). This condition is often considered an important condition to study bodily self-consciousness, especially concerning self-attribution of the affected extremity. Yet, Blanke and Metzinger (2009) have argued that the aspects of bodily self-consciousness that are crucial for the conscious self (SL, 1PP, SI) are rather based on global body representations in the brain. Yet, as no change in 1PP, SL (the self is experienced as localized within the patient’s body, e.g. head-trunk) and SI (e.g. only the contralesional hand is not attributed to the self, whereas the patients still do self-identify with their body) is noted in somatoparaphrenia. Thus, the study of such patients does not allow to study the conscious self as defined by SL, 1PP, SI. This differed in both present patients who suffered from abnormal self-identification. Patient 1 experienced a strange and disconnected contralesional hemi-body including trunk, neck, head and extremities and patient 2 experienced a disconnection from his neck, trunk and both lower extremities, that were felt as less present and numb. In addition, patient 1 reported that the perceptually altered body segments were experienced as invaded by (the body of) another person. Especially the latter aspect, but also the involvement of the midline body region (that affected the head in patient 1) and the larger extension of the affected body surface, suggests the presence of a disorder of self-identification with the body of the patients at a more global level than somatoparaphrenia. These observations of abnormal self-identification with one’s own head and/or trunk are reminiscent of earlier clinical observations in patients with somatoparaphrenia, in whom not only a limb, but also the contralesional hemi-body was affected (Gerstmann, 1942; Glonning, Jellinger, & Tschabiter, 1963; Hoff & Pötzl, 1935/1988; Lhermitte, 1939; MenningerLerchenthal, 1935; Pötzl, 1925, for reviews see Blanke et al., 2008; Brugger, 2006). Thus, Pötzl (1925; case 1) described a patient with left-sided hemiplegia who not only claimed that his left arm belonged to an unknown person, as seen in many patients with somatoparaphrenia, but also that there was another (unseen) person lying in his bed to his left and that this person tried to push him out of the bed. Engerth and Hoff (1929) described a 71 year-old man with left-sided hypoesthesia, hemianopia (with hemianopic hallucinations), and anososognosia who experienced a constant left-sided person who was most often localized next or behind the patient. Biancone described a 72 year-old female patient with left-sided hemiplegia and hemianesthesia that claimed that her left hemi-body belonged to another person that was lying in her bed (quoted after Lhermitte, 1939). In some rare cases, the body of the felt other is even experienced as invading the patient’s body, as noted by patient 1. Thus, a patient with a parasagittal meningioma adjacent to the right posterior parietal cortex reported by Nightingale (1982) described that his left hemi-body felt strange, seemed to have shifted backwards and was invaded (and controlled) by external agents (mostly the patient’s father). Another patient reported by Gloning (1963, case 1) also reported that her left hemi-body felt strange compared to her right hemi-body and was shifted backwards, due to a glioma in right posterior occipito-parietal cortex. These reports point to the importance of the right posterior and medial parietal
708
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
cortex for illusory own body perceptions that do not only affect the contralesional hand but also the midline trunk/head region making self-identification and self-other distinction more ambiguous. We also note that perceptual aberrations and abnormal self-identification of these patients and patient 1 are also comparable to some patients suffering from heautoscopy and the feeling of a presence (for review, see Blanke et al., 2008) that may be associated with an experienced split of the trunk and/or head (e.g. Blanke et al., 2008; Brugger, Blanke, Regard, Bradford, & Landis, 2006). Finally there might also be a partial overlap with the concept of ego-pathology as described by Scharfetter (1981), which has mainly been applied to psychiatric patients. 4.2. Self-identification and neural full-body representation Here, we argue that disturbances in self-identification and perceptual aberrations with respect to the head and/or trunk region are observed when head and/or trunk representations in medio-dorsal parietal or prefrontal cortex are affected. This contrasts functionally and anatomically with damage to more ventro-lateral regions in parietal and prefrontal cortex that are more typically lesioned in patients with perceptual aberrations affecting a certain body part (Arzy et al., 2006; Berti et al., 2005; Hécaen & Ajuriaguerra, 1952). This dissociation between head and trunk versus arm and hand is compatible with the major functional properties of single neurons in posterior parietal cortex. Neurons in primary somatosensory cortex (for example area 3b) have small tactile receptive fields that respond preferentially or only to stimuli applied to a specific and contralateral body part (such as the finger tip or the toe (Gardner, 1988). Neurons in area 1 and 2 have larger receptive fields and respond to stimuli applied to one hand or foot and neurons in unimodal somatosensory association cortex (area 5; (Taoka, Toda, Iriki, Tanaka, & Iwamura, 2000) have even larger receptive fields responding to cutaneous stimuli applied to proximal parts of an extremity or the trunk. Yet, neurons with somatosensory receptive fields that encode the hemi-body or the entire body (global somatosensory receptive fields) have also been described in the ventral intraparietal region (VIP). These neurons respond to and encode preferentially the head and trunk region, but may encode also the left or right hemi-body, the upper or lower hemi-body, or the entire body surface (Duhamel, Colby, & Goldberg, 1991, 1998). These VIP neurons have not only large somatosensory receptive fields, but also large visual receptive fields, and contain many bimodal neurons integrating visual and somatosensory stimuli (Bremmer, Klam, Duhamel, Ben Hamed, & Graf, 2002; Duhamel et al., 1998; Schlack, Hoffmann, & Bremmer, 2002). VIP in humans has been suggested to be localized in the intraparietal sulcus (Bremmer et al., 2001) and thus overlaps with the regions affected in patient 1. Cells with similar functional properties are likely to exist in premotor cortex and the SMA (Graziano, 1999; Penfield & Jasper, 1954). We therefore speculate that processing of VIP in patient 1 and premotor cortex and SMA (and adjacent regions with similar functional properties) in patient 2 was abnormal and gave rise to disturbances in self-identification and perceptual aberrations affecting head and/or trunk and thus full body representations. 4.3. Cognitive neuroscience of bodily self-consciousness The data in patient 2 may further be of relevance for the neurobiology of depersonalization, as he also reported a non-perceptual detachment of the self and loss of self-relevance. Patients suffering from depersonalization often feel detached and alienated from their body and/or mental processes, feeling as an outside ‘‘observer”, while having no or less control over their actions (DSM-IV., 2000). Although somewhat reminiscent of autoscopic phenomena, such as out-of-body experience or heautoscopy (Brugger, 2002), there is generally no experienced perceptual change in 1PP or in SL in patients with depersonalization. Interestingly, a recent model regarding the neurobiological correlates of depersonalization suggests an involvement of the prefrontal cortex (Sierra & Berrios, 1998) proposing that increased prefrontal activity may lead to limbic inhibition that results in decreased autonomic output and hypo-emotionality towards the self and the world. Together with neuroimaging data showing the medial prefrontal cortex’s role in cognitive aspects of self-related processing (Northoff et al., 2006), its recruitment during behavioral tasks demanding imagined perspective transformations (Vogeley & Fink, 2003), it may be suggested that activity in the medial prefrontal cortex may reflect more cognitive than spatial aspects of the 1PP, including self reference, self concept and a mental representation of oneself as a subject of experience that may be disturbed in patients with depersonalization. Although bodily self-consciousness and its three components, SI, 1PP, and SL, have been linked to multisensory integration at the temporo-parietal junction, careful study of patients with depersonalization as well as patients with disorders of body representation affecting the head and/or trunk region may provide additional insights into the neurobiology of selfhood. Such self-related brain activity is likely to be distributed in a network of brain regions including the temporoparietal junction, the dorso-medial parietal cortex, the prefrontal cortex and the default network (Ehrsson, Spence, & Passingham, 2004; Gusnard, Akbudak, Shulman, & Raichle, 2001; Hanakawa et al., 2003; Kircher et al., 2000; Macrae, Moran, Heatherton, Banfield, & Kelley, 2004; Maguire et al., 1998; Northoff et al., 2006; Ruby & Decety, 2001; Vogeley & Fink, 2003). Acknowledgments LH and MS are supported by the Swiss National Science Foundation (Grants 33CM30-124089, 33CM30-123115, 320030122073, 323530-123718). OB was supported by the Swiss National Science Foundation (Sinergia Grant CRSII1-125135: Balancing Self and Body).
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
709
References Altschuler, E. L., & Ramachandran, V. S. (2007). A simple method to stand outside oneself. Perception, 36(4), 632–634. Arzy, S., Overney, L. S., Landis, T., & Blanke, O. (2006). Neural mechanisms of embodiment: Asomatognosia due to premotor cortex damage. Archieves of Neurology, 63(7), 1022–1025. Baier, B., & Karnath, H. O. (2008). Tight link between our sense of limb ownership and self-awareness of actions. Stroke, 39(2), 486–488. Berti, A., Bottini, G., Gandola, M., Pia, L., Smania, N., Stracciari, A., et al (2005). Shared cortical anatomy for motor awareness and motor control. Science, 309(5733), 488–491. Blakemore, S. J., & Frith, C. (2003). Self-awareness and action. Current Opinion in Neurobiology, 13(2), 219–224. Blakemore, S. J., Frith, C. D., & Wolpert, D. M. (1999). Spatio-temporal prediction modulates the perception of self-produced stimuli. Journal of Cognitive Neuroscience, 11(5), 551–559. Blanke, O. (2004). Out of body experiences and their neural basis. BMJ, 329(7480), 1414–1415. Blanke, O., Arzy, S., & Landis, T. (2008). Chapter 22 Illusory reduplications of the human body and self. Handbook of Clinical Neurology, 88, 429–458. Blanke, O., Landis, T., Spinelli, L., & Seeck, M. (2004). Out-of-body experience and autoscopy of neurological origin. Brain, 127(Pt 2), 243–258. Blanke, O., & Metzinger, T. (2009). Full-body illusions and minimal phenomenal selfhood. Trends in Cognitive Sciences, 13(1), 7–13. Blanke, O., & Mohr, C. (2005). Out-of-body experience, heautoscopy, and autoscopic hallucination of neurological origin implications for neurocognitive mechanisms of corporeal awareness and self-consciousness. Brain Research Brain Research Reviews, 50(1), 184–199. Blanke, O., Ortigue, S., Coeytaux, A., Martory, M. D., & Landis, T. (2003). Hearing of a presence. Neurocase, 9(4), 329–339. Botvinick, M., & Cohen, J. (1998). Rubber hands ’feel’ touch that eyes see. Nature, 391(6669), 756. Bremmer, F., Klam, F., Duhamel, J. R., Ben Hamed, S., & Graf, W. (2002). Visual-vestibular interactive responses in the macaque ventral intraparietal area (VIP). European Journal of Neuroscience, 16(8), 1569–1586. Bremmer, F., Schlack, A., Shah, N. J., Zafiris, O., Kubischik, M., Hoffmann, K., et al (2001). Polymodal motion processing in posterior parietal and premotor cortex: A human fMRI study strongly implies equivalencies between humans and monkeys. Neuron, 29(1), 287–296. Brugger, P. (2002). Reflective mirrors: Perspective-taking in autoscopic phenomena. Cognitive Neuropsychiatry, 7(3), 179–194. Brugger, P. (2006). From phantom limbs to phantom bodies: Varieties of extracorporeal awareness. In G. Knoblich, I. M. Thornton, M. Grosjean, & M. Shiffrar (Eds.), Human body perception from the inside out (pp. 171). Oxford: Oxford University Press. Brugger, P., Blanke, O., Regard, M., Bradford, D. T., & Landis, T. (2006). Polyopic heautoscopy: Case report and review of the literature. Cortex, 42(5), 666–674. Brugger, P., Regard, M., & Landis, T. (1997). Illusory reduplication of own’s own body: Phenomenology and classification of autoscopic phenomena. Cognitive Neuropsychiatry(2), 19–38. Craig, A. D. (2002). How do you feel–now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10(1), 59–70. Critchley, M. (1953). The parietal lobes. New York, London: Hafner Publishing Company. Damasio, A. (1999). The feeling of what happens. London: Random House. de Vignemont, F. (2010). Body schema and body image–pros and cons. Neuropsychologia, 48(3), 669–680. Devinsky, O., Feldmann, E., Burrowes, K., & Bromfield, E. (1989). Autoscopic phenomena with seizures. Archieves of Neurology, 46(10), 1080–1088. Dieguez, S., Staub, F., & Bogousslavsky, J. (2007). Asomatognosia. In O. Godefroy & J. Bogousslavsky (Eds.), The behavioral and cognitive neurology of stroke (pp. 215–253). Cambridge: Cambridge University Press. DSM-IV (2000). Diagnostic and statistical manual of mental disorders–DSM-IV (4th ed.). Washington, DC: American Psychiatric Association. Duhamel, J. R., Colby, C. L., & Goldberg, M. E. (1991). Congruent representations of monkey ventral intra-parietal cortex (area VIP). In J. Paillard (Ed.), Brain and space. Oxford University Press. Duhamel, J. R., Colby, C. L., & Goldberg, M. E. (1998). Ventral intraparietal area of the macaque: Congruent visual and somatic response properties. Journal of Neurophysiology, 79(1), 126–136. Ehrsson, H. H. (2007). The experimental induction of out-of-body experiences. Science, 317(5841), 1048. Ehrsson, H. H., Spence, C., & Passingham, R. E. (2004). That’s my hand! Activity in premotor cortex reflects feeling of ownership of a limb. Science, 305(5685), 875–877. Engerth, G., & Hoff, H. (1929). Ein Fall von Halluzinationen im hemianoptischen Gesichtsfeld. Beitrag zur Genese der optischen Halluzinationen. Monatschr Psychiatr Neurol, 74, 246–256. Fourneret, P., & Jeannerod, M. (1998). Limited conscious monitoring of motor performance in normal subjects. Neuropsychologia, 36(11), 1133–1140. Frederiks, J. A. M. (1969). Disorders of the body schema. In G. W. Bruyn & P. J. Vinken (Eds.). Handbook of clinical neurology (Vol. 4, pp. 207–240). . Gardner, E. P. (1988). Somatosensory cortical mechanisms of feature detection in tactile and kinesthetic discrimination. Canadian Journal of Physiology and Pharmacology, 66(4), 439–454. Gerstmann, J. (1942). Problem of imperception of disease and of impaired body territories with organic lesions. Relation to body schema and its disorders. Archives of Neurology and Psychiatry, 48, 890–913. Glonning, I., Jellinger, K., & Tschabiter, H. (1963). Über einen obduzierten Fall von optischer Körperschemastörung und Heautoskopie. Neuropsychologia, 1, 217–231. Grave de Peralta Menendez, R., Gonzalez Andino, S., Lantz, G., Michel, C. M., & Landis, T. (2001). Noninvasive localization of electromagnetic epileptic activity. I. Method descriptions and simulations. Brain Topography, 14(2), 131–137. Graziano, M. S. (1999). Where is my arm? The relative role of vision and proprioception in the neuronal representation of limb position. Proceedings of the National Academy Sciences of the United States of America, 96(18), 10418–10421. Grüsser, O.-J., & Landis, T. (1991). The splitting of ‘I’ and ‘me’: Heautoscopy and related phenomena. In T. Grüsser & O.-J. Landis (Eds.), Visual agnosias and other disturbances of visual perception and cognition (pp. 297–303). Amsterdam: MacMillan. Gusnard, D. A., Akbudak, E., Shulman, G. L., & Raichle, M. E. (2001). Medial prefrontal cortex and self-referential mental activity: Relation to a default mode of brain function. Proceeding of the National Academy of Sciences of the United States of America, 98(7), 4259–4264. Hanakawa, T., Immisch, I., Toma, K., Dimyan, M. A., Van Gelderen, P., & Hallett, M. (2003). Functional properties of brain areas associated with motor execution and imagery. Journal of Neurophysiology, 89(2), 989–1002. Hécaen, H., & Ajuriaguerra, J. (1952). Méconnaissances et hallucinations corporelles. Paris: Masson. Hoff, H., & Pötzl, O. (1935/1988). Transformation between body image and external worl. In J. Brown (Ed.), Agnosia and apraxia: Selected papers of liepmann, lange, and pötzl (pp. 251–262). Hillsdale, NJ: Lawrence Erlbaum. Ionasescu, V. (1960). Paroxysmal disorders of the body image in temporal lobe epilepsy. Acta Psychiatrica Scandinavica, 35, 171–181. Jeannerod, M. (2007). Being oneself. Journal of Physiology Paris, 101(4–6), 161–168. Khateb, A., Simon, S. R., Dieguez, S., Lazeyras, F., Momjian-Mayor, I., Blanke, O., et al (2009). Seeing the phantom: A functional magnetic resonance imaging study of a supernumerary phantom limb. Annals of Neurology, 65(6), 698–705. Kircher, T. T., Senior, C., Phillips, M. L., Benson, P. J., Bullmore, E. T., Brammer, M., et al (2000). Towards a functional neuroanatomy of self processing: Effects of faces and words. Brain Research Cognitive Brain Research, 10(1–2), 133–144. Lantz, G., Grave de Peralta Menendez, R., Gonzalez Andino, S., & Michel, C. M. (2001). Noninvasive localization of electromagnetic epileptic activity. II. Demonstration of sublobar accuracy in patients with simultaneous surface and depth recordings. Brain Topography, 14(2), 139–147. Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317(5841), 1096–1099. Lhermitte, J. (1939). L’image de notre corps (Editions de la Nouvelle Revue Critque ed.). Paris. Lippman, C. W. (1952). Certain hallucinations peculiar to migraine. Journal of Nervous and Mental Disease, 116(4), 346–351. Lukianowicz, N. (1963). ‘‘Body image‘‘ disturbances in psychiatric disorders. British Journal of Psychiatry, 113(494), 31–47.
710
L. Heydrich et al. / Consciousness and Cognition 19 (2010) 702–710
Macrae, C. N., Moran, J. M., Heatherton, T. F., Banfield, J. F., & Kelley, W. M. (2004). Medial prefrontal activity predicts memory for self. Cerebral Cortex, 14(6), 647–654. Maguire, E. A., Burgess, N., Donnett, J. G., Frackowiak, R. S., Frith, C. D., & O’Keefe, J. (1998). Knowing where and getting there: A human navigation network. Science, 280(5365), 921–924. Menninger-Lerchenthal, E. (1935). Das Truggebilde der eigenen Gestalt. Berlin: S. Karger. Michel, C. M., Murray, M. M., Lantz, G., Gonzalez, S., Spinelli, L., & Grave de Peralta, R. (2004). EEG source imaging. Clinical Neurophysiology, 115(10), 2195–2222. Mizumoto, M., & Ishikawa, M. (2005). Immunity to error through misidentification and the bodily illusion experiment. Journal of Consciousness Studies, 12(7), 3–19. Nightingale, S. (1982). Somatoparaphrenia: A case report. Cortex, 18(3), 463–467. Northoff, G., Heinzel, A., de Greck, M., Bermpohl, F., Dobrowolny, H., & Panksepp, J. (2006). Self-referential processing in our brain – A meta-analysis of imaging studies on the self. Neuroimage, 31(1), 440–457. Pegna, A. J., Qayoom, Z., Gericke, C. A., Landis, T., & Seeck, M. (1998). Comprehensive postictal neuropsychology improves focus localization in epilepsy. European Neurology, 40(4), 207–211. Penfield, W., & Jasper, H. (1954). Epilepsy and the functional anatomy of the human brain. Boston: Little, Brown and Company. Podoll, K., & Robinson, D. (1999). Lewis Carroll’s migraine experiences. Lancet, 353(9161), 1366. Pötzl, O. (1925). Über Störungen der Selbstwahrnehmung bei linksseitiger Hemiplegie. Zeitschrift für die gesamte Neurologie und Psychiatrie, 93, 117–168. Ruby, P., & Decety, J. (2001). Effect of subjective perspective taking during simulation of action: A PET investigation of agency. Nature Neuroscience, 4(5), 546–550. Scharfetter, C. (1981). Ego-psychopathology: The concept and its empirical evaluation. Psychological Medicine, 11(2), 273–280. Schlack, A., Hoffmann, K. P., & Bremmer, F. (2002). Interaction of linear vestibular and visual stimulation in the macaque ventral intraparietal area (VIP). European Journal of Neuroscience, 16(10), 1877–1886. Sierra, M., & Berrios, G. E. (1998). Depersonalization: Neurobiological perspectives. Biological Psychiatry, 44(9), 898–908. Sierra, M., & Berrios, G. E. (2000). The Cambridge Depersonalization Scale: A new instrument for the measurement of depersonalization. Psychiatry Research, 93(2), 153–164. Sierra, M., Lopera, F., Lambert, M. V., Phillips, M. L., & David, A. S. (2002). Separating depersonalisation and derealisation: The relevance of the ‘‘lesion method‘‘. Journal of Neurology Neurosurgery and Psychiatry, 72(4), 530–532. Taoka, M., Toda, T., Iriki, A., Tanaka, M., & Iwamura, Y. (2000). Bilateral receptive field neurons in the hindlimb region of the postcentral somatosensory cortex in awake macaque monkeys. Experimental Brain Research, 134(2), 139–146. Todd, J. (1955). The syndrome of Alice in Wonderland. Canadian Medical Association Journal, 73(9), 701–704. Vallar, G., & Ronchi, R. (2009). Somatoparaphrenia: A body delusion. A review of the neuropsychological literature. Experimental Brain Research, 192(3), 533–551. Vogeley, K., & Fink, G. R. (2003). Neural correlates of the first-person-perspective. Trends in Cognitive Sciences, 7(1), 38–42. Vuilleumier, P., Reverdin, A., & Landis, T. (1997). Four legs. Illusory reduplication of the lower limbs after bilateral parietal lobe damage. Archives of Neurology, 54(12), 1543–1547.
Consciousness and Cognition 19 (2010) 711–720
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Grasping language – A short story on embodiment q Doreen Jirak a,e, Mareike M. Menz a, Giovanni Buccino b, Anna M. Borghi b,c,1, Ferdinand Binkofski a,d,*,1 a
Department of Systems Neuroscience and Neuroimage Nord, University Medical Center Hamburg Eppendorf, Hamburg, Germany Magna Graecia, University of Catanzaro, Catanzaro, Italy c Institute of Cognitive Science and Technology, CNR, Rome, Italy d Section for Neurological Cognitive Research, RWTH Aachen University, Aachen, Germany e MIN-Faculty, Department of Informatics, University of Hamburg, Germany b
a r t i c l e
i n f o
Article history Available online 23 August 2010 Keywords: Embodiment Language Perception Action fMRI Meta-analysis ALE
a b s t r a c t The new concept of embodied cognition theories has been enthusiastically studied by the cognitive sciences, by as well as such disparate disciplines as philosophy, anthropology, neuroscience, and robotics. Embodiment theory provides the framework for ongoing discussions on the linkage between ‘‘low” cognitive processes as perception and ‘‘high” cognition as language processing and comprehension, respectively. This review gives an overview along the lines of argumentation in the ongoing debate on the embodiment of language and employs an ALE meta-analysis to illustrate and weigh previous findings. The collected evidence on the somatotopic activation of motor areas, abstract and concrete word processing, as well as from reported patient and timing studies emphasizes the important role of sensorimotor areas in language processing and supports the hypothesis that the motor system is activated during language comprehension. Ó 2010 Elsevier Inc. All rights reserved.
1. The idea of embodiment ‘‘Grasp the subject, the words will follow” was advised by Cato the Elder. Grasping an explanation, giving an example, posing a threat – language is full of actions and objects, and the ties between language and motion are under continuous investigation. Embodied cognition theories are becoming more and more popular in cognitive (neuro)science, as well as in philosophy, anthropology, cognitive psychology, and robotics (e.g. Nolfi & Floreano, 2000; Ziemke, 2002). According to the embodied view, there is no separation between the so-called ‘‘low” cognitive processes, such as perception and action, and ‘‘high” cognitive processes, such as language and thought. Generally, embodiment links the individual sensorimotor experiences with higher cognitive functions such as language processing and comprehension. Connecting motor abilities with cognitive capacities contradicts with the classic amodal view, which assumes a clear-cut separation between low and high level processes and which states that cognition derives from computational processes in separate domains. When applied to language, embodied cognition views claim that when we understand words, the same sensorimotor areas are recruited as for interacting with the objects and entities the words refer to. Similarly, when we comprehend sentences, we internally simulate the state of the world the sentences describe (Zwaan, 2004). In the past years much behavioral
q
This article is part of a speical issue of this journal on Self, Other and Memory. * Corresponding author. Address: Section for Neurological Cognitive Research, RWTH Aachen University, Pauwelstrasse 30, D-52074 Aachen, Germany. Fax: +49 241 80 82139. E-mail address:
[email protected] (F. Binkofski). 1 These authors contributed equally. 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.020
712
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
and neural evidence has been collected, showing that the process of language comprehension activates a motor simulation (Gallese, 2008) and involves the motor system (see Barsalou, 2008; Fischer & Zwaan, 2008; Pulvermüller, 2005). In a neuroscientific context, this perspective also implies that brain areas related to action and language can no be longer be seen as independent, but rather working in concert. Areas traditionally regarded as pure motor areas as e.g. the primary motor or the premotor cortex, as well as areas that have traditionally been assigned to the processing of language, e.g. Broca’s or Wernicke’s region, are not modularized, but rather provide the linkage of action and language (Pulvermüller, 2005). Participation of Broca’s region has already been revealed in different motor experiments, such as grasping experiments (Grafton, Arbib, Fadiga, & Rizzolatti, 1996), object manipulation (Binkofski et al., 1999) and action imitation (Rizzolatti & Arbib, 1998). Based on numerous results, it turns out, that Broca’s region is more than just representing a language processing area (Menz & Binkofski, 2008). In this short story on embodied language we want to review recent literature on language and embodiment – first along the main lines of argumentation in the ongoing debate, and second in an activation likelihood estimation (ALE) meta-analysis on reported neural activity patterns. 1.1. Somatotopic activation during processing of action words and sentences One of the most important issues in the ongoing debate about embodied cognition is the somatotopy of activation. If the hypothesis held true that language recruited the same sensorimotor areas as for action and interaction, motor areas should display the same somatotopy for processing language as for processing actions. Although the general involvement of premotor and motor cortices has been demonstrated repeatedly, the issue of somatotopy still needs to be clarified. There is strong evidence for a somatotopic activation of premotor cortices from studies with different techniques (fMRI, MEG, etc.). Tettamanti et al. (2005) investigated the underlying neural processes while presenting sentences expressing actions performed with the mouth, the hand or the foot. Specifically, hand action and related words were found activated in the left precentral gyrus, the posterior intraparietal sulcus and the left posterior inferior temporal area. In contrast, leg activity has been identified in the left dorsal premotor and left intraparietal sulcus, but located more dorsally and rostrally in relation to the parietal hand activities. In addition, detection of a bilateral pattern in the posterior cingulate shows clear distinction of activation in processing abstractness. Activity in Broca’s region has been detected detached from any effector-specificity, thus implying a special role in language processing. Summarized, the fMRI results presented in that study display activity in a frontal-parietal circuit with temporal participation in the left hemisphere. Furthermore, Pulvermüller (2005) recorded neurophysiological and behavioral responses to verbs referring to actions performed with the face, the arms and the legs. Using a lexical decision task, they found topographical differences in the brain activity patterns generated by the different verbs, starting 250 ms after word onset. Consequently, the English verbs ‘lick’, ‘pick’, and ‘kick’ engage different neural sites in a topographical pattern. Moreover, a near-simultaneous activity pattern in the inferior frontal gyrus and the superior temporal gyrus could be identified, which supports both speech production as well as word comprehension. Another study on somatotopic organization of the motor cortex is presented in Hauk, Johnsrude, and Pulvermüller (2004). The differentiation of arm-related and leg-related action, respectively action words revealed distinct patterns in the middle frontal and the precentral gyrus for arm actions and, on the contrary, activations in dorsal areas in left and midline of the pre-and postcentral gyrus and dorsal premotor cortex. Induced by a dissociation on word category processing, the cortical activity found in this study displays effector-dependent processing along the motor strip. In addition, EEG-recordings showed an activation of the effector-specific motor regions occurring quite early, less than 200 ms after word onset (Pulvermüller, Lutzenberger, & Preissl, 1999). A combined behavioral and TMS study by Buccino et al. (2005) strengthens this suggestion. A decrease in amplitude of MEPs was recorded from hand muscles when listening to hand-action related sentences, and from foot muscles when listening to foot related sentences. In line with this evidence, further results obtained with behavioral tasks (Borghi & Scorolli, 2009; Scorolli & Borghi, 2007) suggest that the simulation activated during combinations of nouns and verbs is sensitive to the congruency between the effector implied by the sentence (e.g. mouth vs. foot) and the effector used to produce the motor response. Also other studies are able to identify body-part specificity in premotor cortices, but not in other motor areas. In an fMRI study with a lexical decision task Willems, Hagoort, and Casasanto (2010) found a preferential activation of the left premotor cortex for right handers, and of the right premotor cortex for left handers, while responding to manual-action verbs (compared to nonmanual action verbs). This suggests that the simulation evoked during language processing is body-specific. However, whereas imagery activated both motor and premotor cortices in a differential way, language comprehension activated only the premotor cortex. In a similar vein, Tomasino, Werner, Weiss, and Fink (2007, 2008) found activation of the primary motor cortex during explicit mental motor imagery, whereas no activation was found in a letterdetection task. Although several studies showed evidence towards action word comprehension in connection with somatotopy, a more critical view concerning this topic is discussed in Postle, McMahon, Ashton, Meredith, and de Zubicaray (2008).
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
713
Focussing on different motor areas Postle et al. (2008) used fMRI in cytoarchitectonically regions of interest (primary and premotor cortices) to compare action verbs related to different effectors (hand, foot and mouth) with other concrete nouns unrelated to body parts and actions, as well as to non-words and to strings of hashes. Action and non-action words were matched for imageability. Whereas an expected somatotopic organization for observation of simple movements could be identified, i.e. activations of motor areas BA 4 and BA 6 in a ventral-to-dorsal fashion according to the succession mouth, hand and foot. The same applies to a posterior-to-anterior pattern across the lateral surfaces of the mentioned motor areas. Although there was no evidence of a somatotopic organization for action words, the pre-supplementary motor area (preSMA) displayed a different activation for foot-related action compared to non-action words, thereby possibly playing a rather cognitive-motor role instead of a pure motor one. A main difference of Postle et al. in comparison to the other studies cited above is the use of cytoarchitectonically defined probability maps. This suggests that studies on somatotopy connected to word meaning extraction should be also related to cytoarchitectural information and functional criteria, in order to correctly interpret activation distribution as somatotopy. Summarized, several language comprehension studies show evidence on an at least effector-specific activation during language comprehension, though not primarily investigated in all studies introduced above. In general, but also considering some critical aspects by Postle et al. (2008), it can be stated, that somatotopy can be identified in premotor regions, but not consistently in primary motor cortices. This might be due to the complexity of tasks, but also to the different roles the motor cortices play in conceptualizing and execution of actions. 1.2. Embodiment and the mirror neuron system Along with a somatotopy in processing language, also the involvement of the mirror neuron system in processing language is based on the theoretical principle, that the processing of language is grounded in the same neural units as the actions the words refer to. According to embodied theories, canonical and mirror neurons represent the neural basis of the simulation activated during language comprehension (Gallese, 2008). Mirror neurons, which were originally identified in the ventral premotor cortex of monkeys, are not only firing during active motion, but they are also triggered by observing a conspecific performing actions with objects (Rizzolatti & Craighero, 2004). Similar investigations in humans lack the single-neuron resolution, thus it is rather referred to as the mirror neuron system (MNS; Buccino et al., 2001, 2005). The linking element between monkey and man is the area F5 of the monkey ventral premotor cortex which is regarded as a homolog to Broca’s region, the inferior frontal region in the human cortex, which is primarily known as a speech processing area (Rizzolatti & Craighero, 2004). This leads to the assumption, that homolog to F5, also Broca’s region contains mirror neurons (Buccino et al., 2005) and Broca’s region is no longer regarded as a pure language area, but as also as a region linking action and language (Binkofski & Buccino, 2004; Menz & Binkofski, 2008). As Broca’s region is the core region of the MNS and implies an additional link between motor processing and speech, the whole MNS possibly also has an impact on language processing and comprehension. Glenberg et al. (2008) show mirror neuron activation in an experimental setting by either presenting a typical action sound or a verbal action description. In differential contrasts of typically MNS-activating tasks, as e.g. object observation compared to language processing, the activation of Broca’s region to either seems to differ. Aziz-Zadeh, Koski, Zaidel, Mazziotta, and Iacoboni (2006), Aziz-Zadeh, Wilson, Rizzolatti, and Iacoboni (2006) and Aziz-Zadeh and Damasio (2008) describe close but not completely overlapping patterns for action observation and reading phrases. They draw the conclusion that mirror neurons are not directly mediating the understanding of language, but possibly play an important role as a precursor in the development of language (Rizzolatti & Arbib, 1998). However, this does not contradict an integrating role of Broca’s region in processing both sounds and actions. In monkey, Kohler et al. (2002) detected firing of mirror neurons in the presence of action specific noise. Mirror neurons showed a specialization for an action and the sound it produces, e.g. they fire when breaking a peanut and also when only the sound is played. In the human domain, D’Ausilio et al. (2009) reported a facilitation of the perception of a given speech sound when the motor articulator responsible for that sound was stimulated with TMS for motor cortex controlling. Apart from Broca’s region, the left inferior parietal lobe (IPL), also part of the MNS (Buccino et al., 2005) was reported to have an important role in the integration of sounds and actions. McNamara et al. (2008) asked subjects to learn associations between previously unrelated novel sounds and meaningless gestures. Both IPL and Broca’s region showed a strong, bilateral, negative correlation of BOLD response with learning of sound–action associations during data acquisition. Together with decrease due to the sharpening of the network, connectivity between the areas increased and the strongest learning related connectivity between regions was found in Broca’s region and left IPL. This leads to the conclusion that the involvement of motor regions in language processing is closely linked to regions of the mirror neuron system, thus possibly relying on using mirror neurons to integrate sounds and actions or even to simulate in order to understand action words. However, this claim is strong and will need further evidence. 1.3. Abstract and concrete word processing A second strong claim of embodied theories relates to the grounding of abstract language. Embodied theories assume that abstract concepts, just like concrete ones, are grounded in the sensorimotor system. Within this general framework, at least three different explanations of abstraction have been proposed (see Glenberg et al., 2008). A key issue in the literature is to what extent concrete and abstract words (e.g., ‘‘bottle” vs. ‘‘truth”) are represented differently. However, starting from this
714
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
general assumption different explanations have been proposed, the most influential one is based on metaphors. Lakoff and Johnson (1980) proposed up to 50 metaphor schemes in everyday language. ‘‘Cancer finally caught him up” serves as example for personification, whereas ‘‘Your claims are indefensible” has yet another categorization. In line with the work, Casasanto and Boroditsky (2008) show connection of metaphorical language processing and experiences built on perception and action. The argumentation of metaphors contributing to access meaning and also appealing to the conceptual system, which is also involved by action experiences and thinking, speaks in favor of the embodiment perspective of language comprehension. According to very recent proposals, abstract words involve more emotional aspects than concrete ones (Vigliocco, 2009); in addition, abstract words rely more on linguistic and social information as their modality of acquisition employs more linguistic information as compared to concrete words (Borghi & Cimatti, 2009). According to the strong version of the embodied framework, both the literal and the more abstract meaning of language (e.g., ‘‘grasp” in the context of ‘‘grasping an apple” and in the context of ‘‘grasping a notion”) are processed in the same neural units. Thus, action words should be represented in the same sensory-motor areas as their simple motor analog. As an example, Aziz-Zadeh and Damasio (2008) proposed that the verb ‘to kick’ (literal) and ‘kick of the year’ (abstract) imply the same ‘kick’ motor representation. There is TMS evidence supporting this view. Glenberg and collaborators (2008) have shown that abstract transfer sentences (e.g., give some news) activate motor information exactly as concrete transfer sentences (e.g., give a pizza). A more specific description of the processing of abstract and concrete words is given by a study of Rüschemeyer, Brass, and Friederici (2007). They performed an fMRI study comparing verbs with motor meaning, such as ‘‘to beat”, and verbs with abstract meanings, such as ‘‘to guess”. Participants had to respond by pressing a key to pseudo-words, while no response for words was required. The authors found enhanced activation of motor verbs compared to abstract verbs in the posterior premotor, primary motor, primary and secondary somatosensory cortex; this activation was bilateral but was higher in the left hemisphere. However, they did not find any difference in the activation of frontal mirror neurons areas, the ventral premotor cortex, and in the inferior parietal lobule for processing simple motor verbs and abstract verbs. In addition, no difference was found while comparing German verbs with motor stems and verbs with abstract stems. Tettamanti et al. (2005), however, describe a premotor activation during the processing of action-related sentences as compared to their control condition (e.g. ‘‘now I appreciate loyalty”), thus reporting a unique activation of a motor area to concrete sentences containing a manipulable object as opposed to sentences containing abstract objects. Another recent fMRI study, in turn suggests that the abstract words and sentences activate motor representations (Raposo, Moss, Stamatakis, & Tyler, 2009). The authors compared single verbs and literal and idiomatic sentences in order to verify whether involvement of motor regions is automatic and invariable or whether it is modulated by the sentential context. In the task they used, particularly adequate in case of ambiguous sentences, participants listened to sentences; on half of them they were presented with a visual probe and had to determine by pressing a key whether the visual probe was related or not to the sentence meaning. Listening to leg-related and arm-related action verbs (e.g., grab, kick) activated a fronto-parietal system typically involved in action execution. This leads to two assumptions. First, verbs and nouns are possibly processed differently regarding their abstractness, and second, that different levels of derivation from a word’s literal meaning might lead to different activations. Hence, abstract (‘‘to kick around an idea”), metaphorical (‘‘to kick in the dugout”), idiomatic (‘‘to kick the bucket”), and morphological (rare in English but in German ‘‘treten/to kick” and ‘‘eintreten”/to occur”) should be investigated separately and not be subsumed under the term ‘‘abstract” or ‘‘non-motor” or even be regarded as homogeneous control conditions. Altogether we can only conclude that demonstrations of the sensorimotor grounding of abstract words have so far been confined to rather specific domains and further evidence for grounding of abstract language contents is needed (for further discussion of this issue, see Borghi & Cimatti, 2009). 1.4. Are sensorimotor areas essential for language comprehension? One core issue in discussing embodied language lies in the question whether the involvement of sensorimotor areas is auxiliary, concomitant, or necessary for language processing and comprehension. Even though the majority of studies demonstrate that the motor system is activated during words and sentences processing, there is some controversial evidence, and some issues remain open (see also Willems & Hagoort, 2007). Partly this might be related to the afore mentioned disagreement as to the definition of abstractness and the usage of control conditions. However there are two points of view contributing to an essential role of sensorimotor areas to language processing: evidence from patients and evidence on timing. The first promising evidence is given by studies on patients with impairments of the motor system. Boulenger et al. (2008) found no priming effect for action verbs for patients affected by Parkinson disease off dopaminergic treatment, i.e. when there is no normal activation level in premotor and motor areas. The priming effect, instead, was present in both controls and Parkinson patients after dopaminergic intake. This study provides strong evidence that the integrity of the motor system is necessary for verb processing. Along with this evidence, Bak et al. (2006) found selective deficits in verb processing in two patients, father and son, with a familial motor disorder; in addition, Bak and Hodges (2004) found a selective difficulty for verb processing in motor neuron disease, a neurodegenerative disease of the motor system. Even if this evidence is not conclusive, a number of studies report action comprehension deficits in patients with premotor and parietal lesions (for a review and discussion, see Aziz-Zadeh & Damasio, 2008). Altogether, it can be claimed that lesions of the motor system selectively impair language processing, and particularly verb comprehension. This evidence can support the argument that an integer motor system might be part of the comprehension process.
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
715
The second good argument in favor of the involvement of the motor system as an essential part of language comprehension would be an early activation of the motor system. Papeo, Vallesi, Isaja, and Rumiati (2009) recorded TMS-induced motor-evoked potentials from right hand muscles in order to measure M1 activity during comprehension of action verbs. They found an increase of M1-activity only at 500 ms, while no increase was present as they delivered single pulse TMS at 170 and 350 ms after action words appearance. This suggests that M1 is involved during post-conceptual processing of action words, and it is not necessarily implied and does not contribute to words comprehension. However, other studies report an early involvement of motor areas (for a review and a model reproducing results on both early and late activation of motor areas, see Chersi, Thill, Ziemke, & Borghi, 2010). In a lexical decision task, presenting action verbs and nouns, Pulvermüller et al. (1999) detected neural activity recorded from EEG less than 200 ms after word onset and in an automatic way. Even in this short time window, the signals highlighted electrocortical differences. As discussed by the authors, the study revealed nonclassical language areas involved in language processing, and proposed additional cell assemblies in the motor cortex for action verbs and neural signals from the visual cortex for nouns, respectively. Thus, we think that evidence collected so far is promising and that in the near future the issue of the necessity of activation of the motor system concerning semantic processing will be definitively solved. 2. Embodiment – a meta-analysis Opposed to narrative reviews or label-based anatomical approaches, the coordinate-based meta-analysis methods statistically aggregate activation foci (‘peaks’) derived from neuroimaging data and emphasize specific neuronal patterns across multiple studies following a common paradigm or hypothesis. Recent experimental series can be evaluated to a meta-statement. For the meta-analysis included in this review we used the ALE approach as implemented in the Ginger ALE software provided by BrainMap (Laird et al., 2005). 2.1. Literature search and criteria An exhaustive literature search was conducted on fMRI studies indexed in the Medline database. We focused on studies, which contained the pre-specified keywords ‘‘embodiment”, ‘‘language”, ‘‘motor”, ‘‘action”, and ‘‘perception”. The filter criteria do not distinguish between concrete and abstract word processing, as the objective of the analysis is to cover the whole range of embodied language. In addition, the selection of experiments took into account both extremities (hand/arm, foot/ leg) to increase the variance of particular action-related patterns in literal and abstract meanings. Furthermore, we included studies on words (verbs, nouns) and on sentences or both. 2.2. Statistical procedure A statistical map was generated by using a collection of 468 foci from the 21 studies reported in Table 1 after transferring them into Talairach space (Talairach & Tournoux, 1988). In order to account for the uncertainty, that is technically inherent to the actual location of the peaks, each coordinate was modeled not as a single point, but by a three-dimensional (3D) Gaussian function with 12 mm FWHM. Thus, the localization probability distributions describe the probability that a given focus actually lay within a particular voxel. Statistical significance is gained via a permutation test of randomly generated foci using the same FWHM and number of foci. The voxel-wise comparison is tested against the null-hypothesis of uniformly distributed peaks, giving a set of ALE-values necessary for thresholding the probability map. Using the False Discovery Rate (FDR) with q = .01, the test was corrected for multiple comparisons. 2.3. Results of the meta-analysis The activation clusters of the meta-analysis are summarized in Table 2 and Fig. 1. All coordinates are in Talairach space and anatomical labels as well as Brodmann areas were obtained with the Talairach Daemon (Lancaster et al., 1997, 2000). Major activity sites are displayed in the left hemisphere, predominantly in the frontal lobe, comprising the inferior frontal gyrus (cluster 1, BA 44, BA 46) and the precentral gyrus (cluster 1, cluster 3, cluster 9, BA 4, BA 6). In the parietal lobe, distinct clusters could be detected in the left supramarginal gyrus (cluster 1, BA 40), as well as in the right superior parietal lobulus (cluster 5, BA 7) and in the left precuneus (cluster 1, BA 19) area. In the left temporal lobe, activations could be found in the middle temporal gyrus (cluster 2, BA 22, BA 39) and more prominently in the fusiform area (cluster 8, BA 37). Further findings include the insula (cluster 2, BA 13) in the left hemisphere as well as the bilateral cerebellum. Additionally, the analysis reveals two clusters comprising the posterior cingulate (cluster 6, cluster 11, BA 30) in both hemispheres. 2.4. Integrating the meta-analysis into previous findings The main advantage of the ALE meta-analysis is to give an overview on previously reported findings, to un-weight interpretations, and to re-weight results within, but even more outside of regions of interest. Due to the method, there
716
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
Table 1 Publications included in the meta-analysis, task they employed, number of subjects that were investigated and number of selected foci for the ALE metaanalysis. Study
Task
N
Foci
Bedny et al. (2008) Boulenger et al. (2008) Canessa et al. (2008)
Action verb comprehension Arm- and leg-related action verbs Semantic decisions on picture pairs of (non-)manipulable objects (e.g. tools) Nouns (objects), verbs (action) Classification of real (e.g. tools) and nonsense objects (e.g. artificial shapes) Arm- and leg-relates action words Hand- and leg-related action verbs Action-/object-naming, verbs or nouns Responses on hand manipulation and phonological word decisions Hand- and foot-related action verbs Arm- and leg-related action verbs Lexical decision between functional and volumetric manipulability (by hand) Go/NoGo tasks to words stimuli Motor- and abstract verbs/wordstems Action- and object naming (verbs, nouns),hand- and leg-related Action- and object naming (verbs, nouns) Word reading and generation from semantic categories Hand- and foot related sentences Action-related and abstract sentences, supplemental data to Tettamanti (2005) Motor and non-motor verbs, hand related Word decision on concrete hand actions
12 18 15
6 43 35
17 20
5 48
14 16 15 14 18 22 15
15 45 62 25 6 5 5
19 20 13 12 12 17 18
7 19 25 6 18 14 10
15 20
29 40
Gennari, Mac Donald, Postle, and Seidenberg (2007) Gronau et al. (2008) Hauk et al. (2004) Kemmerer et al. (2008) Liljeström et al. (2008) Meister and Iacoboni (2007) Postle et al. (2008) Raposo et al. (2008) Rüschemeyer, von Rooij, Lindemann, Willems, and Bekkering (2009) Rüschemeyer, Pfeiffer, and Bekkering (2010) Rüschemeyer et al. (2007) Saccuman et al. (2006) Siri et al. (2007) Tremblay and Gracco (2006) Tettamanti et al. (2005) Tettamanti et al. (2008) Tomasino et al. (2007) Willems, Ozyürek, and Hagoort (2009)
Table 2 Results from the ALE meta-analysis. Clusters of activation connected above threshold, activation sites, Talairach-Coordinates (x, y, z) of maximum ALE-value, and maximum ALE-value of this cluster. Cluster 1
2
3 4 5 6 7 8 9 10 11
Area L. Inferior frontal gyrus (BA 44) L. Inferior frontal gyrus (BA 45) L. Precentral gyrus (BA 6) L. Supramarginal gyrus (BA 40) L. Postcentral (BA 3) L. Precuneus (BA 19) L. Precentral gyrus (BA 4) L. Insula (BA13) L. Middle frontal gyrus (BA 6) L. Middle temporal gyrus (BA 22) L. Middle temporal gyrus (BA 39) L. Insula (BA 13) L Medial frontal gyrus (BA 6) L. Cerebellum anterior lobe R. Superior parietal lobe (BA 7) L. Posterior cingulate (BA 30) L. Middle temporal gyrus (BA 30) L. Fusiform gyrus (BA 37) R. Precentral gyrus (BA 4) R. Fusiform gyrus (BA 37) R. Posterior cingulate (BA 30)
x
y 46 46 40 44 46 28 52 32 26 52 50 54 4 36 26 20 52 46 40 38 16
z 12 26 6 38 20 66 10 22 6 40 58 32 0 42 64 62 4 56 16 48 64
20 8 48 36 40 42 24 4 58 2 6 18 54 22 42 4 10 14 36 18 10
Cluster size
ALEmax
32,144
0.0310 0.0239 0.0199 0.0187 0.0186 0.0168 0.0165 0.0148 0.0127 0.0268 0.0204 0.0120 0.0341 0.0515 0.0165 0.0145 0.0162 0.0143 0.0140 0.0112 0.0123
8856
5760 1224 1072 848 816 736 312 136 136
will be no new evidence, however, the focus on activated areas might change and a more general pattern can be identified. As described in the first section of this short review, embodied cognition theories propose several ideas. The first important one is the assumption that there is no separation between low and high cognitive processes. This assumption is tightly linked to the second claim, that sensorimotor systems are recruited, when verbal material is processed. Indeed, the main finding from this meta-analysis shows the clear involvement of a variety of regions, including mainly temporal (cluster 1, 2, 8) and frontal (cluster 1, 3), but also cerebellar activity (cluster 4, 10). Moreover, there is a clear predominance of activations in the (language and motor areas of the) left hemisphere. This could be due to a variety of reasons. First, language processing naturally occurs in the left hemisphere. In addition, the majority of participants in the present studies had right dominant effectors, hand and foot, which are processed contra-laterally. In
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
717
Fig. 1. ALE results of the meta-analysis. Images display maximum ALE-values thresholded at p < .01 (FDR corrected). Left column displays 3D surface renderings from posterior-left, left, frontal, and right viewpoints. Lightbox images illustrate sections at x, y ,z = 38, 42, 20, x, y ,z = 38, 10, 26, x, y ,z = 4, 5, 56, and x, y ,z = 42, 16, 36 (from top to bottom).
an fMRI study with a lexical decision task Willems et al. (2010) found a preferential activation of the left premotor cortex for right handers, and of the right premotor cortex for left handers, while responding to manual-action verbs (compared to nonmanual action verbs). However, it should be noted that during the experiments, participants were not performing actions. Rather, they had to read, listen and name actions without concomitant motor activity. Hence, it is important to emphasize that the motor areas such as SMA, the precentral gyrus or the premotor cortex were active when language only was processed. The meta-analysis also detects neural activations in the right hemisphere, especially in frontal (cluster 9) and parietal (cluster 5) regions. Indeed, there is evidence that the right hemisphere contributes to the processing of semantic components (Canessa et al., 2008; Gronau, Neta, & Bar, 2008; Kemmerer, Castillo, Talavage, Patterson, & Wiley, 2008). Another interesting result speaking in favor of motor activation irrespective of a distinction between low and high cognitive processes is the here identified participation of the cerebellum in language processing and, consequently, in language comprehension. This region has not yet been extensively focused on in the context of language, although it is well known that is plays a crucial role in motor learning. However, our results show a significant contribution of the left cerebellar hemisphere to language processing. The neural connections in participation with the cerebellum could be another interesting aspect pointing to the neural interoperability of different brain areas in embodied language. By relocating this result to the studies in the literature collection (Table 2) it becomes evident, that the cerebellum activity could be recorded within experiments which investigate action comprehension (Boulenger, Hauk, & Pulvermüller, 2009), action naming (Liljeström et al., 2008) or semantic processing (Saccuman et al., 2006). Therefore, the activation of the cerebellum suggests that language is embodied not only because sensorimotor areas are active, but also because words are processed along a frontal-parietal-temporal network including (clusters 1, 2) subcortical activity. An explanation for the distributed character of activity might lie in the distinction between the concepts retrieved from semantic knowledge and the perceptual component of words as hypothesized in Bedny, Caramazza, Grossman, Pascual-Leone, and Saxe (2008). The bilateral activity in the temporal lobe, namely the fusiform gyri, suggest that the posterior part of the temporal lobe is an area organizing concepts, rather than visual properties of words, which is in line with findings from Hauk and Pulvermüller (2004). Also Rüschemeyer et al. (2007) highlight the role of the right temporal area processing rather complex verbs with abstract meanings. Although they propose further work on that topic, our work provides supporting results for the implication of temporal areas and conceptualization or categorization.
718
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
Only partly reflected in this meta-analysis are the two proposals of embodied cognition theories, namely the somatotopic activation during the processing of action words and sentences, as well as the grounding of abstract concepts in the sensorimotor system. Neither has been investigated per se, as the number of included studies did not allow for a differential contrast between different effectors. 3. Concluding remarks This short review was conceptualized to give an overview along the lines of argumentation in the ongoing debate on the embodiment of language. We further employed an ALE meta-analysis to illustrate and relate previous findings. Both the narrative review, and the meta-analysis confirm the connection of language and motor areas. Primary motor, supplementary motor and premotor cortices are repeatedly reported to be active during language processing, and they are significantly present in the ALE results. If the hypothesis held true that the same neural units are responsible for action and simulation of action during language processing, classical characteristics of motor activation should be shared by language comprehension. Namely, language should produce the same somatotopy in activation as real action does. There is growing evidence supporting this view, especially for premotor areas. However, there is not enough data to validate this claim in a meta-analysis, which could theoretically be of immense help clarifying this issue. The idea, that the mirror neuron system plays an important role in the embodiment of language, is backed by a variety of findings in recent publications. Additionally, a set of peak points in the meta-analysis are very close to the locations reported for the mirror neuron system (Buccino et al., 2001, 2005). However, the strong claim of embodied theories, that mirror neurons represent the neural basis of the simulation activated during language comprehension, needs further thorough investigation. A region of interest in this can be Broca’s region, as this seems to have a core role in the integration of motion and sound. The claim of embodied theories is the grounding of abstract language in sensorimotor areas. Currently the findings on abstract words and sentence processing provide inconsistent results which are possibly due to the varieties in the definition of abstractness, but also to the variety of stimuli used in control conditions. Although our meta-analysis is not able to further enlighten this debate, the regions identified would be well suitable as regions of interest for further analyses on the processing of abstract language. The strongest results in all lines of argumentation are provided by the findings from patients and timing, both of which argue in favor of a necessary instead of an auxiliary role of sensorimotor areas in language processing. This very fast activation, its automaticity, taken together with the likely somatotopic organization render the hypothesis advanced among others by Mahon and Caramazza (2008), that information is first transduced in an abstract format and then influences the motor system, rather unlikely. The hypothesis that the motor system is activated in a direct and straightforward way is much more plausible and economical, even if evidence on timing and somatotopy still leave some unsolved issues. Finally, it should be noted that the discussion on embodiment should take into account in a sufficient way the strong plasticity and distributed character of the human brain. Consider some of the results we discussed. Even if abstract words are not represented in the same motor areas as concrete words, this would not necessarily render a problem. For example, it is possible that abstract words, due to the fact that they do not have a concrete referent, activate more language related areas (for a discussion, see Borghi & Cimatti, 2009; submitted). Even if the activation during language processing pertains the premotor cortices and not the primary motor ones, this would not undermine the embodied hypotheses. If we found that the activation of explicit imagery differs from that elicited by language comprehension, this result would even strengthen the embodied hypothesis. If some patients preserve their ability to comprehend language despite their motor disabilities, as it might happen with apractic patients, this might suggest that the brain is distributed and plastic enough. Acknowledgments This work was supported by the FP7 project ROSSI, Emergence of communication in Robots through Sensorimotor and Social Interaction, Grant Agreement No: 216125 and by German Ministry of Research BMBF (01GW0571 and 01GW0752). We thank the anonymous reviewers for helpful comments on an earlier version of the manuscript. References Aziz-Zadeh, L., & Damasio, A. (2008). Embodied semantics for actions: Findings from functional brain imaging. Journal de Physiologie – Paris, 102(1–3), 35–39. Aziz-Zadeh, L., Koski, L., Zaidel, E., Mazziotta, J., & Iacoboni, M. (2006). Lateralization of the human mirror neuron system. Journal of Neuroscience, 26(11), 2964–2970. Aziz-Zadeh, L., Wilson, S. M., Rizzolatti, G., & Iacoboni, M. (2006). Congruent embodied representations for visually presented actions and linguistic phrases describing actions. Current Biology, 16(18), 1818–1823. Bak, T. H., & Hodges, J. R. (2004). The effects of motor neurone disease on language: Further evidence. Brain and Language, 89(2), 354–361. Bak, T. H., Yancopoulou, D., Nestor, P. J., Xuereb, J. H., Spillantini, M. G., & Pulvermüller, F. (2006). Clinical, imaging and pathological correlates of a hereditary deficit in verb and action processing. Brain, 129, 321–332. Barsalou, L. W. (2008). Grounded cognition. Annual Review of Psychology, 59, 617–645.
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
719
Bedny, M., Caramazza, A., Grossman, E., Pascual-Leone, A., & Saxe, R. (2008). Concepts are more than percepts: The case of action verbs. Journal of Neuroscience, 28(44), 11347–11353. Binkofski, F., & Buccino, G. (2004). Motor functions of the Broca’s region. Brain and Language, 89, 362–369. Binkofski, F., Buccino, G., Posse, S., Seitz, R. J., Rizzolatti, G., & Freund, H. (1999). A fronto-parietal circuit for object manipulation in man: Evidence from an fMRI-study. European Journal of Neuroscience, 11(9), 3276–3286. Borghi, A. M., & Cimatti, F. (2009). Words as tools and the problem of abstract words meanings. In N. Taatgen & H. van Rijn (Eds.), Proceedings of the 31st annual conference of the cognitive science society (pp. 2304–2309). Amsterdam: Cognitive Science Society. Borghi, A. M., & Scorolli, C. (2009). Language comprehension and hand motion simulation. Human Movement Science, 28, 12–27. Boulenger, V., Hauk, O., & Pulvermüller, F. (2009). Grasping ideas with the motor system: Semantic somatotopy in idiom comprehension. Cerebral Cortex, 19(8), 1905–1914. Boulenger, V., Mechtouff, L., Thobois, S., Broussolle, E., Jeannerod, M., & Nazir, T. A. (2008). Word processing in Parkinson’s disease is impaired for action verbs but not for concrete nouns. Neuropsychologia, 46(2), 743–756. Buccino, G., Binkofski, F., Fink, G. R., Fadiga, L., Gogassi, L., Gallese, V., et al (2001). Action observation activates premotor and parietal areas in a somatotopic manner: An fMRI study. European Journal of Neuroscience, 13, 400–404. Buccino, G., Riggio, L., Melli, G., Binkofski, F., Gallese, V., & Rizzolatti, G. (2005). Listening to action-related sentences modulates the activity of the motor system: A combined TMS and behavioural study. Brain Research, Cognitive Brain Research, 24(3), 355–363. Canessa, N., Borgo, F., Cappa, S. F., Perani, D., Falini, A., Buccino, G., et al (2008). The different neural correlates of action and functional knowledge in semantic memory: An FMRI study. Cerebral Cortex, 18(4), 740–751. Casasanto, D., & Boroditsky, L. (2008). Time in the mind: Using space to think about time. Cognition, 106, 579–593. Chersi, F., Thill, S., Ziemke, T., & Borghi, A. M. (2010). Sentence processing: Linking language to motor chains. Frontiers in Neurorobotics. D’Ausilio, A., Pulvermüller, F., Salmas, P., Bufalari, I., Begliomini, C., & Fadiga, L. (2009). The motor somatotopy of speech perception. Current Biology, 19(5), 381–385. Fischer, M., & Zwaan, R. (2008). Embodied language: A review of the role of the motor system in language comprehension. The Quarterly Journal of Experimental Psychology, 61, 825–850. Gallese, V. (2008). Mirror neurons and the social nature of language: The neural exploitation hypothesis. Social Neuroscience, 3, 317–333. Gennari, S. P., Mac Donald, M. C., Postle, B. R., & Seidenberg, M. S. (2007). Context-dependent interpretation of words: Evidence for interactive neural processes. Neuroimage, 35, 1278–1286. Glenberg, A. M., Sato, M., Cattaneo, L., Riggio, L., Palumbo, D., & Buccino, G. (2008). Processing abstract language modulates motor system activity. The Quarterly Journal of Experimental Psychology, 61, 905–919. Grafton, S. T., Arbib, M. A., Fadiga, L., & Rizzolatti, G. (1996). Localization of grasp representations in humans by positron emission tomography. 2. Observation compared with imagination. Experimental Brain Research, 112(1), 103–111. Gronau, N., Neta, M., & Bar, M. (2008). Integrated contextual representation for objects’ identities and their locations. Journal of Cognitive Neuroscience, 20(3), 371–388. Hauk, O., Johnsrude, I., & Pulvermüller, F. (2004). Somatotopic representation of action words in human motor and premotor cortex. Neuron, 41(2), 301–307. Hauk, O., & Pulvermüller, F. (2004). Neurophysiological distinction of action words in the fronto-central cortex. Human Brain Mapping, 2, 1191–1201. Kemmerer, D., Castillo, J. G., Talavage, T., Patterson, S., & Wiley, C. (2008). Neuroanatomical distribution of five semantic components of verbs: Evidence from fMRI. Brain and Language, 107(1), 16–43. Kohler, E., Keysers, C., Umilta, M. A., Fogassi, L., Gallese, V., & Rizzolatti, G. (2002). Hearing sounds, understanding actions: Action representation in mirror neurons. Science, 297(5582), 846–848. Laird, A. R., Fox, P. M., Price, C. J., Glahn, D. C., Uecker, A. M., Lancaster, J. L., et al (2005). ALE meta-analysis: Controlling the false discovery rate and performing statistical contrasts. Human Brain Mapping, 25(1), 155–164. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago Press. Lancaster, J. L., Rainey, L. H., Summerlin, J. L., Freitas, C. S., Fox, P. T., Evans, A. C., et al (1997). Automated labeling of the human brain: A preliminary report on the development and evaluation of a forward-transform method. Human Brain Mapping, 5, 238–242. Lancaster, J. L., Woldorff, M. G., Parsons, L. M., Liotti, M., Freitas, C. S., Rainey, L., et al (2000). Automated Talairach Atlas labels for functional brain mapping. Human Brain Mapping, 10, 120–131. Liljeström, M., Tarkiainen, A., Parviainen, T., Kujala, J., Numminen, J., Hiltunen, J., et al (2008). Perceiving and naming actions and objects. Neuroimage, 41(3), 1132–1141. Mahon, B. Z., & Caramazza, A. (2008). A critical look at the embodied cognition hypothesis and a new proposal for grounding conceptual content. Journal de Physiologie – Paris, 102(1–3), 59–70. McNamara, A., Buccino, G., Menz, M. M., Gläscher, J., Wolbers, T., Baumgärtner, A., et al (2008). Neural dynamics of learning sound–action associations. PLoS ONE, 3(12), e3845. Meister, I. G., & Iacoboni, M. (2007). No language-specific activation during linguistic processing of observed actions. PLoS ONE, 9, e891. Menz, M. M., & Binkofski, F. (2008). Broca’s region beyond language. The omnifariousness of Broca’s region. Rivista Medica, 14, 51–55. Nolfi, S., & Floreano, D. (2000). Evolutionary robotics. MIT Press. Papeo, L., Vallesi, Antonino, Isaja, Alessio, & Rumiati, Raffaella Ida (2009). Effects of TMS on different stages of motor and non-motor verb processing in the primary motor cortex. PLoS ONE, 4(2), e4508. Postle, N., McMahon, K. L., Ashton, R., Meredith, M., & de Zubicaray, G. I. (2008). Action word meaning representations in cytoarchitectonically defined primary and premotor cortices. Neuroimage, 43, 634–644. Pulvermüller, F. (2005). Brain mechanisms linking language and action. Nature Reviews Neuroscience, 6(7), 576–582. Pulvermüller, F., Lutzenberger, W., & Preissl, H. (1999). Nouns and verbs in the intact brain: Evidence from event-related potentials and high-frequency cortical responses. Cerebral Cortex, 9, 498–508. Raposo, A., Moss, H. E., Stamatakis, E. A., & Tyler, L. K. (2009). Modulation of motor and premotor cortices by actions, action words and action sentences. Neuropsychologia, 47, 388–396. Rizzolatti, G., & Arbib, M. A. (1998). Language within our grasp. Trends in Neurosciences, 21(5), 188–194. Rizzolatti, G., & Craighero, L. (2004). The mirror–neuron system. Annual Review of Neuroscience, 27, 169–192. Rüschemeyer, S.-A., Brass, M., & Friederici, A. D. (2007). Comprehending prehending: Neural correlates of processing verbs with motor stems. Journal of Cognitive Neuroscience, 19, 855–865. Rüschemeyer, S.-A., Pfeiffer, C., & Bekkering, H. (2010). Body schematics: On the role of the body schema in embodied lexical-semantic representations. Neuropsychologia, 48, 774–781. Rüschemeyer, S.-A., von Rooij, D., Lindemann, O., Willems, R., & Bekkering, H. (2009). The function of words: Distinct neural correlates for words denoting differently manipulable objects. Journal of Cognitive Neuroscience, 1–8. Saccuman, C. M., Cappa, S. F., Bates, E. A., Arevalo, A., Della Rosa, P., Danna, M., et al (2006). The impact of semantic reference on word class: An fMRI stud of action and object naming. Neuroimage, 32, 1865–1878. Scorolli, C., & Borghi, A. M. (2007). Sentence comprehension and action: Effector specific modulation of the motor system. Brain Research, 1130, 119–124. Siri, S., Tettamanti, M., Cappa, S. F., Della Rosa, P., Saccuman, C., Scifo, P., et al (2007). The neural substrate of naming events: Effects of processing demands but not of grammatical class. Cerebral Cortex, 171–177. Talairach, J., & Tournoux, P. (1988). A stereotactic coplanar atlas of the human brain. New York: Thieme.
720
D. Jirak et al. / Consciousness and Cognition 19 (2010) 711–720
Tettamanti, M., Buccino, G., Saccuman, M. C., Gallese, V., Danna, M., Scifo, P., et al (2005). Listening to action-related sentences activates fronto-parietal motor circuits. Journal of Cognitive Neuroscience, 17(2), 273–281. Tettamanti, M., Manenti, R., Della Rosa, P. A., Falini, A., Perani, D., Cappa, S. F., et al (2008). Negation in the brain: Modulating action representations. Neuroimage, 43, 358–367. Tomasino, B., Fink, G. R., Sparing, B., Dafotakis, M., & Weiss, P. H. (2008). Action verbs and the primary motor cortex: A comparative TMS study on silent reading, frequency judgment, and motor imagery. Neuropsychologia, 48(7), 1915–1928. Tomasino, B., Werner, C. J., Weiss, P. H., & Fink, G. R. (2007). Stimulus properties matter more than perspective: An fMRI study of mental imagery and silent reading of action phrases. Neuroimage, 36, T128–T141. Tremblay, P., & Gracco, V. L. (2006). Contribution of the frontal lobe to externally and internally specified verbal responses: fMRI evidence. Neuroimage, 33, 947–957. Vigliocco, G. (2009). Happiness is an abstract word. In N. Taatgen & H. van Rijn (Eds.), Proceedings of the 31st annual conference of the cognitive science society (pp. 1115–1120). Amsterdam: Cognitive Science Society. Willems, R. M., & Hagoort, P. (2007). Neural evidence for the interplay between language, gesture, and action: A review. Brain and Language, 101, 278–289. Willems, R. M., Hagoort, P., & Casasanto, D. (2010). Body-specific representations of action verbs: Neural evidence from right- and left-handers. Psychological Science, 21, 67–74. Willems, R. M., Ozyürek, A., & Hagoort, P. (2009). Differential roles for left inferior frontal and superior temporal cortex in multimodal integration of action and language. Neuroimage 1.47(4), 1992–2004. Ziemke, T. (Ed.). (2002). Special issue on Situated and embodied cognition. Cognitive Systems Research, 3, 3. Zwaan, R. (2004). The immersed experiencer: Toward an embodied theory of language comprehension. In B. H. Ross (Ed.). Psychology of learning and motivation (Vol. 44, pp. 35–62). New York: Academic.
Consciousness and Cognition 19 (2010) 721–730
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
True lies: Self-stabilization without self-deception q Werner Greve a,*, Dirk Wentura b a b
University of Hildesheim, Germany Saarland University, Saarbrücken, Germany
a r t i c l e
i n f o
Article history: Available online 19 June 2010 Keywords: Self-deception Self-immunization Self-concept Self-stabilization
a b s t r a c t Self-deception entails apparent conceptual paradoxes and poses the dilemma between two competing needs: the need for stability of the self-concept, on the one hand, and the need to accept reality, on the other. It is argued, first, that conceptual difficulties can be avoided by distinguishing two levels of explanation. Whereas, in a personal language, ‘‘the person” deceives him- or her-self, a cognitive (‘‘subpersonal”) approach explains this self-deception by reference to the interplay of cognitive processes of which the person is not aware. Second, the tension between stability and adjustment of the self can be resolved by selfimmunization, which maintains the stability of central self-conceptions by adjusting peripheral aspects and their diagnostic value for the central concepts. Processes of selfimmunization were investigated in a series of studies operating on both levels of explanation. Implications for psychological explanations of personal phenomena such as selfimages and self-insight are discussed. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Whatever one’s self actually ‘‘is” – a structure or a process – it is a compromise between stability and change. On the one hand, we maintain a sense of being one and the same person throughout our whole lifespan, and we manage to remain the same individual with respect to many and various facets of our selves. Innumerable empirical findings – in fact, an entire discipline (i.e., personality psychology; e.g., Costa & McCrae, 1997; Funder, 2001, 2007) – support the notion that we as adults are highly stable in central dimensions of our personality, which is not only represented in but also operationalized through our self-conceptions (see Bengtson, Reedy, & Gordon, 1985). On the other hand, we need to adapt our selves over developmental trajectories. Many of our competencies and skills, including those of which we are proud, change significantly over the lifespan, often in a way we experience as a loss (Baltes, Lindenberger, & Staudinger, 2006). For action to be regulated successfully, the self-concept probably needs not be as accurate as possible, but certainly as realistic as necessary (Greve & Wentura, 2003). We therefore have to adjust our self-conceptions accordingly. In fact, many studies have shown that we can – and do – adjust our selves in response to even minor situational demands (Hannover, 2000; Hannover, Pöhlmann, Springer, & Roeder, 2005). This raises the question of how the demands of self-stabilization and self-adjustment can be reconciled. How do we manage to remain the same person, to keep our identity stable and consistent over time, while adjusting our selves to our everchanging existence? Ever since Anna Freud’s (1936) book on the ego and its defense mechanisms, self-defense has been considered the obvious solution to this dilemma. If we do not acknowledge the changes – and, in particular, the losses – that affect us, we are
q
This article is part of a speical issue of this journal on Self, Other and Memory. * Corresponding author. Address: University of Hildesheim, Institute of Psychology, Marienburger Platz 22, D-31141 Hildesheim, Germany. E-mail address:
[email protected] (W. Greve).
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.05.016
722
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
able to maintain our self-image. Many self-defense phenomena ranging from ‘‘perceptual defense” (e.g., Erdelyi, 1974) and ‘‘denial” (e.g., Breznitz, 1981; Lazarus & Golden, 1981) to self-serving attributions (e.g., Miller, 1976) have been investigated and discussed by several branches of psychology (e.g., Greenwald, 1980; Hilgard, 1949). The ‘‘psychology of the self” (e.g., Greve, 2000a; Leary & Tangney, 2003; Tesser, Felson, & Suls, 2000; Tesser, Stapel, & Wood, 2002) has yielded numerous and diverse findings indicating that self-relevant information is not simply stored or assimilated by the person, but rather actively and systematically processed, resulting in the high – and cumulatively increasing – stability of the adult self (for an overview, see Greve, 2005). The aims of this article are threefold. First, we propose self-immunization as a self-stabilizing process that is able to resolve the tension between the need to adjust to developmental changes and losses, on the one hand, and the need to stabilize the self, on the other. In a nutshell, self-immunization means maintaining the stability of central self-concepts by adjusting the diagnostic value of peripheral aspects for those central concepts. Second, we attempt to demonstrate that self-immunization can be empirically tested, despite the fact that the defense mechanisms of self-stabilization are, in principle, largely unknown to their ‘‘user.” Third, we explore whether self-immunization can resolve the apparent paradoxes of self-deception. 2. Intricacies of self-defense: the costs of ignoring realities The idea that we are good at deceiving ourselves about the degree of our changing involves at least two problems. The first has puzzled generations of philosophers (e.g., Cassam, 1994; Martin, 1986; Mele, 1987): ‘‘Self-deception is usually no great problem for its practitioner; on the contrary, it typically relieves a person of some of the burden of painful thought. . . . But self-deception is a problem for philosophical psychology” (Davidson, 1985, p. 79). At first glance, self-deception is an impossibility: If ‘‘deception” means that someone who knows a certain fact deceives someone else who does not know that fact, then self-deception seems logically impossible (Haight, 1980). Yet, because we do ‘‘deceive” ourselves, it cannot be impossible (Gur & Sackheim, 1979). The plausible solution to the ‘‘enigma of self-deception” (Baumeister, 1993) lies in the dissolution of the unity of ‘‘the self”: There is no single ‘‘ego” that acts, knows . . . or lies. Rather, the self is to be conceived as a complex and dynamic system (e.g., Markus & Wurf, 1987), consisting of highly differentiated cognitive and evaluative structures (‘‘Me”; see James, 1890; Mead, 1934) plus the processes operating on these structures (‘‘I”). These processes serve several functions (e.g., Filipp & Klauer, 1985) that can be grouped into two general categories: the reality principle and the pleasure principle (e.g., Greve, 2005; Higgins, 1997). According to the arguments sketched above, we humans need to be sufficiently realistic in order to act effectively. At the same time, we tend (or want) to be as consistent and positively evaluated as possible. Whereas the former requirement limits our inclination to view ourselves in a gloomy light, the latter is thought to be the reason for self-deception in the first place. Precisely this constitutes the second problem: self-deception comes at a price – it entails disregard of facts. Yet, as mentioned above, the self-concept needs to be sufficiently realistic to usefully regulate action (Dunning, Leuenberger, & Sherman, 1995; Taylor & Gollwitzer, 1995): If I act on false premises, I will usually fail; if I believe I can do something that I actually cannot, it usually results in embarrassment. How do we manage, then, to maintain a sense of self-stability? How does self-defense relate to the subjective intuitions and empirical findings that underscore the stability and continuity of the adult self? Because the stability of the self-concept must not be maintained at the price of a completely unrealistic self-image, self-stabilization cannot rely entirely on protective processes that ignore or modify threatening data (e.g., denial, repression). There must be other, more adaptive processes that ensure the stability of the self and personal identity across the lifespan, without completely disregarding reality. 3. Flexible resistance: self-immunization by peripheral adjustment The heterogeneous and large ‘‘zoo” (Tesser, Martin, & Cornell, 1996) of self-defense processes can be ordered in terms of the radicalism of their defensiveness. Application of a somewhat Procrustean systematization identifies three general ‘‘lines of defense” (Baumeister, 1996; Greve, 2000b). The first can be called the ‘‘rejecting” line of defense. This large category is characterized by the rejection of threatening events or information; it includes mechanisms of perceptual defense, denial, and repression (e.g., Markus & Wurf, 1987). The second can be termed the ‘‘neutralizing” line of defense (e.g., Baumeister, 1995). More than three decades of social psychological research have demonstrated a considerable range of ‘‘reality negotiation” techniques (Snyder & Higgins, 1988; Snyder, Irving, Sigmon, & Holleran, 1992; Swann, 1987), including self-serving biases, processes of reappraisal, doubting the source of threatening information, and the dynamics of ‘‘rationalization” (e.g., Swann & Hill, 1982). Both lines of defense are fairly efficient in reducing the threat of self-discrepant information. Nevertheless, all forms of reality negotiation and interpretive neutralization entail a certain degree of deceit. If I am convinced I have a good memory, but increasingly find myself returning from the shops without everything I need, and if I fail to explain this away (e.g., by denying that I had planned to buy milk or by excusing my omission by reasoning, say, that I was distracted by bumping into a friend), do I necessarily have to admit that my memory has declined? Apparently, to shop efficiently, I badly need a shopping list. Does this concession mean that I have to release the dearly held belief of having a good memory? In other words, does the concession that I forget to buy things while grocery shopping pose an unavoidable and serious threat to my self-concept? We have argued elsewhere (Greve & Wentura, 2003) that a third line of defense may help to avoid this unwelcome consequence without running into the ‘‘paradox” and ‘‘reality costs” of self-deception. An interesting way of protecting a
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
723
threatened self-concept – while accepting the threatening information – is by modifying the concept in question at its periphery. If a decreased level of performance in a certain skill (e.g., remembering the items on a shopping list) cannot be denied, but there is a need to maintain the self-concept with respect to a related general domain or trait (e.g., having a good memory), reducing the diagnosticity of that particular skill for the general domain solves the problem. Thus, I admit that I actually forgot to buy the milk (and accept that I will need to jot down a grocery list in the future), but I dispute that this particular skill says a great deal about my general memory capacity. In other words, I ‘‘trivialize” (Simon, Greenberg, & Brehm, 1995) my failure. Instead, I may reason that my memory is as good as ever, as evidenced (for example) by the fact that I still can recite the beginning of the Odyssey by heart or that I know the title of every single Hitchcock movie. In fact, this maneuver reconciles the reality principle with the pleasure principle: It allows me to concede that I need to write grocery lists but, at the same time, protects my belief that I have a good memory. Following the frequently used metaphor of the self-concept as a ‘‘theory” that a person holds about him- or her-self (e.g., Brim, 1976; Epstein, 1973; Schlenker, 1980), we borrowed the metaphor of immunization used to describe a key concept in the philosophy of science (Popper, 1934/ 1968) to name this self-stabilizing process. ‘‘Immunization” of a theory means a modification of its ‘‘protective belt” (Lakatos, 1970) to protect its ‘‘hard core” against inconsistent empirical evidence. Indeed, by adjusting the operationalizations of my self-concepts (e.g., of ‘‘memory”), I become immune to future forgetful episodes at the grocery store. Why should I tax the noble faculty of memory with such banalities? That’s what notepads are for! Thus, self-immunization integrates stability (of abstract beliefs: ‘‘having a good memory”) and change (of concrete realizations: ‘‘remembering shopping lists”), and thus reconciles defense (of self-related beliefs: ‘‘memory”) and realism (‘‘necessity of writing shopping lists”). 4. Empirical evidence for self-immunization The need to provide empirical evidence for self-immunizing processes raises another problem inherent in the concept of self-deception. If self-defense works, the individuals concerned do not know that (let alone how) it has worked; otherwise, it has clearly failed. Thus, it is impossible to ask people directly whether they have adapted certain aspects of their self-definition to protect more central parts of their self-concept (‘‘Do you deny your failures?” ‘‘Not to my knowledge!”). If the arguments sketched above prove to be tenable, however, the ‘‘personal” phenomenon of self-deception (‘‘I” know and do not know at the same time) can be resolved by differentiating several functions, processes, and structures of the self. Specifically, we propose that structures and changes thereof, as well as the processes that govern these changes, operate (and can thus be identified) on a subpersonal level. In the following, we sum up and discuss a series of studies providing evidence for the argument that it is possible to circumvent these problems of measurement (and apparent conceptual paradoxes) by (a) posing different questions (indirect self-report) and (b) administering cognitive-experimental tasks (e.g., reaction time measures as indicators of cognitive processes). 5. The synchronous perspective: identifying the immune self Our first study was based on the assumption that respondents would rate the skills or abilities they possess as being particularly diagnostic of a desired trait – and the skills or abilities they do not possess as being less diagnostic. In technical terms, the extent to which respondents believe they possess a skill was expected to correlate with their beliefs concerning the diagnosticity of that skill for a given trait. We began by conducting several questionnaire studies. For the domain of memory, for example, respondents were presented with a list of individual skills or abilities (Greve, 1990). They were asked, first, to what extent each skill or ability was indicative of a good memory and, second, to what extent they possessed that skill or ability. Although the strength of the relationships varied, the relationship between each individual skill and its perceived diagnosticity was positive and statistically significant. This finding has been replicated for many other self-concept domains, including intelligence, autonomy, education, attractiveness, and helpfulness (Greve & Wentura, 2003). As a single measure of immunization tendency is clearly preferable to calculating large columns of coefficients, we suggested using the degree of similarity between two individual profiles. For every domain considered, there is one profile for the individual’s skills and one profile for the corresponding diagnosticities. If immunization entails the individual adjustment of the diagnosticities of certain skills for a general domain to reflect perceived competencies in those skills, the degree of similarity of these rating profiles (i.e., profile correlation) should indicate the immunization tendency.1 The main finding from a series of questionnaire studies with a total of more than 700 participants aged 18–81 years is fairly clear: statistically positive immunization scores were obtained for the same domain (in various samples) as well as across various domains (Greve & Wentura, 2003, Study 1). 1 It should be noted that profile scores were calculated using ipsatized values to account for the following problem. If a list of skills relating to a given trait consists of some very easy and some very difficult items (e.g., ‘‘remembering one’s own address” vs. ‘‘remembering the genealogy of Brontë’s Wuthering Heights” for the trait of ‘‘memory”), the average ratings of both skill ability and diagnosticity will be high for the easy items, whereas both ratings will be low for the hard items, resulting in artificially positive intraindividual correlations. Thus, deviations from the average rating for diagnosticities and abilities were used to calculate profile similarities. Note that this is the most conservative measure of self-immunization.
724
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
6. Beyond correlation: the process perspective Although the evidence presented above supports the plausibility of our theoretical assumptions, the studies outlined merely identified traces of the phenomenon under investigation – and not the process themselves. Moreover, we cannot exclude the alternative explanation that the individual skills (and/or the participants’ perception thereof) were adjusted to perceived norms. At best, profile similarities uncover the sediment of self-immunization: the structural patterns that are left in the self-conception after immunization has done its work. In a second step, we therefore sought to demonstrate that change in one’s subjective competence in an individual skill brings about a corresponding change in the perceived diagnosticity of that skill. To this end, we employed two fairly divergent approaches. The first was a longitudinal approach, focusing on a real-life context that can be expected to elicit domain-specific changes in self-concept within a few months. The second was to experimentally induce self-immunization by means of experimentally manipulated performance feedback. The condition for the first approach was the occurrence of subjective change in competence. Unless researchers can afford to assess subjective competence over an entire lifetime, the scope of the investigation must be limited to a period during which respondents are likely to perceive an unexpected change in their ability to perform a certain activity, and during which a change in self-concept change is likely to occur. In a longitudinal questionnaire study, we asked psychology students attending a communication skills course to rate (a) several communication skills and (b) the subjective diagnosticity of these skills with respect to their general communicative ability. Students gave their ratings once at the beginning of the course and once at the end. Because the course entailed several opportunities for practice and critical feedback, a change in self-related beliefs with respect to at least some specific skills was expected. This process can be distressing: most students have high opinions of their ability to communicate effectively, yet invariably make mistakes the first time around. The students then have an entire semester to practice their skills and, hopefully, to acknowledge that good interviewing skills require a great deal of practice. The key result of this quasi-experiment was that not only the synchronous, but also the diachronic immunization scores were significant: the change in subjective competence predicted the change in diagnosticity attributed to each individual skill (Greve & Wentura, 2003, Study 3). In our second approach to the dynamics of self-immunization, we induced self-immunization in an experiment. Participants in a quiz game were informed that their performance in different knowledge domains (e.g., history, natural sciences) was either good or poor (relative to another participant, who was in fact a confederate of the experimenter). Results showed that participants downgraded the diagnosticity of this domain for general knowledge depending on their relative performance (i.e., a participant who performed relatively weakly in, e.g., history gave lower ratings for the item ‘‘Someone with good general knowledge knows a lot in the domain of history”; Greve & Wentura, 2003, Study 2). 7. Beyond self-report: revealing yourself without talking about your self However, all the results reported so far have one critical feature in common: they are all based on self-report data. We needed to go beyond this methodology for two reasons. The first comes in the form of a counterargument to our self-assessment of diagnosticity. In using this approach, we cannot exclude the possibility that participants draw inferences on their own performance and thus produce immunized responses (‘‘As I know that I am erudite, it follows from my poor performance in history that history is not that diagnostic for education”). Given this argument, we should try to assess immunization without forcing participants to construct their responses ad hoc in this way. The second reason is even more fundamental. The assumption of immunization processes is essentially a hypothesis about the mental representation of the self and its change in response to self-threatening events. Mental representations should be assessed by techniques that are more directly related to them than are verbal utterances (see also Wentura, 2005). Moreover, these indicators should be: (1) as invulnerable as possible to self-presentational tendencies (in order to preclude strategic self-presentation) and (2) identify structures of the self – and changes thereof – of which the individual is not aware (in order to avoid the paradoxes of self-deception). In cognitive psychology, the semantic priming paradigm is thought to meet these conditions – at least under certain constraints (see McNamara, 2005; Neely, 1991, for reviews). In the semantic priming paradigm, participants typically work through a series of trials, each consisting of a word or nonword that has to be classified according to its lexical status as quickly as possible. Shortly before presentation of this target word, another word (the prime) that is either semantically related or unrelated to the target stimulus is briefly flashed onto the screen. Response times to the target are faster if the prime is related than if it is unrelated. Although this task has often been used with associatively related prime–target pairs, there is also evidence for its application with categorically related pairs (Hutchison, 2003; Lucas, 2000). The paradigm thus seems suited to capture self-immunization: If, by way of self-immunization processes, a self-related category is more strongly linked to one’s strengths and less strongly linked to one’s weaknesses, this category label should prime the strengths but not the weaknesses. Practically, however, application of the priming technique to investigate self-immunization is somewhat more complicated. In an early attempt (Wentura & Greve, 1996), we instructed student participants to work through an intelligence test, assuming that ‘‘intelligence” was a self-relevant trait for them. The test was presented as consisting of a set of different modules, each identified by short labels (e.g., logic, calculation). Responses were given at the computer screen. The computer
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
725
provided a feedback after each module, supposedly indicating the participant’s performance relative to the average student. Feedback on the first two modules was moderately positive; feedback on the subsequent four modules followed a balanced scheme of two clearly positive and two clearly negative feedbacks. The final module (supposedly testing language abilities) consisted of a lexical decision task, which included the labels of the two positive and the two negative modules. To obscure our intention, we primed ‘‘intelligence” in a rather unconventional way. In a first block, which provided base rates, words and nonwords were presented at the center of the screen; the word list included the module names. In a second block, announced to the participants as being more difficult, items were randomly presented at one of the four corners of the screen. Presentation of the item was always preceded by a valid cue so that participants focused on the appropriate location in advance (with a cue–target asynchrony of 500 ms). The module labels were presented in the upper left corner, directly below the word intelligence, which was part of the supposed name of the test (i.e., ‘‘Intelligence Facets Test”) and constituted part of the test layout throughout the procedure. As hypothesized, response times in this prime condition were faster for names of modules on which participants had received positive feedback than for those on which they had received negative feedback. In the unprimed baseline condition, we did not observe this difference. These results were interpreted as indicating automatic (i.e., not intentionally fabricated) cognitive self-immunizing processes. In a further study (Wentura & Greve, 2005), we used complete sentences as primes (e.g., ‘‘Lisa is good at mental arithmetic”) to capture the gist of propositions expressing observable skills (for other studies using sentences as primes, see, e.g., Conway & Bekerian, 1987; Sharkey & Mitchell, 1985). Following the rapidly presented sentence, trait words were used as targets in a lexical decision task. Identification of potentially related targets (e.g., intelligent) was expected to be facilitated if there was a strong link between the skill and the trait in the individual self-concept. Three months before the laboratory task, participants completed a questionnaire assessing their beliefs about their strengths and weaknesses with regard to skills potentially related to the traits of intelligence, helpfulness, education, and attractiveness. Additionally, they rated the centrality of these traits for their self-concept. During the priming task, no explicit reference was made to the participants’ self-concept (i.e., the sentences referred to fictitious persons). The results supported our self-immunization hypothesis: If a trait was central for the self-concept, a significant priming effect was found for those skills that participants believed themselves to be good at, but not for those skills that participants believed themselves to be poor at. The findings for low centrality traits were reversed (i.e., a significant priming effect was found for those skills that participants believed themselves to be poor at, but not for those skills that participants believed themselves to be good at). This pattern of results corresponds with our hypothesis that processes of self-immunization protect a self-serving definition of central traits. Interestingly, the pattern did not hold for participants with low self-esteem (i.e., the quartile with the lowest scores on the Rosenberg scale, Rosenberg, 1965). For them, central traits seemed to be defined by weaknesses, whereas their own strengths were used to define noncentral traits. This observation strengthens the idea that self-immunization is a means of protecting self-esteem. Of course, this study again provided only a snapshot of the present self-concept, which might have adapted to individual strengths and weaknesses. Again, only an experimental design would allow the causal hypothesis that experiences of loss and failure lead to the adaptation of the self-concept to be tested. The study reported in Wentura and Greve (2004) returned to the cover story of a quiz task used in the experimental study by Greve and Wentura (2003, Study 2; see above). The hypothesis was that confronting student participants with gaps in their general knowledge would result in a pattern of self-immunization in which the trait ‘‘to be erudite” is defined in terms of those aspects of knowledge that the participant has demonstrated. We used the format of popular TV quiz shows, presenting general knowledge questions (e.g., ‘‘Which work was written by Dostoyevsky?”) with a multiple choice response format (response alternatives: [a] Crime and Punishment, [b] War and Peace, [c] The Gulag Archipelago, [d] Eugene Onegin). However, as it is unlikely that a question whose answer is considered to be common knowledge serves the function of tapping eruditeness, we combined the quasi-experimental factor of success (i.e., correct versus incorrect response to the general knowledge questions) with an experimental factor of difficulty. That is, a question was presented with either a relatively difficult set of answers (see above for the Dostoyevsky question) or a relatively easy set of answers (i.e., (a) Crime and Punishment, (b) The German Lesson, (c) Narcissus and Goldmund; (d) Faust; (b)–(d) are all German classics). Everything else being equal, responding correctly to a question that appeared more difficult should be taken as better evidence for being erudite. Additionally, the inclusion of easy tasks ensured that participants had the impression of a good level of overall performance. In the subsequent priming task, the prime sentences described those pieces of knowledge (e.g., Dostoyevsky wrote Crime and Punishment) that were tapped in the quiz task. The relevant target was the word erudite. We expected and found the most pronounced priming effects for difficult tasks that were solved correctly. (In fact, we found significant priming effects only for those pieces of knowledge that referred to difficult, but correctly solved tasks.) This result was replicated in a further experiment that additionally tested the findings against some alternative interpretations (see Wentura & Greve, 2004, for details). Specifically, participants related ‘‘being erudite” only to those propositions that represented – subjectively – noteworthy knowledge. Thus, the two studies by Wentura and Greve (2004, 2005) using the sentence priming task mutually corroborate each other: The first demonstrated the results of self-immunization processes across a range of domains, and the second used an experimental design to capture self-immunization within a specific domain. Finally, in a recent series of studies (Degner & Wentura, 2009), we used a masked affective priming task to assess automatic evaluation of stimuli. Specifically, we presented participants with a series of positive and negative pictures that had to be evaluated as quickly as possible. Shortly before presentation of each target picture, a prime picture followed by a mask was presented. Basic research has shown that the prime picture is evaluated automatically, and that responses to
726
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
valence-congruent targets are thus faster than are responses to valence-incongruent targets (Fazio, Sanbonmatsu, Powell, & Kardes, 1986). This result is even observed if the prime is presented outside of awareness (Draine & Greenwald, 1998). In our study, we used pictures of normal weight and overweight persons as masked primes. We obtained an index of relative positivity of overweight primes by taking the difference in response times for negative targets versus positive targets, both preceded by overweight primes, and subtracting from it the corresponding difference with normal weight primes. Across four studies using a variety of overweight and normal weight prime stimuli, we consistently found a positive relationship between the priming index and the body mass index of participants. Whereas, as expected, underweight and normal weight participants had a negative priming effect (denoting a negative evaluation of overweight), overweight participants showed a significant positive priming effect (denoting relative positivity of overweight). In accordance with other studies, no such relation was found when explicit attitudes were probed in a questionnaire (which revealed a negative overall view of overweight). We can interpret this discrepancy as a result of self-immunization processes: The questionnaire items are statements about obese or fat people in general. Thus, participants have to use their own definition of obesity or fatness to answer the items. It is conceivable that participants set these criteria differently and that overweight participants applied a definition of obesity or fatness that did not include their own weight status. Thus, the negativity they express may apply to individuals far heavier than themselves. In the affective priming task, in contrast, pictures of actual people with distinct weight statuses were presented, but not explicitly categorized as thin versus fat, and may thus not have been processed as fat individuals by the overweight participants. 8. Putting the picture together: self-immunization as (dis-)solution of apparent paradoxes and dilemmas The results presented in the previous section support the idea that we stabilize our selves (and hence ourselves) by means of self-immunization. Our findings from a series of studies with various correlational and experimental paradigms consistently suggest that self-immunization not only represents a part of the self’s self-defense system, but also reconciles the necessary acceptance of realities concerning the self with the desired stabilization of (central) self-conceptions. Thus, selfstabilization can be attained without ignoring realities (with respect to concrete skills or experiences) and without running into any ‘‘paradoxes” of self-deception. 9. Additional evidence The results of a series of empirical studies conducted by Dunning and colleagues (Dunning, 1993, 2005) on social judgments provide further supporting evidence for the concept of self-immunization. Dunning and colleagues convincingly argue that people tailor their judgment of others to maintain and bolster positive beliefs about themselves (Beauregard & Dunning, 1998; Dunning & Hayes, 1996) by referring to ‘‘egocentric definitions” (Dunning & Cohen, 1992) of social categories that lead them to emphasize their own behaviors and capabilities. For instance, data from a study by Dunning, Meyerowitz, and Holzberg (1989) indicated that the assessment of traits and abilities with high conceptual ambiguity (‘‘good leader”) is selfaggrandizing in that people tend to use those trait definitions that support their notion of ability. The tendency to prefer egocentric definitions of social categories was most apparent in studies in which people were asked to articulate their prototypes of socially desirable categories (Dunning, 1993; Dunning, Perie, & Story, 1991). According to Dunning and McElwee (1995), when people change their assessments of whether or not they posses a trait, they may do so because they have been prompted to revise their definition of the trait in question. Accordingly, Dunning and Cohen (1992) demonstrated that people use different definitions to describe themselves and to judge others’ performances. Thus, people’s self-descriptions can differ even their behavior is the same. Although these data clearly demonstrate that people possess a sort of functional flexibility in defining traits, our notion of self-immunization adds a new perspective. It is not only people’s judgments of others but also their self-conceptions that depend on their conceptualization of the respective dimensions or domains (see Dunning, 2005, p. 100ff.). According to the notion of self-immunization, however, these conceptualizations depend on people’s judgment of their own competencies. Thus, beyond addressing changes or individual differences in self-descriptions (influenced by – idiosyncratic – trait definitions), we focus on the intrapersonal stability of self-conceptions, considering changes in trait definitions that are prompted by the need to stabilize one’s judgment of one’s own competencies vis-à-vis the need to adapt one’s self-description to real developmental changes (e.g., declines). Dunning and colleagues – though mentioning the self-serving function of a flexible use of words in judging one’s own and others’ performances – ‘‘found no evidence that these judgments were motivated by a desire to maintain favorable self-concepts” (Dunning & Cohen, 1992, p. 352; see also Dunning, 1993). The theory of selfimmunization thus responds to Dunning’s (1993) call for specific research that demonstrates the individual tendency to use egocentric definitions when driven to maintain one’s self. 10. Limits to conceptual defense – and beyond However, self-immunization cannot be maintained indefinitely. First, like any other means of self-defense, it may become dysfunctional. Sometimes, self-concept revision is absolutely necessary. A case in point is accepting the need for a medical check-up or treatment. For instance, a decisive step in the therapy of alcoholics consists in breaking through the
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
727
immunization dynamic that contrafactually allows patients to believe that they can control their drinking (if they only wanted to). In a questionnaire study (Greve, Wentura, & Zisterer, 1999), we examined immunization of the self-concept with regard to alcoholism. The results showed that patients being treated for alcoholism immunized their self-concept less than did participants in a training program to have their driving license reinstated (after its withdrawal for drunk driving). In fact, the longer the alcoholic patients had been in therapy, the less they immunized. Second, social and even ‘‘epistemological” borderlines we cannot cross pose certain limits to self-immunization. Our social and cultural environments set language frames that we must not leave: It will become increasingly inacceptable if we individually tailor our conceptual definitions ad libitum. People whose memory is fading on all fronts will not only forget grocery lists and telephone numbers, but also addresses, rules, and anniversaries. At some point, it will become socially and conceptually impossible for them to deny that their memory per se – and not just some memory-related skill – has become worse. Many studies have shown that individuals register, both retrospectively and prospectively, the increasing negative ratio of gains to losses in old age (e.g., Brandtstädter, Wentura, & Greve, 1993; Heckhausen, Dixon & Baltes, 1987). Again, the reality principle requires that these changes be acknowledged – if necessary, on more abstract categorical levels. How can people maintain their identity when confronted with such serious developments, losses even? The two-process model of development (e.g., Brandtstädter, 2007; Brandtstädter & Renner, 1990; Brandtstädter & Rothermund, 2002; Brandtstädter & Wentura, 1995; for a discussion, see also Greve & Wentura, 2007) argues that individuals who cannot actively change reality succeed in managing their self-worth by rescaling their goals, reorienting towards different sources of meaning, or readjusting their values – again, without ignoring the realities. Although these ‘‘accommodative adaptations” differ in many respects from immunization processes (Brandtstädter, 1999; Brandtsädter & Greve, 1994), let us, for the purposes of this article, frame them in terms of ‘‘immunization.” Adapting one’s values and goals to accommodate undeniable and unalterable losses and changes can be viewed as an immunization of one’s value as a person: ‘‘I now define this value through different means.” It might be ventured that accommodation is the immunization of identity and self-esteem. 11. Perspectives for future research: development of self-immunization Little is yet known about the development of self-immunization. It seems plausible that younger children do not posses the structural prerequisites for adjusting definitions and operationalizations to self-related experiences and beliefs (Harter, 1999, 2003) and that the capacity for self-immunization develops during adolescence. In a first pilot study (Greve et al., 2009), we found evidence for a shift from a global protective function of self-immunization in early adolescence to more specific functions in later adolescence. There is, however, no linear age progression in the occurrence of self-immunization. This finding suggests that the general process of shifting definitions (i.e., accommodation in a Piagetian sense) changes and develops throughout later childhood and early adolescence. Another way of gaining insight into the generation of self-immunization is to take a closer look at specific psychological interventions. A case in point may be the self-experiential components of psychotherapy training, which are based less on the alleviation of suffering than on the assumption that it may be helpful for therapists to lose a few of the illusions they have about themselves. In a quasi-experimental study, Greve (2008) showed that self-experiential training for therapists altered not their self-conceptions, but their degree of immunization. This finding suggests that people intending to change their selves learn to change their lines of defense. How one interprets the strengthening of these lines of defense depends on the individual case and on the content of the training curriculum; in some cases, it is useful or even necessary, in others, it is counterproductive. Any notion of ‘‘self-insight” must, however, be related to processes not only of self-defense, but also of self-defense change. The ‘‘road to knowing yourself” (Dunning, 2005) is not only blocked and winding, but perhaps not a road at all. 12. Concluding remarks: personal and subpersonal explanatory frames The aims of this article were threefold. First, we argued that the pleasure/reality dilemma of self-defense can be resolved by means of self-stabilization. Self-immunization serves both the reality and the pleasure principle, as it combines stability and change. Stability at a more abstract, conceptual, level is maintained by modifying the concept at its (operational) periphery, which enables the individual (i.e., the self) to accept changes (and losses) in those skills that define the central self-concepts. Second, we presented empirical evidence for this strategy of self-stabilization, which takes into account that processes of self-defense, if not their results, must remain disclosed from the individual’s conscious awareness. Third, we argued that differentiating these two levels of the self resolves the apparent paradox of self-deception. Although the empirical findings presented (and discussed) in the previous sections have been published in various papers, this is the first attempt to combine them into a single picture. This picture may indicate that self-deception is just a sample case of a more general problem. According to the arguments presented here, conceptual and empirical differentiation can resolve the apparent ‘‘paradox” of self-deception and reconcile the reality and pleasure principles: Stability (pleasure) is maintained on an ‘‘upper” level, while a ‘‘lower” level is changed (adjusted) in response to changing realities. Note that this ‘‘upper” level is the focus of people’s self-conceptions and is thus the conscious level (‘‘I have a good memory”). As a rule, the ‘‘lower” level is less often the focus of conscious self-reflections (‘‘My ability to remember shopping lists has decreased in the last ten years”), and people are hardly ever aware of the cognitive link between the two – or, in particular, of changes in such connections
728
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
(i.e., the adjustment of ‘‘operationalizations” of self-conceptions). Thus, the effect of ‘‘self-deception” (‘‘My memory is as good as ever”) is not paradoxical after all: it is true given my current understanding of memory, but a lie in terms of my earlier definition. This self-stabilization maneuver depends on adjustments in cognitive structures that are beyond the horizon of conscious awareness. However, switching between explanatory levels (personal/consciously experienced stability versus subpersonal/cognitively adaptive dynamics) is more than just applying different models to different phenomena. Rather, it can be seen as an attempt to reconcile two possible perspectives on human beings: We are, at the same time, persons and biological systems (Greve & Wentura, 2007; Wentura, 2005; Wentura & Greve, 2006; see also Brandtstädter, 1982, 1984, 2006; Dennett, 1987). Psychological phenomena in the focus of our everyday interest (feeling, thinking, acting), including phenomena of self-conception (‘‘insight,” self-esteem, self-image, etc.), take the personal perspective: Persons (not biological systems) act according to their goals; persons (not biological systems) experience concern if goal attainment seems to be impeded; and persons (not biological systems) have high or low self-esteem. From this point of view, a ‘‘personal” psychology perspective is indispensable at least to describe the phenomena of interest (‘‘Who do you think you are?”). However, researchers interested in the causal mechanics behind complex phenomena (e.g., acts of self presentation or self-related beliefs) must refer to the language of ‘‘subpersonal” cognitive psychology. The present example of realistic self-stabilization demonstrates this very clearly: On a personal level, the paradox of selfdeception cannot be resolved. Certainly, personal and subpersonal approaches to ‘‘psychological” phenomena do not focus on different kinds of processes, but rather represent different theoretical frames (Wittgensteinian ‘‘language games,” as it were). Yet, in a subpersonal language, ‘‘the person” is no longer identifiable: on this level, for conceptual reasons, we are indeed ‘‘no one” (Metzinger, 2003). Even if all the subpersonal (e.g., cognitive) explanations were plausible and empirically corroborated (which they are not), however, this could never cause us to abandon our personal stance towards ourselves: We simply cannot but see ourselves as persons. Perhaps it is worth mentioning that arguing in favor of two tiers of explanation (or even three, taking into account the physical stance that focuses on the biological implementation of the cognitive-functional level; Dennett, 1987) neither entails an ‘‘unnecessary dichotomy” (Anderson, 1991) nor means to argue in favor of reductionism. The personal (intentional) stance can certainly not be reduced to a functional one (let alone a physical level). Rather, we argue in favor of a more careful discussion of different ‘‘kinds” of regulative processes. The respective priority, importance, and even assignment of concrete processes of self-regulation to one or the other category would be much easier if viewed this way. This kind of approach might even lay the foundations for a common theoretical framework that unifies heterogeneous theories of self-regulation. References Anderson, J. R. (1991). Is human cognition adaptive? Behavioral and Brain Sciences, 14, 471–517. Baltes, P. B., Lindenberger, U., & Staudinger, U. M. (2006). Life-span theory in developmental psychology. In R. M. Lerner (Ed.), Handbook of child psychology (6th ed.. Theoretical models of human development (Vol. 1, pp. 569–664). New York, NY: Wiley. Baumeister, R. F. (1993). Lying to yourself: The enigma of self-deception. In M. Lewis & C. Saarni (Eds.), Lying and deception in everyday life (pp. 166–183). New York: Guilford. Baumeister, R. F. (1995). Self and identity: An introduction. In A. Tesser (Ed.), Advanced social psychology (pp. 50–98). New York, NY: McGraw-Hill. Baumeister, R. F. (1996). Self-regulation and ego threat: Motivated cognition, self deception, and destructive goal setting. In P. M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action. Linking cognition to behavior (pp. 27–47). New York, NY: Guilford. Beauregard, K. S., & Dunning, D. (1998). Turning up the contrast: Self-enhancement motives prompt egocentric contrast effects in social judgement. Journal of Personality and Social Psychology, 74, 606–621. Bengtson, V. L., Reedy, M. N., & Gordon, C. (1985). Aging and self-conceptions: Personality processing and social contexts. In J. E. Birren & K. W. Schaie (Eds.), Handbook of psychology of aging (pp. 544–593). New York, NY: Van Nostrand. Brandtstädter, J. (1982). Apriorische Elemente in psychologischen Forschungsprogrammen [A priori elements in psychological research programs]. Zeitschrift für Sozialpsychologie, 13, 267–277. Brandtstädter, J. (1984). Apriorische Elemente in psychologischen Forschungsprogrammen: Weiterführende Argumente und Beispiele [A priori elements in psychological research programs: Further arguments and examples]. Zeitschrift für Sozialpsychologie, 15, 151–158. Brandtstädter, J. (1999). Sources of resilience in the aging self: Toward integrating perspectives. In T. M. Hess & F. Blanchard-Fields (Eds.), Social cognition and aging (pp. 123–141). San Diego, CA: Academic Press. Brandtstädter, J. (2006). Action perspectives on human development. In R. M. Lerner (Ed.), Handbook of child psychology (6th ed.. Theoretical models of human development (Vol. 1, pp. 516–568). New York, NY: Wiley. Brandtstädter, J. (2007). Das flexible Selbst [The flexible self]. Heidelberg, Germany: Elsevier/Spektrum Akademischer Verlag. Brandtstädter, J., & Greve, W. (1994). The aging self: Stabilizing and protective processes. Developmental Review, 14, 52–80. Brandtstädter, J., & Renner, G. (1990). Tenacious goal pursuit and flexible goal adjustment: Explication and age-related analysis of assimilative and accommodative strategies of coping. Psychology and Aging, 5, 58–67. Brandtstädter, J., & Rothermund, K. (2002). The life course dynamics of goal pursuit and goal adjustment: A two-process framework. Developmental Review, 22, 117–150. Brandtstädter, J., & Wentura, D. (1995). Adjustment to shifting possibility frontiers in later life: Complementary adaptive modes. In R. A. Dixon & L. Bäckman (Eds.), Compensating for psychological deficits and declines: Managing losses and promoting gains (pp. 83–106). Hillsdale, NJ: Erlbaum. Brandtstädter, J., Wentura, D., & Greve, W. (1993). Adaptive resources of the aging self: Outlines of an emergent perspective. International Journal of Behavioral Development, 16, 323–349. Breznitz, S. (Ed.). (1981). The denial of stress. New York: International Universities Press. Brim, O. G. (1976). Life-span development of the theory of oneself: Implications for child development. In H. W. Reese (Ed.). Advances in child development and behavior (Vol. 11, pp. 241–251). New York, NY: Academic Press. Cassam, Q. (Ed.). (1994). Self-knowledge. Oxford, UK: Oxford University Press. Conway, M. A., & Bekerian, D. A. (1987). Situational knowledge and emotions. Cognition & Emotion, 1, 145–191. Costa, P. T., & McCrae, R. R. (1997). Longitudinal stability of adult personality. In R. Hogan, J. Johnson, & S. Briggs (Eds.), Handbook of personality psychology (pp. 269–290). San Diego, CA: Academic Press. Davidson, D. (1985). Deception and division. In J. Elster (Ed.), The multiple self (pp. 79–92). Cambridge, UK: Cambridge University Press.
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
729
Degner, J., & Wentura, D. (2009). Not everybody likes the thin and despises the fat: One’s weight matters in the automatic activation of weight-related social evaluations. Social Cognition, 27, 202–220. Dennett, D. (1987). The intentional stance. Cambridge, MA: MIT Press. Draine, S. C., & Greenwald, A. G. (1998). Replicable unconscious semantic priming. Journal of Experimental Psychology: General, 127, 286–303. Dunning, D. (1993). Words to live by: The self and definitions of social concepts and categories. In J. Suls (Ed.). Psychological perspectives on the self (Vol. 4, pp. 99–126). Hillsdale, NJ: Erlbaum. Dunning, D. (2005). Self-insight. New York, NY: Psychology Press. Dunning, D., & Cohen, G. L. (1992). Egocentric definitions of traits and abilities in social judgements. Journal of Personality and Social Psychology, 63, 341–355. Dunning, D., & Hayes, A. F. (1996). Evidence for egocentric comparison in social judgement. Journal of Personality and Social Psychology, 71, 213–229. Dunning, D., Leuenberger, A., & Sherman, D. A. (1995). A new look at motivated inference: Are self-serving theories of success a product of motivational forces? Journal of Personality and Social Psychology, 69, 58–68. Dunning, D., & McElwee, R. O. (1995). Ideosyncratic trait definitions: Implications for self-description and social judgement. Journal of Personality and Social Psychology, 68, 936–946. Dunning, D., Meyerowitz, J. A., & Holzberg, A. D. (1989). Ambiguity and self-evaluation: The role of idiosyncratic trait definitions in self-serving assessments of ability. Journal of Personality and Social Psychology, 57, 1082–1090. Dunning, D., Perie, M., & Story, A. L. (1991). Self-serving prototypes of social categories. Journal of Personality and Social Psychology, 61, 957–968. Epstein, S. (1973). The self-concept revisited: Or a theory of a theory. American Psychologist, 28, 405–415. Erdelyi, M. H. (1974). A new look at the ‘‘new look”: Perceptual defense and vigilance. Psychological Review, 81, 1–25. 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. Filipp, S.-H., & Klauer, T. (1985). Conceptions of self over the life-span: Reflections on the dialectics of change. In M. M. Baltes & P. B. Baltes (Eds.), The psychology of aging and control (pp. 167–205). Hillsdale, NJ: Erlbaum. Freud, A. (1936). The ego and the mechanisms of defense. New York: International Universities Press. Funder, D. C. (2001). Personality. Annual Review of Psychology, 52, 197–221. Funder, D. C. (2007). The personality puzzle (4th ed.). New York, NY: Norton. Greenwald, A. G. (1980). The totalitarian ego. Fabrication and revision of personal history. American Psychologist, 35, 603–618. Greve, G. (2008). Selbsterfahrung in der Verhaltenstherapieausbildung: Inhalt und Prozess [Self-experience as a component of behavior therapy training: Content and process]. Lengerich, Germany: Pabst. Greve, W. (1990). Stabilisierung und Modifikation des Selbstkonzeptes im Erwachsenenalter: Strategien der Immunisierung [Stabilization and modification of self-concepts in adulthood: Strategies of immunization]. Sprache und Kognition, 9, 218–230. Greve, W. (Ed.). (2000a). Die Psychologie des Selbst [The psychology of self]. Weinheim, Germany: PVU. Greve, W. (2000b). Das erwachsene Selbst [The adult self]. In W. Greve (Ed.), Die Psychologie des Selbst (pp. 96–114). Weinheim, Germany: PVU. Greve, W. (2005). Maintaining personality: The active and adaptive self as core of individuality and personhood. In W. Greve, K. Rothermund, & D. Wentura (Eds.), The adaptive self: Personal continuity and intentional self-development (pp. 49–70). Göttingen, Germany: Hogrefe & Huber. Greve, W., & Wentura, D. (2003). Immunizing the self: Self-concept stabilization through reality-adaptive self-definitions. Personality and Social Psychology Bulletin, 29, 39–50. Greve, W., & Wentura, D. (2007). Personal and subpersonal regulation of human development: Beyond complementary categories. Commentary on Kathrin Boerner’s and Daniela Jopp’s Improvement/maintenance and reorientation as central features of coping with major life change and loss: Contributions of three life-span theories. Human Development, 50, 201–207. Greve, W., Wentura, D., & Zisterer, M. (1999). Dynamiken der Selbst-Stabilisierung im Alter [Self-concept stabilization in adulthood: Functionality and limits of functionality of self-immunization]. Zeitschrift für Medizinische Psychologie, 8, 167–174. Greve, W., Zimmermann, S., Meyer, T., Ackermann, J., Hellmers, S., & Pielniok, M. (2009). Die Entwicklung von identitätsstabilisierenden Prozessen im Jugendalter – das Beispiel Immunisierung’ [The development of self-stabilizing processes during adolescence – The sample case of ‘immunization’]. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 41, 26–31. Gur, R. C., & Sackheim, H. A. (1979). Self-deception: A concept in search of a phenomenon. Journal of Personality and Social Psychology, 37, 147–169. Haight, M. R. (1980). A study of self-deception. Sussex, UK: Harvester Press. Hannover, B. (2000). Das kontextabhängige Selbst [The context-dependent self]. In W. Greve (Ed.), Psychologie des Selbst (pp. 227–238). Weinheim, Germany: PVU. Hannover, B., Pöhlmann, C., Springer, A., & Roeder, U. (2005). Implications of independent versus interdependent self-knowledge for motivated social cognition: The Semantic Procedural Interface Model of the Self. Self and Identity, 5, 159–175. Harter, S. (1999). The construction of the self. A developmental perspective. New York, NY: Guilford. Harter, S. (2003). The development of self-representations during childhood and adolescence. In M. R. Leary & J. P. Tangney (Eds.), Handbook of self and identity (pp. 610–642). New York, NY: Guilford. Heckhausen, J., Dixon, R. A., & Baltes, P. B. (1987). Gains and losses in development throughout adulthood as perceived by different adult age groups. Developmental Psychology, 25, 109–121. Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280–1300. Hilgard, E. R. (1949). Human motives and the concept of self. American Psychologist, 4, 347–382. Hutchison, K. A. (2003). Is semantic priming due to association strength or feature overlap? A microanalytic review. Psychonomic Bulletin & Review, 10, 785–813. James, W. (1890). The principles of psychology (authorized edition). New York, NY: Dover. Lakatos, I. (1970). Falsification and the methodology of scientific research programmes. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge (pp. 91–195). Cambridge, UK: Cambridge University Press. Lazarus, R. S., & Golden, G. Y. (1981). The function of denial in stress, coping, and aging. In J. L. McGaugh & S. B. Kiesler (Eds.), Aging: Biology and behavior (pp. 283–307). New York: Academic Press. Leary, M. R., & Tangney, J. P. (Eds.). (2003). Handbook of self and identity. New York, NY: Guilford. Lucas, M. (2000). Semantic priming without association: A meta-analytic review. Psychonomic Bulletin & Review, 7, 618–630. Markus, H., & Wurf, E. (1987). The dynamic self-concept: A social psychological perspective. Annual Review of Psychology, 38, 299–337. Martin, M. W. (1986). Self-deception and morality. Lawrence, KA: University Press of Kansas. McNamara, T. P. (2005). Semantic priming: Perspectives from memory and word recognition. New York, NY: Psychology Press. Mead, G. H. (1934). Mind, self, and society. Chicago, IL: University Press. Mele, A. (1987). Recent work on self-deception. American Philosophical Quarterly, 24, 1–17. Metzinger, T. (2003). Being no one. Cambridge, MA: MIT Press. Miller, D. T. (1976). Ego-involvement and attributions for success and failure. Journal of Personality and Social Psychology, 34, 901–906. Neely, J. H. (1991). Semantic priming effects in visual word recognition: A selective review of current findings and theories. In D. Besner & G. W. Humphreys (Eds.), Basic processes in reading. Visual word recognition (pp. 264–336). Hillsdale, NJ: Erlbaum. Popper, K. R. (1934/1968). The logic of scientific discovery. London UK: Hutchinson. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Schlenker, B. R. (1980). Impression management. Monterey, CA: Brooks. Sharkey, N. E., & Mitchell, D. C. (1985). Word recognition in a functional context: The use of scripts in reading. Journal of Memory and Language, 24, 253–270.
730
W. Greve, D. Wentura / Consciousness and Cognition 19 (2010) 721–730
Simon, L., Greenberg, J., & Brehm, J. (1995). Trivialization: The forgotten mode of dissonance reduction. Journal of Personality and Social Psychology, 68, 247–260. Snyder, C. R., & Higgins, R. L. (1988). Excuses: Their role in the negotiation of reality. Psychological Bulletin, 104, 23–35. Snyder, C. R., Irving, L. M., Sigmon, S. T., & Holleran, S. (1992). Reality negotiation and valence/linkage of self-theories: Psychic showdown at the ‘‘I’m OK” corral and beyond. In L. Montada, S.-H. Filipp, & M. J. Lerner (Eds.), Life crises and experience of loss in adulthood (pp. 275–297). Hillsdale, NJ: Erlbaum. Swann, W. B. (1987). Identity negotiation: Where two roads meet. Journal of Personality, 53, 1038–1051. Swann, W. B., & Hill, C. A. (1982). When our identities are mistaken. Journal of Personality and Social Psychology, 43, 59–66. Taylor, S. E., & Gollwitzer, P. M. (1995). Effects of mindset on positive illusions. Journal of Personality and Social Psychology, 69, 213–226. Tesser, A., Felson, R. B., & Suls, J. M. (Eds.). (2000). Psychological perspectives on self and identity. Washington, DC: American Psychological Association. Tesser, A., Martin, L. L., & Cornell, D. P. (1996). On the substitutability of self-protective mechanisms. In P. M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action. Linking cognition and motivation to behavior (pp. 48–68). New York, NY: Guilford. Tesser, A., Stapel, D. A., & Wood, J. V. (Eds.). (2002). Self and motivation. Washington, DC: American Psychological Association. Wentura, D. (2005). The unknown self: The social cognition perspective. In W. Greve, K. Rothermund, & D. Wentura (Eds.), The adaptive self: Personal continuity and intentional self-development (pp. 203–222). Göttingen, Germany: Hogrefe & Huber. Wentura, D., & Greve, W. (1996). Selbstkonzept-Immunisierung: Evidenz für automatische selbstbildstabilisierende Begriffsanpassungen [Selfimmunization: Evidence for automatic self-stabilizing concept adjustments]. Zeitschrift für Sozialpsychologie, 27, 207–223. Wentura, D., & Greve, W. (2004). Who wants to be. erudite? Everyone! Evidence for automatic adaptation of trait definitions. Social Cognition, 22, 30–53. Wentura, D., & Greve, W. (2005). Assessing the structure of self-concept: Evidence for self-defensive processes by using a sentence priming task. Self and Identity, 4, 193–211. Wentura, D., & Greve, W. (2006). Duality models in social psychology: Different languages or interacting systems? Psychological Inquiry, 17, 210–216.
Consciousness and Cognition 19 (2010) 731–744
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Self-deception as pseudo-rational regulation of belief q Christoph Michel *, Albert Newen 1 Ruhr-Universität Bochum, Institut für Philosophie, GA3/141, Universitätsstraße 150, D-44780 Bochum, Germany
a r t i c l e
i n f o
Article history: Received 28 December 2009 Available online 23 August 2010 Keywords: Self-deception Rationality Belief-revision Self-immunization
a b s t r a c t Self-deception is a special kind of motivational dominance in belief-formation. We develop criteria which set paradigmatic self-deception apart from related phenomena of automanipulation such as pretense and motivational bias. In self-deception rational subjects defend or develop beliefs of high subjective importance in response to strong counter-evidence. Self-deceivers make or keep these beliefs tenable by putting prima-facie rational defense-strategies to work against their established standards of rational evaluation. In paradigmatic self-deception, target-beliefs are made tenable via reorganizations of those belief-sets that relate relevant data to target-beliefs. This manipulation of the evidential value of relevant data goes beyond phenomena of motivated perception of data. In selfdeception belief-defense is pseudo-rational. Self-deceivers will typically apply a dual standard of evaluation that remains intransparent to the subject. The developed model of self-deception as pseudo-rational belief-defense is empirically anchored. So, we hope to put forward a promising candidate. Ó 2010 Elsevier Inc. All rights reserved.
1. A distinction: auto-manipulation vs. self-deception 1.1. Pluralism Peter believes that he and John, a famous and celebrated actor and (in reality) a quite distant friend of Peter, are involved in a very deep friendship. They do know each other, and John has even invited Peter to some of his parties. But Peter has never even come close to John’s inner circle. Nevertheless their recent interaction has motivated Peter to believe that he is John’s favorite. However, no one else, including John himself, would come to the conclusion that he and Peter are close friends. Examples like this illustrate that self-deception is a familiar phenomenon. But the analysis of self-deception has proven to be a tricky endeavor. There is little consensus on the proper analysis of the cognitive state and the cognitive dynamics of selfdeception. The two basic reasons for the high diversity in theory are, firstly, that there are various ways in which motivation can influence acceptance and, secondly, that the pre-analytic folk-intuition of self-deception is heavily under-determined. To put things right, we suggest a distinction between motivational dominance on acceptance in general and a more constrained notion of self-deception in particular. Most people are willing to call a subject S a ‘self-deceiver’ in loose, ordinary folk-talk, if the following is true of his or her mind and behavior:
q
This article is part of a speical issue of this journal on Self, Other and Memory. * Corresponding author. Fax: +49 234 32 14963. E-mail addresses:
[email protected] (C. Michel),
[email protected] (A. Newen). 1 Fax: +49 234 32 14963.
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.019
732
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
1.1.1. Pattern of Motivational Dominance (PMD)2 (1) S is of normal intelligence and has normal rational capacities. (2) S is strongly motivated to evaluate p as true. (3) S has access to a sufficient amount of relevant information about the subject matter SM and this information is supportive of not-p. (4) S honestly accepts the proposition p. (5) If S had been neutral towards SM3, the same evidence would have led S to accept not-p instead of p. A few preliminary remarks about what makes those conditions intuitively necessary: Condition (1) says that there can be no general absence or breakdown of S’s rational capacities, including S’s ability to draw simple inferences. S is in that sense sufficiently rational. Condition (2) pragmatically excludes cases in which subjects come to be deceived that p without having any motivation to believe that p. The typical motivational source of self-deception is that the subject matter at stake is of high individual importance to S. Furthermore, as a potential self-deceiver, S cannot be completely blind to the relevant evidence. If S did not access the evidence, S could not be blamed for anything like self-deception, and this is why we need (3). Additionally, S has to accept p. Here ‘acceptance’ is a liberal notion of belief. S’s acceptance of p need, e.g. not entail that S always consistently acts on p. The degree or nature of acceptance is left open here. Finally, (5) claims that it is motivation which causes p-acceptance. S would judge not-p in similar cases when he is neutral with regard to SM.4 Note that (1) and (3) should imply that self-deceivers are evidence-sensitive and in principle capable of rational belief-adjustment with regard to the subject matter. This excludes belief-like phenomena like pathological confabulation and various forms of delusion from the realm of everyday self-deception. PMD provides the necessary and sufficient conditions for a pre-analytic folk-notion of self-deception in a broad sense. But PMD—as it stands—only says that motivation dominates acceptance of p in a rational subject S. PMD is a coarsegrained pattern; it does not specify how the motivational impact on acceptance is accomplished. Therefore PMD is compatible with nearly all the established theoretical analyses of self-deception.5 There are multiple ways motivational dominance in the sense of PMD could be implemented in S. If we do not want all of these equally to go through as ‘selfdeception’, it is clear that the pre-analytic inclination to label a given case of motivational dominance as ‘self-deception’ will not yet tell us whether this type of cases actually captures proper self-deception in contrast to other forms of PMD. The differences between forms of motivational dominance are remarkable. The debate on self-deception has produced various moreor-less appealing psycho-philosophical stories about how motivation contributes to S’s acceptance of p. We can choose freely between models that analyze self-deception as a form of intentional deception, as a case of motivational bias, models which describe self-deception as a gap between belief and avowal, as a form of attention control, as a form of pretense, a form of repression, as a phenomenon related to confabulation or as false meta-belief, just to name a few. This variety of phenomena requires us to specify all of the PMD-conditions given above and see what additional criteria need to be provided. The underdeterminacy of PDM clearly suggests a methodological limitation to a simply example-based approach to self-deception. Example-based approaches look for what is sufficient to explain cases we would pre-theoretically label as ‘self-deception’.6 This method thereby takes it that the pre-analytic intuition is sufficient to identify those cases that are examples of ‘real selfdeception’. However, the above considerations show that all it might finally do is to pick one of various types of PMD-cases. Thus, more work needs to be done. Going beyond PMD we will fix ‘self-deception in a narrow sense’7 by dint of further reasonable constraints. The basic distinction we propose is the distinction between auto-manipulation8 in general and self-deception in particular. PMD is a general model of auto-manipulation in sufficiently rational subjects in all of its varieties, and it is basically an empirical question as to which different forms of motivational dominance actually occur. Auto-manipulation occurs in any case where motivation dominates acceptance of a proposition p. ‘Self-deception’ refers to a class of special phenomena within the broader frame of auto-manipulation. In what follows, we will show that this distinction is useful and argue that being a case of PMD is, while necessary, not sufficient for being a case of self-deception. We intend to show with regard to two examples that selfdeception in a narrow sense is clearly different form other forms of motivational dominance. In the next section, we introduce three fundamental criteria of adequacy for self-deception.
2 Convention for the rest of this paper: ‘‘p” designates the content of the target belief, favored by the subject S but unsupported by evidence; ‘‘not-p” is the content of the belief actually justified by evidence. 3 I.e. if condition (2) is not fulfilled. 4 This comprises the counterfactual claim that, if S had been neutral towards the subject matter, S would have judged not-p, as well as the claim that typically S will judge not-p in parallel cases on the basis of equivalent evidence if those cases are not evaluated by S as being of any subjective importance. 5 In particular, the conditions of PMD do not even address the question that has driven the debate among philosophers: What’s the subject’s or the self’s very own contribution to the manipulation of his or her attitude? PMD, as it stands, is neutral between what the intentionalist and the deflationist camps disagree on. 6 An example-based approach explicitly receives methodological preference by Mele (2001, p. 5f). 7 Simply ‘‘self-deception” in the following. 8 The term ‘auto-manipulation’ is supposed to be neutral between different theories of self-deception and implies no intentionalist preference.
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
733
1.2. Three constraints on self-deception Giving a satisfactory account of self-deception is a joint-venture of normative and descriptive elements. Basically, an adequate model should incorporate reasonable theoretical and conceptual constraints and capture a real phenomenon. Defining self-deception requires clarifying all five conditions of PMD and providing supplemental conditions. Three criteria of adequacy for an analysis of self-deception suggest themselves from the very start. (i) An adequate model of self-deception must be psychologically viable9 Criterion (i) presupposes that a model is coherent. It is difficult to judge from the armchair what is actually ‘psychologically possible’, but a model of self-deception must, at the very least, be in line with some minimal requirements of rationality, i.e. it should show how self-deception could be plausibly implemented in a sufficiently rational subject. So-called literalist ‘deception-models’10 of self-deception where S deceives himself knowingly and purposefully require that: (a) S believes that not-p, (b) S intends to believe that p, and (c) S comes to believe that p as a result of his intentional efforts to deceive himself. Self-deception, so conceived, involves a simultaneous endorsement of both beliefs, p and not-p, as well as an intention to deceive oneself. This has obvious difficulties. Therefore, refined intentionalism as well as non-intentionalism about self-deception both suggest that to be plausible, a model of everyday self-deception should, firstly, avoid presupposing that self-deceivers must hold incoherent beliefs of the form ‘‘p and not-p”.11 Secondly, self-deception should not require any intention of the form ‘‘Deceive yourself!” since such intentions strongly tend to undermine their accomplishment in sufficiently rational subjects.12 In this sense, the so-called ‘paradoxes of self-deception’ impose limitations on how self-deception can be plausibly characterized as a state and as a process, and virtually all accounts are tailor-made to avoid them. Beyond avoiding the ‘paradoxes’, theorists who aim at describing the paradigmatic garden-variety self-deception would also require the following: (ii) An adequate model of self-deception should provide a description of a phenomenon that is both real and commonplace Condition (ii) says that a model of self-deception need not only be viable but also capture an actual and commonplace phenomenon and explain intuitive examples of garden-variety self-deception. Condition (ii) by itself does not exclude the possibility that forms of auto-manipulation may exist, which are more common or widespread than our target phenomenon. Self-deception occurs frequently enough to be familiar, but it still seems the exception rather than a regular condition of belief. Empirical grounding will be useful to show that a model satisfies (ii). It is, of course, beyond the scope of this article to discuss in detail all intentionalist and non-intentionalist proposals on how to navigate around the cliffs of paradox. And we do not dispute that there are possible cases of intentional auto-manipulation that satisfy the essential requirements of the deception-model. So called ‘mental partitioning’,13 ‘temporal partitioning’,14 or other sophisticated methods of inducing false beliefs in oneself offer ways to retain the essentials of the deceptionmodel. But since a critique of intentionalism is not in the focus of this article, we have to rest content with the observation that intentionalist constructions all share a tendency to satisfy (i) for the price of failing (ii) in different respects.15 We suggest that both forms of partitioning do not describe what one would call intuitive and common examples of self-deception, although they can be seen as actual cases of auto-manipulation. For example if mental partitioning (two subsystems in one mind) is to be the strategy to prevent the paradoxes, it is questionable whether this would capture a commonplace phenomenon, even if such divisions are possible.16
9
For an elucidation of the paradoxes see Section 2.1. In its strongest version the intentionalist ‘deception-model’ of self-deception parallels intentional interpersonal deception. Interpersonal deception requires a deceiver, A, who intentionally leads a finally deceived individual, B, into believing a proposition p that A himself considers false. In the standard case, A believes not-p and leads B to believe that p is true. Similarly, in self-deception S intends to deceive himself. S is motivated to do so due to his desire to evaluate p as true, paired with his belief that not-p. 11 Which is not to say that the mind could not harbor inconsistencies. 12 If A and B are identical, the constitutive asymmetry in knowledge between A and B which enables literal deception gets lost. The deceiver and the deceived both know about the actual facts as well as about the deceptive intention. This spoils self-deception in the same way A will spoil his plan to deceive B by starting out saying, ‘‘Hello, might I deceive you into believing that I’m a professor at Harvard?” In other words, a dynamic paradox is lurking. How could the intention ‘‘Deceive yourself!” bring about a successful belief change when we appear to be required to be ignorant of this very intention as well as of the facts at issue, if we are to become deceived? 13 So-called ‘mental-partitioning’-accounts explain self-deception by postulating more or less independent sub-centers of agency that allow to implement self-deception very close to the model of other-deception. 14 See Bermúdez (1997, 2000). 15 José Bermúdez (2000) has presented a form of self-induced long-term attitude-shift as an instance of intentional self-deception. Brian McLaughlin refers to the ‘diary-case’, a ‘‘memory exploiting stratagem” of self-induced deception (McLaughlin, 1988, Section 31f). A person can mislead herself about the date of an appointment she wants to miss by intentionally making a false entry in her diary and trusting her notoriously bad memory. She’s right about her memories qualities and and misses the appointment, believing that it is set as the diary says. 16 In this article we cannot discuss in detail refined modifications of divisionism like the one provided by Pears (1991). 10
734
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
Motivational influence on our beliefs turns deceptive when normally rational subjects manipulate their own attitudes in direct response to strong evidence to the contrary. S can only be blamed for self-deception, if the evidence is strongly asymmetrical with regard to the subject matter at stake. If the body of evidence was really ambiguous, we could hardly blame S for self-deception. If S notices that p has some evidence, if p is integrated in a belief-network and if p has proven pragmatically useful, then S has a perfectly rational motivation not to give up or not to reject his belief that p prematurely. We therefore stress that over and above (i) and (ii), the following condition is a crucial one. (iii) In self-deception rational subjects establish or re-establish p-acceptance in the face of strong evidence to the contrary Self-deceivers must deal with strong evidence against p. We develop the notion of strong counter-evidence and its implications for the dynamics of self-deception in the following section. 2. Criterial evidence and the doxastic dynamics of self-deception From what has been said thus far, a core challenge for a theory of self-deception has become clear. Being a commonplace phenomenon, self-deception is a form of auto-manipulation that has to work in rational and evidence-sensitive subjects. At the same time, self-deception, as a matter of fact, requires that evidence be strongly asymmetrical. So, (iii) will shape how we have to specify PMD-condition (3) in a case of self-deception, i.e. S’s access and treatment of evidence. The point that notp-evidence has to be sufficiently strong before a belief that p can become self-deceptive has been at least implicitly acknowledged by many different approaches to self-deception. These comply with (iii) insofar they presuppose that sufficiently rational self-deceivers do truly believe what is suggested by the evidence, i.e. not-p. We will suggest a weaker alternative: It is sufficient if S considers or suspects that not-p in the face of an undeniable evidential challenge. But let us first evaluate the reasons one may have to ascribe a belief that not-p to self-deceivers. The simple reason for requiring S to believe that not-p is that – since S is basically rational and sensitive to evidence – S will form a belief that accords with the evidence. In addition, behavioral facts have been regarded as requiring S to believe that not-p. Audi (1988) and Rey (1988), among many others, have suggested that a belief that not-p is responsible for supposedly typical forms of self-deceptive behavior such as S’s deliberate avoidance of situations that will confirm not-p to S as well as S’s hesitance to act in accordance with p in some, if not in most or even all situations.17 A different line of motivation for the assumption that not-p is believed by S comes from the intentionalist camp. Davidson (1985, 1998), Pears (1991) and Bermúdez (1997, 2000) claim that S’s belief that not-p plays a crucial role in motivating and initiating the process of self-deception. What else, so one might ask, should move S to deceive himself in the first place, if not his believing that p is false? Thus, in addition to S’s strong motivation to believe that p, S must believe that p is actually false in order to engage in self-deception. Furthermore, the belief that not-p is actually true causes the cognitive dissonance and explains the psychological tension that have been proclaimed a characteristic feature of self-deception by various authors. So, it appears that there could be at least four good reasons to suppose that a person who is self-deceived in accepting p will also believe that not-p: – – – –
Rationality: Since S is sufficiently rational and evidence-sensitive, S simply recognizes that not-p. Behavior: The assumption that not-p is believed by S explains typical self-deception behavior. Motivation and activation: Only a belief that not-p explains why a process of self-deception is initiated. Phenomenology: S’s experience of a characteristic psychological tension is caused by his believing that actually not-p.
We need not fully endorse all four theoretical motivations for assuming a belief that not-p in order to recognize that they reflect a crucial question which any theory of self-deception has to face. In which way and to what result is not-p-evidence actually accessed and how can it be treated by S? This question is at the very heart of a theory of self-deception, since to answer it is to give a model of the cognitive dynamics of self-deception. Before we can provide such a model, we need to clarify the notion of strong evidence we presuppose in (iii). Counter-evidence is sufficiently strong if, under a neutral evaluation, the not-p-data are clear enough to destroy p-confidence. The kind of evidence we have in mind is best characterized as ‘criterial evidence’ against p (also referred to as criterial negative evidence ‘E ’ in the following). E is directly perceived as evidence against p and is a criterion for believing not-p. If S evaluates information as criterial evidence against p, S entertains a background-belief that this type of information makes the truth of p strongly unlikely. This meta-belief is part of S’s standard-rational evaluation of p-related information and defines what sort of data regularly defeats p. On the basis of this notion of criterial evidence we can develop a preview of what we take to be a very plausible model of the doxastic dynamics of self-deception. The process of self-deception is set off by recognized E in combination with a strong motivation to evaluate p as true. At the outset, if S is rational and sensitive to evidence, S will directly register relevant 17 To give an example: The fact that Peter knows that he and John are not actually best friends not only explains Peter’s absence of trust in John with regard to important matters of privacy but also his systematic avoidance of situations which strengthens the evidence – e.g., Peter now avoids joining when John and some of his friends go hang out, since he had already noticed on several occasions that John treated some of them with more privilege and in a much more attentive manner than him. Whereas psychological literature has investigated on the phenomenon of ’self-immunization’ in this context, we prefer to speak more generally about ’belief-immunization’.
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
735
data as E , that is, S will perceive and process these data as strong evidence against p. Accessed E triggers a belief-formation tendency towards not-p in S (at least at S’s first encounter with the data), i.e. S’s rational tendency of belief-formation will strongly suggest not-p and undeniably challenge p-confidence. However much S may desire that not-p be false, S is forced to consider or suspect strongly that not-p, if he is capable of a rational evaluation of p-relevant data in the first place. If we tried to imagine S as ‘blinded’ to E or E as being filtered out of S’s overall body of evidence, the conditions of rationality and evidence-sensitivity as well as the access-condition would be implausibly weakened. As described so far, the process is perfectly rational. S enters self-deception only by a subsequent pseudo-rational maneuver in which S attempts to invalidate the undeniable challenge put to p by E . This process has been described as a process of ‘‘immunization” in the psychological literature18 and we will characterize its pseudo-rational quality in Section 4. Assuming that not-p only strongly suggests itself to S, without assuming that S actually believes that not-p has two main advantages. Firstly, the belief-status of p can be maintained, there is no conflict with (i), i.e. there is no threat of a static paradox. Secondly, the aforementioned four reasons for assuming settled not-p beliefs can be accommodated by the alternative attitude without exception. Considering on the basis of accessed E that not-p is very likely true will do. First, in order to account for S’s rationality and evidence-sensitivity, it is sufficient to assume that S suspects not-p on the basis of E . Second, if S is forced to suspect that p is very probably false, this will at some point cause typical behavioral asymmetries between S and S, where S is a self-deceiver and S believes p without strong evidence to the contrary. It will, for example, easily suffice to explain why and how S can identify and avoid the situations he expects to defeat p.19 Thirdly, criterial counter-evidence and S’s subsequent considering of not-p are perfectly sufficient to motivate and initiate a process that can be characterized as selfdeceptive. Lastly, E exerts pressure on S’s desired belief that p. S’s believing p together with his considering not-p is sufficient to evoke cognitive dissonances and psychological tension, if S wants to evaluate p as true. In summary, we suggest that S’s consideration of not-p on the basis of criterial evidence against p is the sufficient and adequate way of spelling out the condition of access to evidence. No settled belief that not-p is necessary. We have argued that a commitment-free consideration of not-p by S is a plausible consequence of S’s access to criterial not-p-evidence. In our approach to self-deception we assume that the only belief self-deceivers need to commit themselves to is p. What needs explanation then is exactly how minimally rational subjects manage to hold or establish non-rationally motivated beliefs against undeniable challenge and to attain confidence in p. As we will argue in the following sections, this is possible by pseudo-rational reorganizations within S’s ego-centric belief-system. These reorganizations serve the purpose of defending p by invalidating the evidential challenge. In a case of self-deception, contrary to rational belief-revision, the desired invalidation of E is achieved by adaptation-strategies that are objectionable from the rational point of view. Pseudo-rational reorganization of p-relevant belief-sets amounts to a rationalization of S’s commitment to p in the face of criterial counter-evidence. Severe evidential challenge forces S to rationalize his commitment to p if S wants to retain or establish p-confidence. We take it as fairly plausible to assume that being a sufficiently rational subject, S’s beliefs are anchored in a rational belief-system, i.e. S takes p to be answerable to evidence and reasons. Even if our reasons for beliefs are not the only or not even the dominating causes of our beliefs that p, it will be rather difficult for us to retain p-confidence if (a) p faces a massive challenge and (b) in facing it, we find ourselves unable to explain how we can commit ourselves to p from a rational point of view. In that sense, self-deceptive rationalization should make p look rationally ‘committable’ (at least) to S, i.e. S should be able to see p as supported by evidence and reasons. Sanford (1988) has developed a rationalization-analysis of self-deception about one’s reasons for actions. Most of us require their reasons for action to be ‘ostensible’20, i.e. these reasons should comply with our self-image as rationally and morally responsible actors. In an analogous sense, beliefs must be committable in order to be embraced by a believer who is minimally ‘epistemically responsible’ about his beliefs. Nonrational commitments to the truth of p that are merely based on desire or emotion would be cases of straightforward irrationality. In self-deception S makes P rationally committable via changes among his background-beliefs. This amounts to a rationalization of S’s commitment to p in the face of E . This motivation-based rationalization will immunize p against E -type evidence. S’s rationalization of his commitment turns into self-deception, (a) if that process puts elements of rational thinking to work against the standard-rational evaluation of the subject matter, (b) if the standard-rational evaluation still remains operative when S evaluates neutral cases21 on the basis of equivalent information, and (c) if rationality-based p-immunization is ad hoc and enjoys priority over the pursuit of truth. We will illustrate in Section 4.2, how self-deceivers characteristically engage in such a dual standard of rationality. In self-deceptive rationalization, S’s capacity of rational thinking cannot only be blinded but must be used as a tool of auto-manipulation. To keep or make p rationally committable, S has to adapt his belief-set in a way that invalidates criterial counter-evidence and saves the desired conclusion. Therefore pseudo-rational adaptation marks a level that can outperform alternative forms of auto-manipulation when counter-evidence is strong. The preceeding sketch of the cognitive dynamics of self-deception shows how we can assume that it is sufficient for selfdeception that S holds only p throughout the whole process (and does not believe the contrary). ‘One-belief views’ avoid the puzzles of self-deception and make it easy to satisfy (ii) by rejecting the two-belief-requirement. In the next section we will examine influential, alternative one-belief views. There are two main types of theories. One claims that in self-deception S 18 19 20 21
See Sections 4.1 and 4.2. This point about suspicion has also been made by van Leeuwen (2007, p. 427). Sanford (1988, p. 157). See PMD condition (5).
736
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
believes that not-p, while p-acceptance is not situated on the level of belief. We consider Tamar Szabó Gendler’s recent ‘pretense’-model in this context. The other represented most prominently by Alfred Mele, holds that motivationally biased belief-acquisition only involves a belief that p. 3. What do self-deceivers believe? 3.1. The ‘pretense’-view Audi (1982, 1988), Rey (1988) and others have proposed that there is a gap between what self-deceivers believe and what they avow. They claim that self-deceivers actually believe that not-p and that p-acceptance lacks the essential property of belief-states: being action-relevant. Recently, Tamar Szabó Gendler has claimed that self-deceivers do not believe, but merely ‘pretend’ that p. Her view is more intuitive and more adequate with regard to the behavior of self-deceivers than, e.g. the version of Rey. It avoids a strict avowal-action-gap by conceding to ‘pretense’ a scope of action-motivation that is limited, but nonetheless exceeds the level of mere avowal. She characterizes a self-deceptive pretense-mode of mind in the following way: ‘‘A person who is self-deceived about [p]22 pretends (in the sense of makes-believe or imagines or fantasizes) that [p] is the case, often while believing that [not-p] is the case and not believing that [p] is the case. The pretense that [p] largely plays the role normally played by belief in terms of (i) introspective vivacity and (ii) motivation of action in a wide range of circumstances.” (Szabó Gendler, 2007, p. 233f). ‘Pretense’, ‘imagination’, etc. are used in a technical sense here. In contrast to belief, which is defined by Szabó-Gendler as a ‘receptive’ attitude in the service of truth, ‘pretense’ is characterized as ‘projective’. Instead of pursuing truth, self-deceivers mentally escape from the not-p environment and create for themselves a ‘p-world’. This basically makes them inhabitants in two worlds; one in which they rationally believe not-p and another one in which they actively enclose themselves in an imaginary p-world that is buffered against hostile evidence and kept from rational control. Since S cannot believe p, he can just ‘pretend’ or ‘fantasize’ that p. ‘Pretense’ governs action in a kind of encapsulated fantasy-world-mode of self-deception and beliefs govern action in the actual-world-mode. Self-deceivers act on their ‘pretense’ as long as no other motivation overrides the original motivation to stick to p and forces them back into the real-world-belief that not-p. The behavioral profile of ‘pretense’ actually looks very ‘self-deception’-like. And in contrast to the avowal-view, ‘pretense’ clearly satisfies condition (ii); it seems not only real, but also commonplace. Moreover, it addresses (iii), since not-p evidence typically leads to not-p beliefs and p-fantasizing. ‘Pretense’ is a real phenomenon of auto-manipulation. The familiarity of ‘pretense’ (and the fact that ‘pretense’ is clearly not paradoxical) and the similarities between the behavioral patterns of ‘pretense’ and self-deception motivate the general claim: self-deception, if we look at it, is typically only ‘pretense’. However, there are several reasons to be skeptical about this move. Szabó Gendler claims that the view seems ‘‘natural” and she wonders how this obvious explanation has been overlooked. A probable reason is that no one would commonly consider a normal p-pretender to be self-deceived, and therefore, the general claim that all self-deceivers are only engaging in a sort of pretense, imagining or fantasizing is counterintuitive rather than natural. Most, to the contrary, share the intuition that what self-deceivers do, is more like believing p than pretending p. A self-deceptive ‘pretender’ or ‘fantasizer’ is someone who seeks ways of avoiding acknowledging and facing what he knows. He engages in a kind of wishful thinking whereas classically conceived, a self-deceiver seems to be someone who actually gets things wrong and who will defend his doxastic p-commitment, if challenged. Thus, an argument is required to show why it is more intuitive to characterize S as pretending than as believing that p. But apart from the question whether the ‘pretense’-analysis has our intuition on its side, it is not clear whether ’pretense’ is fit to explain self-deception and at the same time to contrast with belief-status. Moreover the phenomenon of ‘pretense’ can easily be acknowledged as a phenomenon of auto-manipulation but remain distinguished from self-deception. In contrast to mere pretenders or imaginative thinkers, Szabó Gendler’s ‘pretenders’ must develop sufficient confidence in p, if they are to avow p sincerely. So, it is obvious that ‘pretense’ must differ from ordinary pretense or fantasizing on the one hand, and from belief on the other. This is the case because: (a) ordinary pretense and fantasizing are by definition confidence-free, non-doxastic states and (b) belief is the thing the account wants to avoid. To play its explanatory role the new concept of ‘pretense’ must be a hybrid that eats its cake and keeps it, too. It has to be belief-like in explaining S’s pbehavior and p-confidence, while being sufficiently imagination-like so as not to conflict with S’s knowledge that p is untrue. This is hard to accept for in imaginative attitudes like ‘pretense’ there is no room for confident and sincere p-avowal as it constitutes the phenomenon of self-deception. Self-deception without p-confidence is not a serious option. A single pretender who’s not a liar, would not display his confidence in defending p, but it is natural to assume that a real self-deceiver would. We assume that it is part of the phenomenon of self-deception that self-deceivers will characteristically be ready to defend p, e.g. when p is directly challenged by other members of S’s community. If we want S to avow p sincerely and develop p-confidence, we should expect that S will explore ways of acquiring p-confidence by invalidating not-p-evidence.23 If S
22 23
Szabó-Gendler’s convention is changed into the convention of this paper, in which ‘‘p” designates the deceived target belief. If the ‘pretense’-mode of mind were to be immune against challenges, evidence-sensitivity and rationality would be lost.
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
737
turns out to be unable to defend his avowal by rationalizing it, ‘pretense’, and along with it self-deception will either break down or S would have to concede a straightforward irrationality in his p-confidence. So, if Szabó Gendler’s ‘pretense’ is like normal pretense, imagining or fantasizing, it conflicts with the requirement of sincere avowal and p-confidence. Moreover one will not be able to explain how self-deceivers establish and defend p against acute evidence and challenges by others. But if self-deceivers do establish and defend p-confidence in the face of challenge, we leave pretense behind and enter the realm of belief. Motivated ‘pretense’, ‘fantasizing’ and ‘imaging’ are ways to actively avert reality. This can be characterized as an especially strong form of wishful thinking. But averting reality and self-deception can and should be distinguished. They are different phenomena, even if they occasionally exhibit behavioral resemblances. Averting from reality is still is still an option, if self-deception breaks down. The ‘pretense’-model remains a static picture in which evidence and motivation go separated ways. S’s actual beliefs are not subject to motivational influence by definition. ‘Pretense’, if it is different from belief in any significant way, neither alters the subject’s confidence in not-p, not does it influence confidence in p either, since p is never believed. Hence, in the ‘pretense’-view, self-deception is a folk-psychological myth from the very beginning. Szabó Gendler’s initial distinction between belief as a merely ‘receptive’ attitude and ‘pretense’ as a ‘projective’ attitude is also too rigid. It underestimates our actual capacity to see or reinterpret a not-p-world as a p-world under the influence of our motivations. Self-deceivers, as we will see in the next sections, are able to manipulate the evidential value of unwelcome information and develop p-confidence against the evidence. In our view, self-deception is a cognitively dynamic process at the belief-level. This process is not properly described as a process of averting reality, but rather as a process where selfdeceivers incorporate the accessed evidence by making use of their capacity of rational reasoning as a tool of auto-manipulation. One might argue that the presupposed ability and readiness to defend and rationalize p does not necessarily indicate belief-level. But, as a simple matter of fact, rationalized p-commitments will often reach belief-status even against evidence. And there is no threat of paradox in assuming that they would. The condition of rationality and evidence-sensitivity and the condition that not-p-evidence must be strong, are compatible with p-acceptance on belief-level, as we intend to demonstrate in the following sections. Alfred Mele has demonstrated that motivation can be efficacious on the level of belief. We examine his account before developing our model of self-deception as pseudo-rational belief-defense. 3.2. Motivational bias In his influential deflationary analysis of self-deception as ‘motivational bias’, Alfred Mele (1997, 2001) holds that it is sufficient and typical that self-deceivers only acquire a belief in what they desire to be true, namely p, without entertaining any belief that not-p. Mele thinks that not-p-beliefs (or weaker replacements) are completely dispensable when it comes to explaining what happens in garden-variety cases of self-deception. Explaining garden-variety cases requires neither any ‘activator’ nor any sort of intentional effort on the part of the agent—nor should we assume that psychological tension is typical of self-deception. Rather, we can think of real self-deception as working much more smoothly and effortlessly. Beliefformation gets silently biased by our motivation. In his model Mele draws on empirical evidence about stable patterns of biased information processing in subjects, so-called ‘cold biases’: In pragmatic heuristics, for example, the ‘vividness’ and ‘availability’ of p-data, or p-‘confirmation bias’ (a stable inclination to search for confirming rather than disconfirming evidence of hypotheses) generally facilitate p-acceptance at a level far below that of rational scrutiny.24 Cold biasing is a form of silent data-selection or data-structuring that is efficacious entirely at a procedural, sub-intentional level. The essential aspect of self-deception according to Mele is ‘hot bias’, i.e. biased processing driven by motivation. S’s individual motivation to believe that p structures S’s accessing and processing of p-relevant information in a manipulative way and it is further claimed that such motivationally biased belief-formation is sufficient to explain garden-variety cases of self-deception. Thus, if S desires to evaluate p as true and if this desire biases the processing of p-relevant data and S is thereby led into accepting p, S enters self-deception. Mele gives the following jointly sufficient conditions for self-deception: Biased belief-formation 1. 2. 3. 4.
The belief that p which S acquires is false. S treats data relevant, or at least seemingly relevant, to the truth value of p in a motivationally biased way. This biased treatment is a non-deviant cause of S’s acquiring the belief that p. The body of data possessed by S at the time provides greater warrant for not-p than for p (Mele, 2001, p. 50f).
First, we want to focus on condition (2) and what it tells us about how S accesses and treats evidence. By speaking about ‘‘relevant data” being treated ‘‘in a motivationally biased way” Mele means that S’s motivation to believe that p shapes the body of p-data via biased misinterpretation of p-relevant data and/or via ‘selective focusing and attending’ and/or ‘selective evidence gathering’ (Mele, 2001, pp. 25–31). This way S’s motivation basically functions as a not-p-data filter and as a p-data amplifier: In biased evaluation of the body of p-relevant-data, not-p-supporting data are kept out of focus, get sorted out or downplayed, whereas p-supporting data pass through and receive high promotion. This motivational data-structuring determines S’s perception of the overall evidence and leads S directly into believing that p.
24
Mainly Nisbett and Ross (1980)
738
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
The question we would like to stress is: What does S actually ‘see’, when his belief-formation falls victim to such motivationally biased data-processing? What is S’s very own informational basis for confidence in p in that model? Mele explicitly accepts as an exception (2001, p. 52) that the biased mechanisms of data-selection can lead to a misperception of where the weight of the evidence truly lies. In this case, S non-intentionally accumulates an incomplete set of information about the subject matter. The available body of data which supports not-p from unbiased point of view is selected and gathered in a way that S finds the data supporting p since the information S actually has is incomplete. In that case, selective evidence gathering inhibits S’s collecting all relevant not-p-data. Then, contrary to (4), S is actually not in possession of all relevant data and as a consequence, will not access the evidential asymmetry indicated in (4) and consequently, S will not further have to deal with the evidence. Instead, S will develop a belief that accords with a set of data that is incomplete due to manipulation. In this case, S’s belief that p can be described as justified. In Mele’s view, S can nevertheless be self-deceived, even if the relevant not-p information is not in his possession due to motivational influence on the input of data. In the special case just described, the fact that the evidence supports not-p is not accessed by S. We will now elaborate further why in our view motivationally biased data-processing as described in Mele (2001) has general problems with the requirement that the relevant evidence has to be accessed by S. In the special case described above, contrary to condition (4), S is not in the possession of data that speak against p. However, a ‘misperception’ of the evidence seems to be rather typical for biased belief-acquisition, even if S is in possession of all relevant data. According to (4), Mele assumes that typically, S is in possession of the relevant data and treats them in a motivationally biased way. But what we think to be relevant here, is not access to data, but access to (counter)evidence. Even if S has access to all p-relevant data, it is not provided that S himself actually registers the objective evidential trend that the data would suggest to an unbiased subject, if S’s treatment of data is subject to the types of biased processing that Mele mentions. This is possible because data and evidence are situated at different epistemic levels. Two subjects can perceive the same set of data but perceive different evidential values of these data. Only evaluated p-data are evidence for or against p, and one and the same datum or fact can be perceived as having different evidential values. It is not perceptual data that we manipulate but their evidential quality. Desires can doubtlessly influence how we perceive the evidential status of some data. If John desires that Mary be interested in him, a smile from her will be an important datum for John that she might be interested in him, too. If he did not care about her, he might hardly notice it, let alone perceive it as a sign of interest. Motivationally biased perception of data, as characterized by Mele, is situated on the level of data-evaluation. S may collect all the relevant data, but non-intentional mechanisms of biased evaluation of data will highlight p-supportive data and feed them in S’s belief-formation process while data in support of not-p are left aside and downplayed. Therefore, S can access every single datum, while misperceiving25 the evidential tendency of all information. As a consequence, the evidential basis on which S establishes p-confidence is already the result of the biased treatment of data. When being biased, the world as S accesses it is a result of biased treatment of data, i.e. S fails to access the actual evidential value of data due to sub-intentional manipulation. In this sense, S’s motivationally biased treatment of data corrupts the objective weight of evidence and leads S into allocating the weight of evidence to the wrong side. A biased subject can thus collect the relevant data but remain nonetheless blind to the fact that the overall body of data provides greater warrant for not-p than for p. But then, it will be rather the rule than the exception in biased belief-acquisition that S is self-deceived but never aware of the fact that the evidence speaks against p. We think that explaining self-deception against accessed counter-evidence requires a form of treatment of evidence that is not present in Mele’s account. There can hardly be doubted that we possess the capacities of data-manipulation that Mele describes, and it is likely that they influence how we perceive the evidential status of data and that they lead us into false beliefs. But all this only works, as long as the not-p-data S receives is relatively harmless and does not imply a very strong tendency towards not-p. The problem of bias is that a mere biased view on a set of data offers no means to handle instances of criterial counter-evidence.26 Selfdeception, as we have constrained it above, requires a strong asymmetry of evidence, i.e. a strong tendency towards not-p. If that tendency were not strong, ambiguity would remain with regard to the question whether or not p, and hence S could be excused for simply keeping his commitment to p. But S cannot be excused from a rational point of view if he ignores criterial counter-evidence. Criterial counter-evidence, as introduced in Section 2 is directly perceived as evidence against p. Criterial counter-evidence breaks down p-confidence and forces S to consider that not-p is very likely true. Peter has believed that wife Mary is faithful. If he sees Mary passionately embracing John when they think they are unobserved, this will give Peter criterial evidence that he is wrong. Criterial evidence implies that S holds a stable meta-belief such that a perceived event of type X is not compatible with holding the desired belief p. The rule ‘‘If X, then very likely not-p” is a constitutive part of the standard-rationality that Peter endorses in everyday hypothesis-testing. Biased perception-structuring will hardly help him here, because criterial evidence simply cannot be merely misperceived in a biased way. In order to regain confidence in his belief, Peter is forced to update or reorganize his belief-system in a manipulative way if he is to regain confidence in p.27 The task in such reorganization can only be to invalidate the received criterial evidence against p. Hence, if S shall be able to believe that p against criterial counter-evidence, S’s treatment of this evidence will have to consist in the reorganization of his belief-system.
25 With regard to the perception of evidence or of evidential value, we use the term ‘‘perception” in a broad sense. The evidential value attributed to data is the result of an evaluation. In the case of morivational bias this includes top-down influences of motivation on attention and data-interpretation. 26 Similarly, ‘cold bias’ in data-processing will rather facilitate acceptance of p in S in the absence of sufficient evidence than in would be able to establish acceptance of p, if S possesses good evidence against p. 27 Mele’s own examples for ‘‘positive” and ‘‘negative misinterpretation” can also be read as examples of self-serving adaptations in belief-systems, but we don’t have the space to show this here. See Mele (2001, p. 26f).
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
739
Could not biased evidence filtering alone be strong enough to block even criterial counter-evidence and thereby S’s beliefformation tendency towards not-p? The answer is obviously no, because then, the necessary condition of evidence-sensitivity would not be satisfied, nor would the condition of minimal rationality; believers who are simply insensitive to criterial evidence would be considered victims of mono-thematic irrationality. Cases of lost sensitivity to criterial evidence are located outside the field of garden-variety self-deception even in from a pre-analytic point of view. A strong lack of sensitivity to evidence indicates a phenomenon of delusion rather than a phenomenon of garden-variety self-deception. Thus, biased structuring of evidence-perception as described by Mele is not sufficient to satisfy the criteria of adequacy we have put forward. To achieve that aim we need a reorganization of the subject’s belief-system. Such a reorganization makes believers capable of invalidating criterial evidence. We will explain below, how pseudo-rational reorganization of belief-systems works and why it makes sense to view it as the true criterion of commonplace self-deception. To summarize the argument thus far: Pretense and bias are real cognitive processes. Both of them exhibit considerable similarities to self-deception but both finally fail to describe self-deception in a satisfying way. Both have problems in conceiving of self-deception as a dynamic process of belief, albeit for different reasons. Trying to avoid the classical paradoxes, pretense must deny p-acceptance belief-status. Mele’s account does justice to the intuition that self-deceivers believe that p, but biased evidence-perception cannot handle the constraint that self-deception requires strong counter-evidence. Selfdeceivers are not only biased and they are not only pretenders for they essentially make use of their rational capacity of adapting belief-systems. Neither the ‘motivational bias’-model nor the ‘pretense’-model offers an account of how a strong evidential tendency towards not-p could be handled by self-deceivers. To deny that it could, would amount to denying paradigmatic cases of self-deception. As a commonplace phenomenon, biased perception-structuring will typically contribute to processes of self-deception. But the manipulation of criterial counter-evidence exceeds the level of perception-structuring and exploits our capacity to defend target-beliefs against certain types of undeniable challenge. This process has been described in the empirical literature by Wentura and Greve (2003, 2005). In what follows, we characterize this process as ‘pseudo-rational’ revision in ego-centric belief-systems. 4. Ego-centric belief-systems and dual rationality 4.1. Adaptation in ego-centric belief-systems In this section, we seek to spell out a rationalization model of self-deception and to provide evidence for the claim that it satisfies (i), (ii), and (iii)—namely, that it is free of paradox, that it is an actual and commonplace phenomenon and that it explains how self-deception can cope with criterial counter-evidence, respectively. Our suggestion is to model self-deception as a special class of revisions in ego-centric belief-systems. In ego-centric beliefsystems, beliefs are ranked not only by rational criteria of importance such as their explanatory role, their being highly integrated in a coherent belief-network and their being tested and proven, but also by levels of subjective importance. If a belief has a high subjective importance for S, S has a strong motivation to maintain that belief. According to psychological evidence, motivated belief-formation will be the rule rather than the exception in humans. In everyday belief-revision, there is evidence that the organization of our belief-systems is frequently governed by subjective importance in addition to rational considerations. Typical examples of beliefs of high subjective importance are beliefs about ourselves that are related to self-esteem, but also beliefs that are of emotional value or desirable to S in other ways.28 These may include evaluations relevant to one’s self-image like pride and shame, but also the evaluation of other persons or matters.29 If a belief that p plays a central role in S’s ego-centric belief-system due to its high subjective importance to S, the revision-threshold for p will be correspondingly high. Models of everyday hypothesis-testing30 indicate that, dependent on S’s degree of motivation to believe p, S will in general require more and weightier evidence to reject p than to reject not-p and conversely require less and thinner evidence to embrace p than to embrace not-p.31 This asymmetry in revision-threshold is a first constant of motivated hypothesistesting. Another component of motivated belief-acquisition concerns processes of biased perception-structuring like selective focusing and attending as well as tendentious evaluation of the weight of evidence. They have been described in different ways by theorists like Mele (1997, 2001), Talbott (1995) and van Leeuwen (2008). These processes or strategies serve a non-rational immunization of beliefs of subjective importance. But as long as S is sufficiently rational and sensitive to evidence, we are in need of an additional strategy if we want S to gain p-confidence in the face of criterial counter-evidence. An asymmetrical revision-threshold and biased evaluation of data would not do in cases where the evidence is too plain to be overlooked. We claim that this can only be managed by reorganizing our local belief-sets about the subject matter in question. Self-deceivers characteristically adapt those background-assumptions which constitute the criterial function of E for p. The notion of selfdeception as an immunization of beliefs (typically about ourselves) of high subjective importance against criterial evidence
28
For the role of emotion in self-deception see Sahdra and Thagard (2003) and Mele (2000). For example, if Peter is deceiving himself about Mary, this may be due to the fact that his relationship with Mary deeply affects his self-image or that a certain image of Mary or his relationship to her has high emotional significance to him. Each motivation can be sufficient to trigger a process of self-deception, but they may also coincide. If Mary is Peter’s wife and gives strong evidence that she is cheating on him, Peter’s self-deception that she is faithful can be motivated by his self-esteem as well by the purely emotional significance this fact has to him. 30 Trope and Liberman (1996). 31 In addition, in biased hypothesis-testing S will go for confirming p instead of not-p. See Mele (2001, p. 32). 29
740
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
is well-supported by empirical investigation. Experiments by Wentura and Greve (2003, 2005) have convincingly revealed that there is an adaptation of trait-definitions that serves self-immunization. They were able to show that subjects automatically adapt trait-definitions in a self-serving way (like, e.g., ‘being erudite’) if (a), those traits are highly valued parts of their selfimage, and if (b), the subject is confronted with criterial counter-evidence. Via questionnaires and semantic priming tasks, Wentura and Greve confirmed the hypothesis that subjects re-shape trait-definitions according to their actual skills: ‘‘Confronting (student) participants with gaps in their general knowledge will result in a pattern of self-immunization that defines the trait eruditeness in terms of those pieces of knowledge that the participant has successfully demonstrated. Faced with undeniable challenges (failing in a general education quiz), subjects preserve their self-image and incorporate the data by modifying what counts as evidence for ‘being erudite’.” (Wentura & Greve, 2003, p. 33). According to Wentura and Greve, subjects develop idiosyncratic concepts of the trait ‘eruditeness’ in response to what we call criterial counter-evidence. This strategy allows S to keep the valued or esteemed parts of his self-image intact.32 S will adjust his belief-system as a consequence of accessing criterial evidence against beliefs that have high subjective importance. Relevant adjustments will allow S to commit himself to p regardless of the serious challenge posed by counter-evidence. If S has run out of ad hoc-explanations for why p is unaffected by what looks like paradigmatic instances of counter-evidence, S can immunize p against certain types of counter-evidence. In the example above, S has the following doxastic profile before being confronted with his failure: (p) ‘‘I am erudite” (rated by S as an important part of his self-image). (q) ‘‘Knowing about history is necessary for being an erudite person”. (r) ‘‘I have good knowledge of history”. The belief to be preserved is p; the belief that gets refuted by E (the history-test failure) is r. The two beliefs p and r are inferentially related via a connecting belief q. This belief q is the trait definition, i.e. it defines what counts as criterial evidence against p. In this sense, ‘being erudite’ is, as Wentura and Greve note, a theoretical term, and q is part of S’s theory TE about eruditeness in which S fixes all criterial evidence for or against p. 33 In self-deception, S will typically change TE in order to save p. How can S change his theory of eruditeness TE? If S takes r to be defeated by E , S will typically adapt or simply waive q in a way which blocks the inference from not-r to not-p. Additionally, S may introduce supplemental beliefs s and t that justify changing or waiving q. By TE changes, E -type counter-evidence is discarded (e.g. by giving up the claim that historical knowledge is necessary for being erudite). But in large, q can even remain intact (and imply r) if S finds other loopholes for denying that the received information is valid against p. S might say: ‘‘True, being erudite essentially includes knowing about history, but since I’m German, e.g. American history is not part of core-proficiency in history. Besides, knowing about exact dates of battles etc., is not the essence of a true understanding of historical processes at all.” In similar fashion, S can introduce additional conceptual distinctions and subtleties which serve the function of making the inference to not-p questionable: given that E has the effect of establishing not-r, then conceptual distinctions may change the claim in q, thereby restoring confidence in p. In experiments by Wentura and Greve, subjects established congruency between TE and their actual abilities: ‘‘Accordingly, self-immunization means assigning less weight to those observables an individual believes himself to be poor at and more weight to those observables he believes himself to be good at. As a consequence, the immunized concept (‘erudite’) is maintained against a certain failure (‘‘I don’t know when Caesar was murdered”) by reshaping its connotation (‘‘It’s not historical education, but, say, geographical knowledge what counts”).” (Wentura & Greve, 2003, p. 31). Beliefs that are connected to observables only via a set of connecting beliefs about a criterial function of r for p invite selfdeception. S will hardly succeed in deceiving himself about the immediately observable fact that he has failed the history test. In general, q-type connecting beliefs are not as easy to rebut as r-type beliefs. They provide self-deceivers with room for adapting their definitions and for constructing subtleties and ambiguities that they have never stressed before which allows them to reject that the target-belief that p is finally defeated. In our special case, TE, the theory of ‘being erudite’, is notoriously under-determined. TE provides S with the space for re-interpretation that lies at the basis of the stability of self-deception and in practice avoids a regress: TE and q are answerable to evidence, but they contain complex, theory-based terms that are only remotely observational (Newen & Bartels, 2007). Therefore TE-adaptations are able to provide a relative stability for self-deceptive beliefs. S could for example also increase stability of self-deceptive adaptations by conceding to himself superior insights into what true eruditeness does comprise.
32 There is additional empirical evidence for a strong tendency of test-subjects to modify their standards of evaluation in a way that allows them to maintain their self-esteem. See cf. Beauregard and Dunning (2001). If we attribute properties to others like being intelligent or being good in mathematics, we do that according to an idiosyncratic standard that bears on our own abilities. If we are not good in mathematics we establish a low standard for being good in mathematics such that we can self-attribute to share that property if we have advanced competences, we develop a high standard for being good in mathematics such that almost no one besides ourselves passes it. These trait definitions have an egocentric standard and thereby allow us to keep our selfesteem. 33 We want to stress that we do not presuppose a rich concept of ‘theory’ here. A theory T in this case is merely a local set of beliefs with regard to a special subject matter SM. It involves no requirement of strong systematicity.
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
741
4.2. Dual rationality as a constitutive element of self-deceptive pseudo-rationality Up to this point it is still not entirely obvious whether S is self-deceived. As yet we have developed no criterion as to when the revision of TE is irrational or self-deceptive. It can be rational to defend important beliefs against prima-facie evidence, even if this evidence is criterial according to the old version of TE. What would make S’s move problematic is if S endorses two inconsistent standards of belief revision at the same time. Wentura and Greve (2003, 2005) as well as Beauregard and Dunning (2001) focus on self-related adaptation, but do not consider self-/other-asymmetries. But given the PMD-condition (5)34 and the constraint that self-deceivers endorse standard-rationality, we predict that self-deception will typically involve a dual standard of rationality (where S does not notice the duality). The phenomenon of dual rationality consists in the following disposition to evaluate cases asymmetrically. A revision of the relevant belief-set to neutralize criterial evidence applies only to matters of subjective importance. In dual rationality, subjects will keep their rational standards in general and at the same time introduce a new standard of evaluation with respect to particular matters of subjective importance. With regard to self-esteem, this will typically result in first-/third-person asymmetries. Peter will continue to regard himself as erudite, but the next day during lunch he judges his boss Frank by the old version of TE when Frank shows an embarrassing gap in his knowledge of history. Evidence supports that we are more generous with our ego-centric definitions of desirable traits than we are with the general definition we apply to others. There is evidence for a general tendency in human reasoning that supports the plausibility of such dual rationality in selfdeceivers. Apparently, we do not only use two systems of reasoning but also develop dual standards of evaluation, e.g. when we evaluate human abilities. Beauregard and Dunning (2001) not only prove that there is an ego-centric bias in evaluating abilities like being intelligent but they also make clear that there are two standards of evaluation involved, namely an intersubjective social standard of evaluating mathematical abilities, (which can be characterized by the results of standardized tests) and the ego-centric standard that we use if we evaluate unreflectively. In addition, there is evidence that we can establish dual rationality intrapersonally. There is a strong line of research arguing that people use two systems of reasoning (Evans & Over, 1996; Kahneman & Frederick, 2002, 2005; Reyna & Brainerd, 1994; Sloman, 1996; Stanovich & West, 2000): a primitive, intuitive heuristic, associative reasoning system on the one hand and a cognitively demanding analytic, rule-based reasoning system on the other hand.35 The question remains whether there is evidence that we actually use dual standards internally for evaluating relevantly similar packets of information. Evidence for intrapersonal dual standards of evaluation is for example given by Dunning and Cohen (1992) and Beauregard and Dunning (1998), who have demonstrated that persons refer to different definitions in describing their own performances and the performances of others. Furthermore, if the dual reasoning systems are used for the same task, then we should notice intrapersonal conflicts, since the two systems work independently and in parallel. Hence, they should produce different, competing outcomes. In fact, there is evidence for competing outcomes since measureable monitoring processes are implicitly involved in cases of dual rationality (de Neys & Glumicic, 2008). This supports our analysis nicely, since we have evidence that dual rationality can be established intrapersonally: in the demanding subjectively relevant cases (which are typically first-person), the intuitive, associative reasoning is allowed to triumph over the monitoring process, while in the neutral third-person cases the implicit monitoring process allows the reflective, rule-based reasoning to dominate. The latter especially demands universal employment of rules, while the former remains implicit, involves an ego-centric standard of rationality in subjectively relevant cases and leads to dual rationality. Dual rationality will typically be observable in ‘automatic’ or unreflected adaptation and this is one reason for criticizing S’s activity from a rational point of view. In the case of ‘automatic’ TE-adaptation, S’s own dual standard will not be transparent to S. However, room should be left for the possibility that reflective self-deceivers seek to avoid a dual standard and begin to generalize the revised version of TE also over parallel cases.36 But a strong generalization of TE involves the danger of revealing self-deception. Theoretically, S can iterate self-deception but extending self-deception in iterating the process and creating more dual standards will make the process less economical and make it more and more unlikely that S’s can retain pconfidence. Thus, dual rationality remains characteristic for self-deception, but the dual standard must basically remain opaque to S. What makes TE-change in the above described fashion pseudo-rational (in addition to the typical dual standard of evaluation) is the fact that the change is ad hoc, i.e. it can be identified as made on the basis of the subjective importance of p instead of on the basis of the pursuit of truth. To follow a simplified Quinean standard, rational processes of revision are mind-to-worlddirected, i.e. the primary touchstone for the whole system is the external world. If adaptation-processes obviously violate this primacy and tend towards belief-immunization, the revision will be criticized for following a motivation external to the norms
34
Condition (5) states: If S had been neutral towards SM, the same evidence would have led S to accept not-p instead of p. The normal cognitive development seems to involve a process in which the domain of reasoning is first treated by associative processes before we develop more and more rule-based analytic reasoning. Furthermore, there is evidence that a specific mistaken evaluation in an everyday task called ‘base-rate neglect’ is a consequence of using the associative reasoning system. The effect is reduced if we acquire rule-based reasoning for that domain (Barbey & Sloman, 2007). We can diminish the impact of strong bias—that is constituted by ignoring relevant knowledge and relying on associative reasoning—by learning to reason in a rulebased way. 36 This strategy will include cases which are not immediately self-related into the scope of subjective importance. If not only Peter but also Mary failed the history-test, Peter might generalize TE and thereby also immunize Mary’s status of being an erudite woman against history-test evidence. This strategy is compatible with self-deception but it will be risky for Peter to pursue it too strongly since it direcly exposes revised TE to falsification. If Mary is honest and not willing to participate in collective self-deception, John will encounter immediate counter-evidence against the viability of revised TE. If Peter is insistent, he will be forced to iterate the process and, e.g. discredit Mary’s opinion. Typically, the iterated process will create a dual standard again. 35
742
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
of rationality. Moreover, in self-deception the resulting beliefs will typically be false. As a matter of fact, S will typically be wrong about p, since in most cases there is good reason why a connecting beliefs q is part of the standard form of TE. Dual rationality and the tendency towards irrationally motivated immunization justify the characterization of S’s maneuver as ‘pseudo-rational’. 4.3. Conclusion, advantages and scope of the model We have argued that self-deception consists in p-commitments on the basis of pseudo-rational adjustments of the beliefsystem. The adjustments invalidate criterial evidence against p. Self-deception typically involves a dual standard of evaluation. The proposed model reconciles two requirements: Self-deceivers are sufficiently rational and sensitive to evidence and they can manipulate their beliefs in the face of criterial evidence to the contrary. Self-deception can be characterized as a cognitively dynamic process of keeping or establishing belief in p without having to assume that S must believe not-p. As has been pointed out, pseudo-rational adaptation is particularly suitable to meet the strong counter-evidence requirement. But there is one more reason for seeing rational reorganization as a necessary feature of self-deception. As a technique of auto-manipulation, motivated reorganization obviously differs from phenomena like attention control, selection of evidence, control of memory retrieval and control over contents of conscious thought as described by Talbott (1995) and van Leeuwen (2008). It can be questioned whether those revision-free phenomena of belief-control would not perform equally well against criterial counter-evidence in sufficiently rational subjects. We cannot discuss this here. But in our view, there is nevertheless one strong intuitive aspect of the phenomenon of self-deception that makes pseudo-rationality seem indispensable. As indicated in Section 3.1, we see it as part of the phenomenon that self-deceivers defend their target-beliefs in social interaction. If challenged by someone’s pointing out the weak evidence for p, a true self-deceiver will argue for his commitment rather than shy away from it and will defend his confidence. In many cases, self-deceivers might resist giving up p even more than normal unmotivated believers would. If we want self-deceivers to be able to resist rational belief-adaptation, there seems hardly a way around the rationalization strategy. By pseudo-rationality, motivated beliefs can be successfully anchored in rational discourse at least to a limited extent. Pseudo-rationality is not only a successful ‘technique’ to arrive at desired conclusions but also a requirement for self-deception among members of rational discourse. Several forms of auto-manipulation can be active at the same time and lend mutual support in S’s auto-manipulation. However, pseudorationality is necessary for self-deception in rational subjects in addition to lower levels of motivated informationprocessing. The ‘pseudo-rationality’-model satisfies all three strict criteria of adequacy: (i) the developed picture of the cognitive dynamics of self-deception is paradox-free. Pseudo-rational immunization of p against counter-evidence involves neither a dynamic nor a static paradox. (ii) The developed view describes a real psychological phenomenon, as the empirical evidence about the adaptation of trait-definitions and dual rationality suggests. (iii) The model explains how self-deception (in contrast to other forms of auto-manipulation) in rational subjects can operate against criterial evidence. Beyond the strict criteria of adequacy, the model provides ways of doing justice to a set of further intuitions and observations that are connected to self-deception. When engaging in more laborious pseudo-rational defense of p, subjects can feel that self-deception did not just happen to them, but that they have played an active part in a manipulative process. But, as Wentura’s and Greve’s studies indicate, pseudo-rational defense also may proceed in a rather automatic way. Conscious defense of p against p-hostile evidence seems possible, as long as (a) the pseudo-rational quality and dual rationality remain opaque to the subject and (b) as long as S need not think or recognize that his p-commitment is improperly founded.37 We prefer not to characterize self-deception in contrast to other forms of auto-manipulation by requiring it to be intentional. From our point of view, the question of whether self-deception is intentional in any good sense is an issue of secondary importance. All we require for self-deception is pseudo-rational adaptation within a belief-system, no matter whether adaptation is automatized or not. All that has to remain opaque to S is the pseudo-rational quality of his p-commitment. Since the model describes vivid cognitive dynamics, it can account for psychological tension as a typical feature of selfdeception. However, in many cases of self-deception, psychological tension may be low or absent, as Mele suggests. The likely reason for an absence of tension is that self-deception—like any mental activity—becomes habitual. If subjects develop idiosyncratic patterns of interpretation for certain subject matters, they become ‘trained’ self-deceivers. A repeated application of the revised theory of ‘eruditeness’ may reduce tension, e.g. in future history-test failures. Moreover, the fact that subjects register criterial counter-evidence explains why self-deceivers can sometimes be only slightly surprised when self-deception breaks down and they acknowledge that they got it wrong. Finally, the ‘pseudo-rationality’ model even provides a parallel between self- and other-deception: Making p acceptable for others requires, in principle, abilities and strategies similar to those we use for making p committable for ourselves. The model is developed as a model of self-deceptive belief-defense. It is important to note that it can cover self-deceptive belief-acquisition and self-deceptive belief-retention as well. If the matter of p has subjective importance for S, S can develop pconfidence against a not-p environment by a similar strategy in both cases. It is no necessary condition that a belief that p must have been established in S beforehand in order for S to exploit the strategy of pseudo-rational adaptation with regard
37 Talbott (1995) has indicated how far self-deception might involve non-paradoxical intentional aspects that differ from straightforwardly self-deceptive intentions, i.e. intentions with the literal self-deceptive content ‘‘deceive yourself!”. Our account is open for like notions of intentionality.
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
743
to an issue of subjective importance. Pseudo-rational adaptation can be used to manipulate the evidential value of any data to fit any target-belief, no matter whether this target-belief is to be retained or yet to be established. In both cases self-deception is a rationalization of commitments that S is making on motivational grounds. In an ego-centric belief-system, establishing new beliefs about matters of subjective importance can for example also serve as a stabilization of the self-image, to the effect that single beliefs are acquired in order to stabilize and defend an already established set of motivational beliefs. We have dispositions to embrace certain types of new beliefs and to commit ourselves to p on motivational grounds (e.g. for the purpose of stabilizing an already existing set of motivated beliefs), even if S has never considered p before. If S has a motivation to accept p and p enters consideration, contradicting evidence can be invalidated and p-confidence can be developed on the basis of pseudo-rational adaptation. To demonstrate the flexibility of pseudo-rational adaptation, let’s look at an example of belief-acquisition via Tx-change in which S acquires a belief to the contrary of what he had believed before: John, an advanced graduate student, has never considered himself an even reasonably good philosopher. He has always felt behind his peers in most of the aspects that commonly count as relevant for having any career-prospects in the field. Therefore, he himself has believed for almost all of his graduate period that academic philosophy is definitely not his terrain. However, as John has almost finished his studies, questions about his future become salient and start pressing him. But John is far too lazy to even think about starting anything new and he is basically unwilling to acknowledge that he did not make the right kind of investment for his life. But John also knows well that even if the fact of someone’s being rather weak as a philosopher combined with his having poor talents in explaining philosophical problems to students and his being a fairly misanthrope personality does not make it impossible to get a philosophy job, it will nevertheless for sure mean having an awful life. In his situation, John is strongly motivated to acquire the belief that he is actually an exceptional philosopher, endowed with basically just what it takes to become a good academic philosopher. And in fact, he does change his opinion about what is criterial for someone’s being likely to become a successful academic philosopher and sincerely comes to believe that his great days in philosophy are about to come quite soon. It is possible for him to do so by revising his old theory about the issue, convincing himself that his former belief that he has no talents in the field was the result of a corruption of his mind by misguided standards and unreliable criteria for good philosophy, that his talents and insights have been overlooked because his individual approach does not conform with the expectations, that his achievements have not been properly evaluated due to the mainstream’s lack of sensibility for his approaches. He develops the belief that the whole education- and career-system builds on false priorities, that fundamental mistakes are involved in how philosophy is pursued in departments nowadays, etc. This way John brings his plan into harmony with the evidence about his performance, talents and personal traits. In the absence of job alternatives, John confidently decides to pursue a career as a philosophy professor. Motivated belief-change via change of background-theories is common, and it is the involved pseudo-rationality including dual rationality that will decide whether S’s belief-change is finally to be evaluated as self-deceptive. In principle, the technique of pseudo-rational T-revision seems capable of transforming the evidential status of any information towards a desired result, independent of the subject’s current state of belief and the source of motivation. Neutral data can be transformed to E+ or E , in addition, both can be invalidated to E0 and even E turned into E+ or vice versa38. This flexibility can be used to account for the problematic case of twisted self-deception39 as well. Peter deceives himself into believing that his wife Mary is unfaithful despite having rather good evidence for her faithfulness. It is difficult to clarify the motivational structure of such cases from the armchair, since the acquired belief seems undesirable and it is controversial as to how such cases are correctly characterized from a psychological point of view. But regardless of how the motivational structure of these cases is to be analyzed, Peter could basically proceed as described above. Peter’s revision of his view about what types of behavior indicate unfaithfulness can attribute to neutral or weak data a high evidential status for ‘‘unfaithful”, (e.g. Mary’s admiring a common friend and her engaging in intimate conversation). As a consequence of his new evaluation of evidence, Mary’s in fact unaltered behavior will now be perceived by Peter as indicating unfaithfulness. Another women’s similar behavior would not, since Peter, in line with his old theory, perceives her behavior not as diagnostic of unfaithfulness. Admittedly, twisted cases are tricky but the ‘pseudo-rationality’-model can offer a strategy in case such twisted self-deceivers are sufficiently rational and evidence-sensitive. It seems that the technique of pseudo-rational adaptation is basically available for all kinds of self-deceptive endeavors. Thus, the ‘pseudo-rationality’-model seems to provide an attractive account of self-deceptive auto-manipulation. Acknowledgments The authors wish to thank Alex Byrne, Amber Griffioen, Eduardo García Ramírez, Gottfried Vosgerau and Louise RöskaHardy for very helpful comments on earlier drafts and for discussions on the topic that have led to improvements of the paper. The paper has also benefited much from the remarks of two anonymous reviewers. References Audi, R. (1982). Self-deception, action and will. Erkenntnis, 18(2), 133–158. 38 An example of the latter is Mele’s example of ‘positive misinterpretation’ where Sid’s interpretation turns Roz’ obvious and unambiguous refusal to date him into an encouragement (Mele, 2001, p. 26). 39 The term is introduced by Mele (1999).
744
C. Michel, A. Newen / Consciousness and Cognition 19 (2010) 731–744
Audi, R. (1988). Self-deception, rationalization, and reasons for acting. In A. O. Rorty & B. P. McLaughlin (Eds.), Perspectives on self-deception (pp. 92–120). Berkeley: University of California Press. Barbey, A. K., & Sloman, S. A. (2007). Base-rate respect: From ecological rationality to dual processes. Behavioral and Brain Sciences, 30, 241–297. Beauregard, K. S., & Dunning, D. (1998). Turning up the contrast: Self-enhancement motives prompt egocentric contrast effects in social judgments. Journal of Personality and Social Psychology, 74, 606–621. Beauregard, K. S., & Dunning, D. (2001). Defining self-worth: Trait self-esteem moderates the use of self-serving trait definitions in social judgment. Motivation and Emotion, 25(2), 135–161. Bermúdez, J. (1997). Defending intentionalist accounts of self-deception. Behavioral and Brain Sciences, 20, 107–108. Bermúdez, J. (2000). Self-deception, intentions, and contradictory beliefs. Analysis, 60(4), 309–319. Davidson, D. (1985). Deception and division. In E. LePore & B. McLaughlin (Eds.), Actions and events (pp. 138–148). New York: Basil Blackwell. Davidson, D. (1998). Who is fooled? In J.-P. Dupuy (Ed.), Self-deception and paradoxes of rationality (pp. 1–18). Stanford: CSLI. De Neys, W., & Glumicic, T. (2008). Conflict monitoring in dual process theories of thinking. Cognition, 106, 1248–1299. Dunning, D., & Cohen, G. L. (1992). Egocentric definitions of traits and abilities in social judgment. Journal of Personality and Social Psychology, 63, 341–355. Evans, J. St. B. T., & Over, D. E. (1996). Rationality and reasoning. Hove, UK: Psychology Press. Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute substitution in intuitive judgement. In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgement (pp. 49–81). Cambridge, MA: Cambridge University Press. Kahneman, D., & Frederick, S. (2005). A model of heuristic judgement. In K. J. Holyoak & R. G. Morrison (Eds.), The Cambridge handbook of thinking and reasoning (pp. 267–293). Cambridge, MA: Cambridge University Press. McLaughlin, B. P. (1988). Exploring the possibility of self-deception in belief. In A. O. Rorty & B. P. McLaughlin (Eds.), Perspectives on self-deception (pp. 29–62). Berkeley: University of California Press. Mele, A. R. (1997). Real self-deception. Behavioral and Brain Sciences, 20, 91–102. Mele, A. R. (1999). Twisted self-deception. Philosophical Psychology, 12, 117–137. Mele, A. R. (2000). Self-deception and emotion. Consciousness & Emotion, 1(1), 115–137. Mele, A. R. (2001). Self-deception unmasked. Princeton: Princeton University Press. Newen, A., & Bartels, A. (2007). Animal minds and the possession of concepts. Philosophical Psychology, 20(3), 283–308. Nisbett, R., & Ross, L. (1980). Human inference: strategies and shortcomings of social judgments. Englewood Cliffs, N.J.: Prentice Hall. Pears, D. (1991). Self-deceptive belief formation. Synthese, 89, 393–405. Rey, G. (1988). Toward a computational account of akrasia and self-deception. In A. O. Rorty & B. P. McLaughlin (Eds.), Perspectives on self-deception (pp. 264–296). Berkeley: University of California Press. Reyna, V. F., & Brainerd, C. J. (1994). The origins of probability judgment: A review of data and theories. In G. Wright & P. Ayton (Eds.), Subjective probability (pp. 239–272). Oxford, England: John Wiley & Sons. Sahdra, B., & Thagard, P. (2003). Self-deception and emotional coherence. Minds and Machines, 13, 213–231. Sanford, D. H. (1988). Self-deception as rationalization. In A. O. Rorty & B. P. McLaughlin (Eds.), Perspectives on self-deception (pp. 157–169). Berkeley: University of California Press. Sloman, S. A. (1996). The empirical case for two systems of reasoning. Psychological Bulletin, 119, 3–22. Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: Implications for the rationality debate. Behavioral and Brain Sciences, 23, 645–726. Szabó Gendler, T. (2007). Self-deception as pretense. Philosophical Perspectives, 21, 231–258. Talbott, W. J. (1995). Intentional self-deception in a single coherent self. Philosophy and Phenomenological Research, 55, 27–74. Trope, Y., & Liberman, A. (1996). Social hypothesis testing: Cognitive and motivational mechanisms. In E. Higgins & E. Kruglanski (Eds.), Social psychology: A handbook of basic principles (pp. 239–270). New York: Guilford Press. Van Leeuwen, N. (2007). The product of self-deception. Erkenntnis, 67, 419–437. Van Leeuwen, N. (2008). Finite rational self-deceivers. Philosophical Studies, 139, 191–208. Wentura, D., & Greve, W. (2003). Who want to be. . . Erudite? Everyone! Evidence for automatic adaptation of trait definitions. Social Cognition, 22(1), 30–53. Wentura, D., & Greve, W. (2005). Assessing the structure of self-concept: Evidence for self-defensive processes by using a sentence priming task. Self and Identity, 4, 193–211.
Consciousness and Cognition 19 (2010) 745–750
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Approaching self-deception: How Robert Audi and I part company q Alfred Mele Florida State University, Department of Philosophy, Tallahassee, FL 32306-1500, United States
a r t i c l e
i n f o
Article history: Available online 7 July 2010 Keywords: Belief Deception Experimental philosophy Self-deception
a b s t r a c t This article explores fundamental differences between Robert Audi’s position on selfdeception and mine. Although we both depart from a model of self-deception that is straightforwardly based on stereotypical interpersonal deception, we differ in how we do that. An important difference between us might be partly explained by a difference in how we understand the kind of deceiving that is most relevant to self-deception. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Robert Audi and I have respectfully disagreed about self-deception for a long time (see Mele, 1982). In fact, Audi’s work on self-deception is what first attracted my attention to the topic. In the present article, I return to my roots in a way. My point of departure is a relatively recent exchange of ideas between Audi and me on self-deception (Audi, 2007; Mele, 2007, pp. 252– 254). In Mele, 2007, I concentrate on one of the novel ideas in Audi’s work on self-deception, his position on the relationship between self-deception and delusion. Here I focus on a more basic disagreement between us – a disagreement about what self-deceived people do and do not believe. In my view, at least in typical cases of self-deception, people believe a false proposition regarding which they are self-deceived (Mele, 2001). In Audi’s view, although the self-deceived person sincerely avows a false proposition p, he does not actually believe that p (see below). Some philosophers defend positions on this particular point of disagreement that are more similar to Audi’s than to mine (Bach, 1981; Funkhouser, 2005; Gendler, 2007; Rey, 1988); and the converse is true of many other philosophers. However, my concern here is my disagreement with Audi in particular.
2. Orienting models and a basic disagreement Audi writes: Given our wide agreement on data, why do we differ on self-deception? I credit the influence of a kind of difference more common in philosophy than many people realize: a difference in orienting models. Mele’s model is apparently the act of deceiving, in which the deceived forms a false belief and so does not see the truth in question; mine is the state of being deceived – altered to apply, so far as possible, to a person who is both deceived and deceiver and hence, apart from inconsistent beliefs, must see the truth (2007, p. 253). There is something to this. But, in my opinion, our most basic disagreement about self-deception is not explained by the difference in orienting models Audi describes here. A distinctive feature of Audi’s view is that no one who is self-deceived with q
This article is part of a special issue of this journal on Self, Other and Memory. E-mail address:
[email protected]
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.009
746
A. Mele / Consciousness and Cognition 19 (2010) 745–750
respect to a false proposition, p, actually believes that p (1982, p. 147); instead, the self-deceived person ‘‘sincerely” (p. 137) – or ‘‘non-lyingly” (p. 139) – avows p or is disposed so to avow it (also see Audi, 1985, pp. 174–175 and 1997a, p. 144). I find this highly counterintuitive (see below), and I take Audi’s thesis that no one who is self-deceived with respect to a false proposition, p, actually believes that p to be the most basic point on which we disagree. Elsewhere, I write: ‘‘Stock examples of self-deception, both in popular thought and in the literature, feature people who falsely believe – in the face of strong evidence to the contrary – that their spouses are not having affairs, or that their children are not using illicit drugs, or that they themselves are not seriously ill” (Mele, 2001, p. 9). If Audi is right, the cases at issue, as I describe them, are not actually cases of self-deception. In his view, because these people have the false beliefs I mentioned, they are deluded and not self-deceived (In Mele, 1982, I examine arguments presented in Audi, 1982 for this thesis; interested readers may consult both articles.). Consider the following four cases, all of which I offer elsewhere as examples of ways in which desiring that p can contribute to believing that p in instances of self-deception:1 CASE 1. Don just received a rejection notice on a journal submission. He hopes that his article was wrongly rejected, and he reads through the comments offered. Don decides that the referees misunderstood a certain crucial, complex point and that their objections consequently do not justify the rejection. He believes that the paper ought to have been accepted. In fact, however, the referees’ criticisms are warranted; and a few days later, when Don rereads his paper and the comments in a more impartial frame of mind, it is clear to him that this is so. CASE 2. Sid is very fond of Roz, a college classmate with whom he often studies. Wanting it to be true that Roz loves him, he interprets her refusing to date him and her reminding him that she has a steady boyfriend as an effort on her part to ‘‘play hard to get” in order to encourage Sid to continue to pursue her and prove that his love for her approximates hers for him. As Sid interprets Roz’s behavior, not only does it fail to count against the hypothesis that she loves him, it is evidence for the truth of that hypothesis. This contributes to his believing, falsely, that Roz loves him. CASE 3. Beth’s father died a short time ago, not long after her twelfth birthday. Owing partly to her desire that she was her father’s favorite, she finds it comforting to attend to memories and photographs that place her in the spotlight of her father’s affection and unpleasant to attend to memories and photographs that place a sibling in that spotlight. Accordingly, she focuses her attention on the former and is inattentive to the latter. This contributes to Beth’s coming to believe – falsely – that she was her father’s favorite child. In fact, Beth’s father much preferred the company of her brothers, a fact that the family photo albums amply substantiate. CASE 4. Betty, a political campaign staffer who thinks the world of her candidate, has heard rumors from the opposition that he is sexist, but she hopes he is not. That hope motivates her to scour his voting record for evidence of political correctness on gender issues and to consult people in her own campaign office about his personal behavior. Betty misses rather obvious and weighty evidence that her boss is sexist – which he in fact is – even though she succeeds in finding less obvious and less weighty evidence for her favored view. As a result, she comes to believe that her boss is not sexist. I do not offer these cases as the best or clearest examples of self-deception featuring a false belief that p. Their purpose is to illustrate how desiring that p can contribute to believing that p in cases of self-deception. But, as I see it, each of these four cases is a case of self-deception. And on Audi’s view, none of these cases is a case of self-deception: more precisely, Don is not is not self-deceived about whether his paper was wrongly rejected, Sid is not self-deceived about whether Roz loves him, and so on. Both Audi and I reject the idea that self-deceived people believe that p is true while also believing that p is false. And each of us might try to explain why the other is confused about what the self-deceived person does and does not believe. Each of us might think that the other has certain theoretical commitments that lead him astray or a flawed orienting model. Other philosophers have weighed in on our views of self-deception, and we have replied. But I know of no published work on what nonspecialists think about cases such as the ones above. Audi says that he has tried ‘‘to do justice to self-deception as pre-theoretically understood” (2007, p. 252). So I decided to ‘‘poll the folk,” as experimental philosophers say. I conducted two studies. In study 1, participants were 89 students in a basic philosophy class at Florida State University. Self-deception was not on the course agenda. Participants were given the following written instructions: ‘‘We’re interested in how you understand the expression ‘self-deception.’ Please read the four cases below and circle your answers to the questions about them. As you’ll see, 1 is a strong yes and 7 is a strong no.” Participants read cases 1 through 4 above. After each case, they were presented with the following text: Is this an example of self-deception? Yes, definitely 1 2 3 4 5 6 7 No, definitely not The cases appeared in four different orders: ABDC, BCAD, CDBA, DACB, where A is the story about Don, B is the story about Sid, and so on. The results were as follows:
1
Cases 1 and 2 derive from Mele, 1983, p. 369. Cases 3 and 4 derive from Mele, 2001, pp. 26–27.
A. Mele / Consciousness and Cognition 19 (2010) 745–750
CASES
MEANS
Don Sid Beth Betty
4.38 2.35 2.52 2.88
747
These results provide evidence that ‘‘self-deception as pre-theoretically understood” (Audi, 2007, p. 252) is not restricted to cases in which people lack the pertinent false belief. Given Audi’s concern ‘‘to do justice to self-deception as pre-theoretically understood” (2007, p. 252), these results should worry him.2 Some readers may wonder – as I did – why the mean response to Don’s story differed markedly from the mean responses to the other stories. In the end (that is, ‘‘a few days later”), Don sees the truth. My hunch was that this fact accounts for at least most of the difference: many readers may have judged that Don’s story is not an example of self-deception because, in the end, he is not self-deceived. In study 2, I tested this hypothesis with the following story: CASE 1a. Don just received a rejection notice on a journal submission. He hopes that his article was wrongly rejected, and he reads through the comments offered. Don decides that the referees misunderstood a certain crucial, complex point and that their objections consequently do not justify the rejection. He believes that the paper ought to have been accepted. In fact, however, the referees’ criticisms are warranted and Don should have known that his paper deserved to be rejected. Participants were 40 students in another basic philosophy class at Florida State University. Self-deception was not on the course agenda. Participants were presented with case 1a alone, and the instructions varied accordingly from those in study 1. The mean response was 3.07. Counting answers of 1, 2, and 3 as signifying agreement and answers of 5, 6, and 7 as signifying disagreement, 27 participants agreed that case 1a is an example of self-deception and 7 disagreed. A majority of these participants, unlike Audi, evidently do not restrict cases of self-deception to cases in which people lack the pertinent false belief. If I were to be persuaded that I am wrong about the meaning of ‘‘self-deception,” that the great majority of the respondents in my studies misapply the concept of self-deception to the cases they are asked about, and that Audi is right, I doubt that I would be terribly disappointed. In my work on self-deception (or what I think of as self-deception), my primary concern is to explain how ordinary instances of it happen. I have suggested that the following conditions are jointly sufficient for entering self-deception in acquiring a belief that p. 1. 2. 3. 4.
The belief that p which S acquires is false. S treats data relevant, or at least seemingly relevant, to the truth value of p in a motivationally biased way. This biased treatment is a nondeviant cause of S’s acquiring the belief that p. The body of data possessed by S at the time provides greater warrant for p than for p. (Mele, 2001, pp. 50–51).3
Drawing on empirical work (in Mele, 2001 and elsewhere), I have articulated and defended a theory about how it happens that people satisfy this collection of conditions (and, of course, I have offered guidance on interpreting conditions 2 and 3). If Audi is right, a person’s satisfying the first of these conditions is incompatible with his being in self-deception with respect to p; again, he holds that the self-deceived person does not actually believe the false proposition with respect to which he is self-deceived. On Audi’s view, a person who satisfies my four conditions may be deluded, but he is not in self-deception with respect to p. If I were to be persuaded that he is right, I might just say that I have learned that, really, what I had been talking about all along was entering self-delusion in acquiring a belief that p; and I might leave it at that. But, of course, partly in light of the data I discussed, I remain unpersuaded that Audi is right about the meaning of ‘‘self-deception.” Indeed, I am persuaded that he is wrong about it. 3. History and more Audi says that the apparent difference between our respective orienting models is related to our disagreement about whether self-deception is an essentially historical phenomenon (2007, p. 254). I believe that he is right about this. After providing a bit more background, I will explain why. I have already reported my proposed set of sufficient conditions for entering self-deception in acquiring a belief that p. As I observe elsewhere (Mele, 1987, p. 131 and 2001, pp. 56–57), people can also enter self-deception in retaining a belief. Here is an example from Mele, 1987 (pp. 131–132). Sam has believed for many years that his wife, Sally, would never have an
2 A referee expressed the worry that study 1 is not a proper test because it includes no control (or contrast) group. However, the point of the study is to gather evidence about whether the claim that being self-deceived with respect to a false proposition, p, requires not actually believing that p – that alleged necessary condition on self-deception – does ‘‘justice to self-deception as pre-theoretically understood” (Audi, 2007, p. 252). Control and contrast groups are not needed for this purpose. Incidentally, in this article I am not concerned to argue that self-deception entails false belief. It is the claim of Audi’s on which I am focusing that concerns me. 3 Obviously, this set of alleged sufficient conditions does not commit me to the view that self-deception requires false beliefs.
748
A. Mele / Consciousness and Cognition 19 (2010) 745–750
affair. In the past, his evidence for this proposition was very good. Sally obviously adored him, she never displayed a sexual interest in another man, she condemned extramarital sexual activity, she was secure, she was happy with her family life, and so on. However, things recently began to change. Sally is now arriving home late from work on the average of two nights a week, she frequently finds excuses to leave the house alone after dinner and on weekends, and Sam has been informed by a close friend that Sally has been seen in the company of a certain Mr. Jones at a theater and a local lounge. Nevertheless, Sam continues to believe that Sally would never have an affair. But he is wrong. Her relationship with Jones is by no means platonic. On the model of my proposed set of sufficient conditions for entering self-deception in acquiring a belief, I propose that the following conditions are conceptually sufficient for entering self-deception in retaining a belief that p: 1. 2. 3. 4.
The belief that p which S retains is false. S treats data relevant, or at least seemingly relevant, to the truth value of p in a motivationally biased way. This biased treatment is a nondeviant cause of S’s retaining the belief that p. The body of data possessed by S at the time provides greater warrant for p than for p.
Whenever one enters self-deception in acquiring or retaining a belief that p, there is a transition from one’s not being selfdeceived in believing that p to one’s being self-deceived in believing that p. In my view, an essential feature of the transition is that it includes one’s deceiving oneself, a process that takes time. This is part of what Audi refers to as my ‘‘orienting model.” (Readers who believe that deceiving is always intentional will be misled by what I have just said. I believe that much deceiving is not intentional, and I argue that self-deceivers typically do not intentionally deceive themselves; see Mele, 2001, chapter 1.) Furthermore, as I see it, one is not in self-deception unless one entered self-deception. If this is right and if I am right about what entering self-deception requires, then being in self-deception has an essential historical feature: one is not in self-deception unless one deceived oneself, and deceiving oneself takes time.4 Some things have essential historical features. For example, a burn is not a sunburn unless it was caused by exposure to the sun, and a piece of paper is not a genuine US dollar bill if it was not produced by the US Treasury Department. I see a person’s being in a state of self-deception as similar to a patch of skin’s being is a sunburned state. Audi asks (2007, p. 254): ‘‘Must we deny that if I am perfectly duplicated, then, if I am now in self-deception, a full, purely psychological description (apart from listing self-deception) would ground (or even entail) my being in self-deception?” Well, must we deny that if I am perfectly duplicated, then, if I now have a sunburn on my back, my duplicate has a sunburn on his back too, even though his burn was produced by a heat lamp? My answer to the second question is yes, we must deny it. My answer to the first question is the same (assuming that the purely psychological description includes nothing historical), but I am open to being talked out of it. As Audi observes, he has ‘‘argued that ‘self-deception with respect to p is a state in which S [as deceiver] unconsciously knows (or has some reason to believe, and unconsciously and truly believes) that [p], [yet, as deceived] sincerely avows or is disposed so to avow, that p, and has at least one want which in part explains why the belief that p is unconscious’” (2007, p. 252; the brackets are Audi’s and the embedded quotation is from Audi, 1982). This is a complex and distinctive state, and I can see why someone who thinks of self-deception along these lines would think that nothing more – including any fact about how the person came to be in this state – is required for being in self-deception with respect to p (at least for many p’s) than being in this state. In my view, the purely here-and-now state that a self-deceived person is required to be in is simpler. I do not require the unconscious true belief, and so, of course, I also do not require the presence of a want that partly explains why the belief is unconscious. Perhaps it is partly because my view is simpler on this person-internal front that I am inclined to see something historical as essential to being ‘‘in self-deception.” The expression ‘‘in self-deception” is not one for which I have much use. As Audi mentions (2007, p. 252), my focus has been on entering self-deception. However, I am happy to say that someone who has entered self-deception in acquiring or retaining a belief that p is in self-deception with respect to p, as long as the person continues to be self-deceived in believing that p. One thing that satisfaction of this last clause requires, of course, is that the person continues to believe that p. But that is not sufficient for satisfying the clause. The person’s evidence may have come to be such that he is now warranted in believing that p. In some such case, it may be true that although the person was self-deceived in believing that p and still believes that p (a false proposition), he is no longer self-deceived in believing it (and is no longer in self-deception with respect to p). Audi writes: ‘‘Self-caused deception need not be self-deception; and even an analysis of self-caused deception may not suffice for an adequate account of self-deception” (2007, p. 253). I agree entirely. As I see it, causing oneself to be deceived is a necessary condition for self-deception, not a sufficient condition. And, as I point out in Mele, 1987, ‘‘if Jones, due to unmotivated carelessness, misreads a clearly printed word in a news article, thereby causing himself to have a false belief (and, hence, to be deceived), neither the deceiving of himself, nor the condition of deception which he has produced in himself, is the sort of thing which we ordinarily call self-deception” (pp. 124–125). This is an example of self-caused deception without self-deception. Audi contends that his model ‘‘does justice to the dissociation self-deception apparently entails and places the element of deception in the tendency to make sincere avowals of propositions one unconsciously knows to be false yet – in misleading
4
As I understand deceiving oneself, it entails causing oneself to have a false belief. (Whether most lay folk understand deceiving oneself as I do is testable.)
A. Mele / Consciousness and Cognition 19 (2010) 745–750
749
but natural terms – ‘consciously believes’ to be true. By virtue of the dissociation, the sincere avowal, which would normally imply belief, manifests a kind of deception” (2007, p. 253). In a related vein, some philosophers, including Audi (1997b), have argued that my proposed sufficient conditions for entering self-deception in acquiring a belief that p are not sufficient because they do not capture a kind of tension that is necessary for self-deception (see Mele, 2001, pp. 52–56 for references and discussion). The quartet of conditions at issue certainly does not entail that there is no tension in self-deception. Nor do I claim that self-deception normally is tension-free. Satisfying my four conditions may often involve considerable psychic tension. The present question is whether tension or dissociation is conceptually necessary for entering self-deception in acquiring a belief that p. My answer is no. Given the details of Sid’s case, for example, even if he is free of psychic conflict during the process of acquiring the belief that Roz loves him and while that belief is in place, he is self-deceived and he enters selfdeception in acquiring that belief. In Audi’s view, again, ‘‘self-deception with respect to p is a state in which”: (1) ‘‘S [as deceiver] unconsciously knows (or has some reason to believe, and unconsciously and truly believes) that [p],” (2) ‘‘[yet, as deceived] sincerely avows or is disposed so to avow, that p,” (3) ‘‘has at least one want which in part explains why the belief that p is unconscious” (2007, p. 252), and (4) does not believe that p. I believe that self-deception is pretty common (see Mele, 2001). How common are cases that satisfy Audi’s four conditions? Suppose we were scientists trying to answer this question. How would we proceed? One thing we would want to be able to do is to figure out a way to tell whether someone who sincerely avows that p also believes that p. Because this would be very difficult, it would be very difficult to be confident, regarding any apparently ordinary instance of self-deception, that it satisfies both condition 2 and condition 4. When I add to that the obstacles to confidence about what alleged self-deceivers unconsciously know or believe, I find that I am reluctant even to guess about the frequency of cases satisfying Audi’s four conditions. One might propose that we arrive at the postulation of the satisfaction of Audi’s four conditions in cases of apparent selfdeception by an inference to the best explanation. We look at what the person does and says and judge that the best explanation of his deeds and words is that he satisfies Audi’s four conditions. Here is an alternative explanatory hypothesis about tension-involving cases of the kind on which Audi focuses: what the person avows – namely, p – he believes; and rather than unconsciously knowing or believing that p, he consciously believes that there is a significant chance that p. Readers are encouraged to make up their own minds about which hypothesis is more plausible. For a defense of the hypothesis I just reported, see Mele, 2001, pp. 67–73. Someone might claim that, owing to the nature of deceiving, self-deception requires that one know (or at least believe) the truth. Recall, in this connection, Audi’s report that his state model of self-deception is meant to apply ‘‘to a person who is both deceived and deceiver and hence, apart from inconsistent beliefs, must see the truth” (2007, p. 253). Obviously, it is not the fact that the person is deceived that is meant to motivate the claim that he ‘‘must see the truth”; the fact that he is a ‘‘deceiver” is meant to do this. Consider the following two propositions: 1. By definition, person A deceives person B (where B may or may not be the same person as A) into believing that p only if A knows, or at least believes truly, that p and causes B to believe that p. 2. By definition, deceiving is an intentional activity: nonintentional deceiving is conceptually impossible. In various places, including chapter 1 of Mele, 2001, I argue that both propositions are false and that acceptance of either of them easily leads to confusion about self-deception. I cannot repeat the arguments here, but I will say something about this issue (drawing on Mele, 2001, pp. 8–9). In a standard use of ‘deceived’ in the passive voice, we properly say such things as ‘‘Unless I am deceived, I left your article on my desk.” Here ‘deceived’ means ‘mistaken’. There is a corresponding use of ‘deceive’ in the active voice. In this use, to deceive is ‘‘to cause to believe what is false” (as the Oxford English Dictionary reports). Plainly, one can intentionally or unintentionally cause someone to believe what is false; and one can cause someone to acquire the false belief that p even though one does not oneself believe that p. Yesterday, mistakenly believing that my daughter’s car keys were on my dresser, I told her they were there. In doing so, I caused her to believe a falsehood. I deceived her, in the sense identified; but I did not do so intentionally, nor did I cause her to believe something I disbelieved. Stereotypical instances of deceiving someone else into believing that p are instances of intentional deceiving in which the deceiver knows or believes truly that p. The point made in the preceding paragraph has little significance for self-deception, if paradigmatic instances of it have the structure of stereotypical instances of interpersonal deception. However, like Audi, I argue that they do not. Of course, we differ in how we depart from a model of self-deception that is straightforwardly based on stereotypical interpersonal deception. Audi observes that he has a state model of self-deception whereas I have an act model (2007, p. 253); but his model seems to be influenced by an assumption about acts of deceiving that I reject. At any rate, one important difference between us is that he requires the self-deceived person to know or believe the pertinent true proposition, and I do not. This difference might be partly explained by a difference in how we understand the kind of deceiving that is most relevant to self-deception. I close with a comment on the survey studies I reported. Although some philosophers find experimental philosophy irritating, I believe that – whether it is irritating or not – its techniques can be useful in some philosophical debates. Here my studies were put to a very limited use. As I reported, Audi says that he has tried ‘‘to do justice to self-deception as pre-theoretically understood” (2007, p. 252). Now, I know too much about motivationally biased belief (see Mele, 2001) to be
750
A. Mele / Consciousness and Cognition 19 (2010) 745–750
confident about my own untested hunches regarding how self-deception is pre-theoretically understood. And just as my hunches about this may be biased partly by my theory, the same is true of Audi. What I tested was the hypothesis that self-deception, as pre-theoretically understood, is such that no one who is self-deceived regarding a false proposition p actually believes that p. And the results definitely do not favor the hypothesis. Although, as I have noted, I am inclined to believe that every case of self-deception involves deceiving and that the kind of deceiving at issue entails causing a false belief, this inclination of mine was not examined in this article and I have not tested the hypothesis that my inclination is in line with majority judgments of lay folk about cases. Others are free to conduct such tests. The results may prove interesting. References Audi, R. (1982). Self-deception, action, and will. Erkenntnis, 18, 133–158. Audi, R. (1985). Self deception and rationality. In M. Martin (Ed.), Self deception and self understanding (pp. 169–194). Lawrence: University of Kansas Press. Audi, R. (1997a). Self-deception, rationalization, and the ethics of belief. In R. Audi (Ed.), Moral knowledge and ethical character (pp. 131–156). New York: Oxford University Press. Audi, R. (1997b). Self-deception vs. self-caused deception: A comment on Professor Mele. Behavioral and Brain Sciences, 20, 104. Audi, R. (2007). Belief, intention, and reasons for action. In M. Timmons, J. Greco, & A. Mele (Eds.), Rationality and the good (pp. 248–259). New York: Oxford University Press. Bach, K. (1981). An analysis of self-deception. Philosophy and Phenomenological Research, 41, 351–370. Funkhouser, E. (2005). Do the self-deceived get what they want? Pacific Philosophical Quarterly, 86, 295–312. Gendler, T. (2007). Self-deception as pretense. Philosophical Perspectives, 21, 231–258. Mele, A. (1982). Self-deception, action, and will: Comments. Erkenntnis, 18, 159–164. Mele, A. (1983). Self-deception. Philosophical Quarterly, 33, 365–377. Mele, A. (1987). Irrationality: An essay on akrasia, self-deception, and self-control. New York: Oxford University Press. Mele, A. (2001). Self-deception unmasked. Princeton: Princeton University Press. Mele, A. (2007). Self-deception and three psychiatric delusions: On Robert Audi’s transition from self-deception to delusion. In M. Timmons, J. Greco, & A. Mele (Eds.), Rationality and the good (pp. 163–175). New York: Oxford University Press. Rey, G. (1988). Toward a computational account of Akrasia and self-deception. In A. Rorty & B. McLaughlin (Eds.), Perspectives on self-deception (pp. 264–296). Berkeley: University of California Press.
Consciousness and Cognition 19 (2010) 751–761
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Descartes discarded? Introspective self-awareness and the problems of transparency and compositionality q Markus Werning Ruhr University Bochum, Department of Philosophy, Universitätsstr. 150, 44801 Bochum, Germany
a r t i c l e
i n f o
Article history: Available online 16 August 2010 Keywords: Introspection Self-awareness Inner self Semantic compositionality Phenomenal transparency Mental quotation Extrospection Inner speech Phonology Descartes
a b s t r a c t What has the self to be like such that introspective awareness of it is possible? The paper asks if Descartes’s idea of an inner self can be upheld and discusses this issue by invoking two principles: the phenomenal transparency of experience and the semantic compositionality of conceptual content. It is assumed that self-awareness is a second-order state either in the domain of experience or in the domain of thought. In the former case self-awareness turns out empty if experience is transparent. In the latter, it can best be conceived of as a form of mental quotation. Various proposed analyses of direct and indirect quotation are discussed and tested regarding their applicability to thought. It is concluded that, on the assumption of compositionality, the inner self is only insofar accessible to awareness as it has an accessible phonological (or otherwise subsymbolic) structure, as apparently only inner speech does. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction ‘‘I am aware that I exist; I ask who is that I of which I am aware”.1 When René Descartes poses this question right after his seminal cogito argument of the Second Meditation, he raises the issue of self-awareness. He does not question the existence of such a self. The existence of the self, rather, is asserted in advance. His question is also not one about the epistemic status of the process of self-awareness: Is it knowledge that we have or not? The Latin original indicates that Descartes, indeed, takes it to be knowledge.2 No, what Descartes asks is: What has the self to be like such that reflexive awareness of it is possible? It is thus presupposed that the essential property of the self is an epistemic property, viz. being the object of reflexive awareness. Descartes’s own answer is twofold. The negative part consists in the claim that the object of self-awareness in the intended sense is not the body. The positive part is rather indirect: ‘‘that which is doubting, understanding, affirming, denying, q
This article is part of a special issue of this journal on Self, Other and Memory. E-mail address:
[email protected] 1 Latin original: ‘‘Novi me existere; quaero, quis sim ego ille, quem novi” (Descartes, 1641/1904, p. 27) – A frequently noted worry roots in a grammatical peculiarity that concerns the unusual application of a demonstrative – ille ‘that’ – to a personal pronoun – ego ‘I’. Some philosophers have blamed Descartes for using ungrammatical language and even accused him of ‘‘a linguistic derailment with terrible consequences” (Beckermann, 2009, p.7). Demonstratives are akin to definite articles in that they are typically applied to generic terms as in that/the man. The Latin ego as the English I is an indexical term – whose referent varies with the speaker and the situational context – and normally neither allows for a demonstrative nor for a definite or indefinite article. The application of a demonstrative to the first person singular pronoun coerces a grammatical type shift and treats the pronoun as a generic term. It so, in a question-begging way, hypostatizes a class of entities: the Is – and consequently my I, your I, her I, etc. To excuse Descartes, one might point out that natural language is very flexible and provides many examples for forced shifts of grammatical type. There, e.g., do exist accepted phrases like the here and now where indexicals occur with a definite article. Grammar alone does not suffice to discard Descartes. 2 Novi is the first person present perfect of noscere which means ‘get to know’. 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.07.003
752
M. Werning / Consciousness and Cognition 19 (2010) 751–761
willing, refusing, imagining, and experiencing.”3 Descartes, so we may justly read, identifies the object of self-awareness with the subject of mental states and he thereby individuates the self as the unity that carries mental contents. This would make selfawareness a case of introspective self-awareness or awareness of an inner self. The idea of an inner self contrasts with the idea of a bodily self. The latter can be regarded as the object of perceptual or doxastic attitudes turned towards one’s own body. These make up the so-called body image. The bodily self can also be regarded as given by a body schema: a collection of motor programs or habits that enable and constrain movement and the maintenance of posture (Gallagher, 2005). It is yet an open question whether and to which extent body image and body schema are intertwined. Experiments known as the rubber hand illusion and the out-of-body illusion induce an illusionary body schema – either a mislocation or a partial or global misidentification of one’s own body – on the basis of visual and tactile stimulation alone (Botvinick & Cohen, 1998; Lenggenhager, Tadi, Metzinger, & Blanke, 2007). This indicates that body image and body schema may in fact be more closely related to each other than the conceptual differentiation suggests. An essentially diachronic view to be distinguished from the notions of the inner and the bodily self is the proposal of a narrative self. It has been suggested to conceive of it is as an abstraction from an autobiographical narrative process that brings together hitherto only minimally coherent pieces of behavior ‘‘enhanced by an illusion of greater unity” (Dennett, 1992, see also Ricoeur, 1992). The present paper takes up Descartes’s original question regarding the nature of the object of introspective self-awareness and asks whether Descartes’s idea of an inner self as a unitary carrier of mental contents can be upheld or whether it has to be discarded. In the latter case the self would very likely reduce to nothing but the bodily self or would have to be replaced by the quasi fictional idea of a narrative self. I would like to stress that metaphysical questions regarding the legacy of a res cogitans as a substance ontologically independent from the physical world are important in their own right, but will not be of concern to us here. The idea that a carrier of mental contents be the object of self-awareness does per se not contradict physicalism, as the various contemporary naturalistic accounts of mental content seem to show (see Greenberg, 2005, for review). Instead, we will discuss Descartes’s legacy by invoking two principles that have been moving ever closer into the center of the debate in the philosophy of mind over the past two decades: the principle of the phenomenal transparency of experience and the principle of the semantic compositionality of conceptual content. Experience is said to be phenomenally transparent just in case having an experience of certain objects, events or situations is for the subject just as if the objects, events, or situations were present (see the following section for elaboration). By saying that a conceptual representation is semantically compositional we mean that the content of a complex concept is a structure-dependent function solely of the contents of the parts of the concept (see below). The two principles cover two aspects – the conceptual and the phenomenal – that are somewhat heterogeneously distributed over the class of mental states. There are mental states like beliefs, desires, recollections, or expectations that are often regarded as conceptual – these will be called thoughts in the course of the paper. There are mental states like perceptions, hallucinations, and proprioceptions that are widely regarded as phenomenal – they will subsequently be discussed under the label of experience. There may also be mental states that have both a conceptual and a phenomenal aspect: many emotions are eventual candidates here. One may even hold that seemingly phenomenal states after some scrutiny reduce to conceptual states. Even though the transparency of experience and the compositionality of conceptual content regard two different aspects in the class of mental states, they, from a more abstract point of view, are somewhat akin because they both deny any particular role that access to the intrinsic properties of the vehicles of mental content might have in the determination of content. As we will see later on, this is the main reason why the idea of introspective self-awareness is so difficult to accommodate for if the two principles hold. Given the main mutually non-exclusive, but probably exhaustive dichotomy in the class of mental states – experiences with phenomenal qualities on the one hand and thoughts with a conceptual structure on the other hand – there are altogether three ways to account for introspective self-awareness: (i) It can be construed as a phenomenon completely in the domain of experience, such that the introspective and the introspected states are experiences. For that case will we show that introspective self-awareness is empty if experience is phenomenally transparent. (ii) Introspective self-awareness can be analyzed fully in the domain of thought. In that case we will argue that introspective self-awareness is best conceived of as a mental analogue of direct quotation. Provided that thought is compositional, we will then show that the inner self must have a cognitively accessible subsymbolic structure which is very likely akin to the phonological structure of natural language. Thus mental states would have to be linguistically structured thoughts in order to be introspected. The inner self would reduce to a stream of inner speech. (iii) Introspective self-awareness is a hybrid phenomenon crossing the domains of thought and experience: it would thus either be a non-conceptual experience directed towards a non-phenomenal thought or a non-phenomenal thought representing a non-conceptual experience. Since this is a rather hypothetical option and to my knowledge nobody has ever tried to analyze introspective self-awareness this way, I will spare a discussion of this option due to space limits. The paper concludes with the conditional claim that if experience is phenomenally transparent and thought is semantically compositional, then introspective self-awareness is either empty or reduces to a representation of inner speech.
3
Latin original: ‘‘Nempe dubitans, intelligens, affirmans, negans, volens, nolens, imaginans quoque, et sentiens‘‘ (Descartes, 1641/1904, p. 28).
M. Werning / Consciousness and Cognition 19 (2010) 751–761
753
Although I will focus more on an explication than on a detailed justification of the conjunctive antecedent of the conditional in this paper, I expect that only few philosophers will be ready to reject the if-clause of the conditional tout court. Whereas compositionality is almost unanimously accepted as a constraint on thought, the situation with regard to the phenomenal transparency of experience, I concede, is more complex. Here the main question is whether perception and eventually other experiential states are regarded as having conceptual or non-conceptual content. If experiences were construed as conceptual (see, e.g., McDowell, 1996), they would count as instances of thought in the sense of this paper. The compositionality conjunct of the antecedent would apply and the truth of the conditional would make its consequent unavoidable even if one were to deny the phenomenal transparency of experience. Only for non-conceptualist views of experience the transparency conjunct is essential to establish the consequent. Note, however, that most non-conceptualists, prominently Dretske (1981/2003b,2003a), do endorse the transparency thesis or related claims (e.g., direct realism). So even if the main claim of this paper is limited to a conditional, the foregoing remarks illustrate that logical space leaves remarkably little room to avoid the antecedent of the conditional and hence its consequent. 2. Phenomenal transparency This issue of phenomenal transparency was first advanced by G. E. Moore. In his Refutation of Idealism he writes: [...] that which makes the sensation of blue a mental fact seems to escape us; it seems, if I may use a metaphor, to be transparent – we look through it and see nothing but the blue; we may be convinced that there is something, but what it is no philosopher, I think, has yet clearly recognized. (Moore 1903, p. 446) In a slightly more poetic way Martin Heidegger formulated the same idea independently and made it a cornerstone of his philosophy: We never originally [...] perceive a throng of sensations, e.g., tones and noises, in the appearance of things [...]; rather we hear the storm whistling in the chimney, we hear the three-engine aeroplane, we hear the Mercedes in immediate distinction from the Volkswagen. Much closer to us than any sensations are the things in themselves. (Heidegger, 1935/1977, p. 156) Gilbert Harman re-introduced the notion of phenomenal transparency into modern analytic philosophy: When you see a tree, you do not experience any features as intrinsic features of your experience. Look at a tree and try to turn your attention to intrinsic features of your visual experience. I predict you will find that the only features there to turn your attention to will be features of the presented tree, [...] (Harman, 1990, p. 39). What the three authors point to is the fact that, when we experience something, we do not experience the intrinsic properties of the vehicles of the experience – what Moore and Heidegger call ‘‘sensations”.4 To give a positive account of phenomenal transparency – what it is for the subject to have such an experience rather than what it is not – and to avoid metaphorical language, I propose a definition that uses a counterfactual as-if construction: Definition phenomenal transparency A subject’s S experience of a scenario X is called phenomenally transparent just in case S’s having an experience of X is for S just as if X were present. As the scenario somebody experiences we understand the totality of the objects and events with their properties and relations that make up the content of the experience at a time. The as-if analysis of transparency exhibits a number of features that allude to aspects prominent in traditional philosophical characterizations of experience and are worth closer investigation. First, the definition presupposes a relation of ownership between the experience of the scenario and the experiencing subject. Moreover, it also covers the aspect of subjectivity in that it claims that, for the owner of the experience, having the experience is as if the scenario were present. For a third person, who studies the experience, say, neurophysiologically, the experience may well be different. Second, our definition contrasts with the position of direct realism which would license the inference from S experiences X to X is present. This inference is blocked because the as-if clause is counterfactual. So even if the scenario is not present, having the experience might well be for the subject as if the scenario were present. ‘‘Experience” in our context is not a success word and the definition is not restricted to veridical experience but may as well hold for hallucination. However, what is sometimes called the ‘‘feeling of realness” (Brown, 2004) is intended to be entailed. This feeling of realness is typical for perceptions and proprioceptions, but also comes with many hallucinations and illusions. The notion of presence is even stronger 4 Metzinger formulates what he calls the ‘‘standard definition” of phenomenal transparency as follows: ‘‘Phenomenal states are transparent in that only their content-properties are introspectively accessible to the subject of experience” (Metzinger, 2003, p. 354). In my eyes it is misleading to presuppose that the content-properties are accessible introspectively. For, it is not introspection, but rather perception or – as in the case of bodily states of one’s own – proprioception that makes the content-properties available to the subject. However, since the idea of phenomenal transparency denies that there are accessible intrinsic properties of the vehicles of experiences, the whole idea of introspection becomes at least questionable and we should not define phenomenal transparency in terms of introspective accessibility in the first place.
754
M. Werning / Consciousness and Cognition 19 (2010) 751–761
than that of realness since it does not only entail that the scenario is real for the subject, but also that there is some spatiotemporal immediacy of the scenario for the subject. Finally, the definition contains an important minimality constraint. The use of the word just implies that in the counterfactual condition nothing is present for the subject, but the objects and events with their properties and relations of the scenario that make up the content of the experience. As we will see, the minimality constraint is crucial for the problem phenomenal transparency creates for purely experiential accounts of introspective self-awareness. 3. The experiential account of introspective self-awareness One way to think of introspection is to view it as a phenomenon fully in the domain of experience. Introspection might be regarded as an experience of the experiences one has at a certain time. The idea is that a person might, e.g., see a tree and then in addition experience that she sees a tree. She might have a toothache or feel a sore tooth and then experience that she has a toothache or feels a sore tooth. Introspection according to this view would be a second-order experience. However, would introspection construed as a second-order experience lead to self-awareness? Would it license the inference to a self? The answer very likely is ‘‘no” if experience is phenomenally transparent. To see this, let’s first turn to proprioceptions like pain. There are two ways of thinking about pain (and most other proprioceptions). According to the first conception, pain does not necessarily involve consciousness, but simply is a state of the body that one occasionally experiences when one feels pain. Under this view feeling pain is a first-order experience and does not at all qualify as an introspective state. According to the alternative conception, pain is conceived of as a state that necessarily involves consciousness. Here, pain is itself a state of experience rather than a content of experience. Having a toothache, according to this view, is the experience of a sore tooth (of one’s own). Turning to the second-order experience we now face a dilemma: Experiencing the presence of a sore tooth and experiencing a toothache – i.e., having the second-order experience – are for the subject either the same or not the same. If they are the same for the subject, there is no reason for the subject to infer the existence of anything over and above the tooth and its soreness on the basis of her second-order experience just as experiencing the presence of a sore tooth would not license any such inferences. Any inference to an internal vehicle of the proprioception as part of an inner self would be unmotivated by the second-order experience. If experiencing the presence of a sore tooth and experiencing a toothache, on the other hand, are not the same for the subject, then phenomenal transparency is violated. For, if the experience of a toothache were phenomenally transparent it should, by definition, be for the subject just as if a toothache were present. Now again, the presence of a toothache for the subject under the second conception of pain should be just the same as experiencing the presence of a sore tooth. Unfortunately, the states of proprioception do not by introspection contribute to an awareness of the self if phenomenal transparency holds. The same is true for perception: Assume that the perception of a tree be itself an object of introspective experience. The phenomenal transparency of experience would entail that being in the perceptive state and the introspective state at the same time would be for the subject just as if a tree and the perception of a tree were present. After another application of the transparency property we see that the latter – the perception of a tree – is for a subject just as if a tree were present. It follows that being in the combined – perceptive plus introspective – state is for the subject just as if a tree and the same tree were present. It hence is just like being in the state of perception without being in the state of introspection. Introspection adds nothing to perception and gives no cues to an inner self if experience in general is phenomenally transparent.5 4. Second-order thoughts Introspective self-awareness might also be analyzed in terms of conceptually structured thoughts rather than as a form of phenomenal experience. The property of phenomenal transparency would not apply. In those terms an introspective state would be a second-order thought of sorts that reports another thought of one’s own. What we are looking for seems closely analogous to – yes, indeed, structurally the same as – quotation in natural language. A quotation in natural language is an utterance that reports another utterance. One generally distinguishes two kinds of quotation: direct quotation and indirect quotation. The reason is that an utterance can either be reported by reference to its vehicle – the actual expression uttered – or by reference to its content or meaning – what was said by the utterance. Compare the following:
Donald said; ‘In Mark’s garden furze is more widespread than iv y’:
ð1Þ
Donald said that in Mark’s garden furze is more widespread than iv y:
ð2Þ
The truth-conditions of the two sentences differ. The words that were actually used in the reported utterance make a difference regarding direct quotation whereas they do not have an effect on the truth of the indirect quotation if content is left unchanged. This holds for all kinds of verbal variations that would leave content unchanged: (a) Choice of language: Had Donald said what he said in German, (1) would no longer be an accurate report of the utterance, while (2) would remain true. (b) Substitution of synonyms: For the truth of (2) it does not matter whether Donald used the word gorse or furze whereas it does matter for (1). (c) Substitution of co-referential rigid designators: Assume that Mark and Samuel be proper
5
The argument of this section largely builds on Dretske (2003). For a closer discussion see Werning (2004a).
M. Werning / Consciousness and Cognition 19 (2010) 751–761
755
names that refer to the same person – as would be the case for Mark Twain and Samuel Clemens. Then the truth of (2) would remain unaffected even if Donald actually used the word Samuel to refer to the garden in question.6 (d) Analytical or logical equivalence: The utterance In Mark’s garden, ivy is less widespread than furze would be truly reported by (2), but not by (1). Furthermore, in specific contexts – where e.g., conversational focus, metrics, dialect, or speech impediments are at issue – the accuracy standards for direct quotation may be raised so that even word order, prosody, phonology, or phonetics play a role. All those aspects are irrelevant for indirect quotation because here only the content, not the vehicle of the utterance is at issue. There are numerous other aspects in which direct and indirect quotation differ – the role of indexicals, anaphoric reference, tense, etc. – but these issues will not be dealt with here. When we now turn back to self-reporting second-order thoughts, we have a choice between conceiving of them as reports of the internal vehicles of other thoughts of the subject – i.e., as direct mental quotations – or as reports only of the contents of those thoughts – as indirect mental quotations. What would those choices imply for the possibility of self-awareness? For second-order thoughts of the latter kind Fernández (2003, 2005) has coined the term extrospection. He argues that any piece of evidence – perception, memory, testimony – that justifies a subject’s belief that p also justifies the subject’s belief that she believes that p if the subject has formed the belief that p and the belief that she believes that p on the basis of this evidence. His argument presupposes a reliabilist notion of justification and identifies the justifiedness of a belief with being the product of a belief formation process that is grounded in some reliable regularity. The argument can roughly be summarized as follows: If a properly functioning epistemic subject has good evidence for the truth of the proposition p, she normally comes to believe that p. Due to this mind-to-mind regularity, a subject that forms the belief that p on the basis of good evidence for the proposition p is also justified, on the basis of this evidence, to believe that she believes that p. Access to the internal vehicle of the first-order thought is thus not required to justify the second-order thought. One only has to have access – this is the role of good evidence – to the content of the first-order thought, the proposition p, to justify the secondorder thought. Thus self-knowledge would be sufficiently grounded in extrospection. I am in general very sympathetic with this account of self-knowledge. The question in our context however is whether extrospective self-knowledge qualifies as knowledge of an inner self. Only thus Descartes’s original idea could be defended. I take it that this was not even in the intentions of Fernández’s. Just as indirect linguistic quotation gives us only few hints as to which linguistic expressions – which language, which words, and which syntactic structures – were actually used by the reported speaker, extrospection, viewed as indirect mental quotation, provides us with almost no clue with respect to the nature, the constituents, and the structure of the reported thought. All it provides us with is the external content of the latter. The inner self remains in the dark. Self-reporting second-order thoughts might also be construed in analogy to direct quotation. Here we may justly speak of introspective thoughts because thoughts of one’s own are reported by reference to their internal vehicles. The challenge here is to find a proper way to analyze direct quotation and apply this analysis to thought. In this regard it is of particular significance that thought – as is widely assumed – is compositional. The principle of semantic compositionality is defined in most general terms for representational structures like language or thought that consist of (i) a set of representational vehicles and (ii) a set of combinational operations mapping arrays of constituent vehicles onto a complex vehicle, and for which (iii) a function of semantic evaluation is defined on the set of vehicles: Definition semantic compositionality A representational structure (i.e., language or thought) is called semantically compositional just in case the semantic value (meaning, or respectively, content) of any complex vehicle (grammatical term or, respectively, mental concept) is a function of the semantic values of its constituent vehicles where the choice of the function depends solely on the operation by which the constituents combine to form the complex.7 As Fodor (2001) has pointed out, compositionality is much less controversial for concepts (including truth-evaluable thoughts) with respect to their contents than it is for linguistic expressions with regard to their meanings. Even though many a priori justifications for the compositionality of meaning in natural language have been put forward more or less convincingly – learnability, productivity, systematicity, communicative efficiency (for a critical discussion of those arguments see Werning, 2005; Pagin & Westerståhl, in press) – it is not keen to say that a final verdict has not yet been spoken. In particular many tenacious linguistic counterexamples have yet to be defused. In the case of thought the case seems clearer: What could matter for the contents of the constituent concepts to determine the content of the complex concept other than their combinational structure? In natural language extralinguistic factors that lie neither in syntactic structure nor in the meanings of the constituent expressions may eventually matter to determine the meaning of a complex expression. Think of ostensive gestures that help to determine the reference of demonstratives (see Kaplan, 1989). Or imagine background situations that play a role for the determination of truth-conditions. In sentences like All guests have gone uttered at a party or Some passengers have died said after a train accident, the situation obviously restricts the domain of the quantifier – if quantifiers ranged over an unrestricted domain, sentences of this kind would either be trivially false or trivially true. In those cases one is typically required to allow for explicit or implicit
6 It is widely, though not unanimously accepted that co-referential rigid designators may be replaced in indirect quotations without changing their truthconditions. This view grounds in the possible-worlds semantics of propositional attitudes. 7 For formalizations see Hodges (2001) and Werning (2004b).
756
M. Werning / Consciousness and Cognition 19 (2010) 751–761
contextual arguments in order to force the semantic analysis of a sentence into a compositional framework. This strategy only works because one may assume that both the speaker and the hearer of such a sentence do represent the relevant factors mentally – typically on the basis of perception or memory – even if the factors are not contained in the learned meanings of the constituent expressions themselves. In the domain of thought this strategy is not viable because contextual factors may matter to determine the content of a complex concept only if they are mentally represented. They are thus not extra-representational, but belong to a larger conceptual structure internal to the mind of the subject. To thought, compositionality applies without exception. If we are hence to construe introspective thoughts in analogy to direct quotations, we have to make sure that the analysis of direct quotation used abides by the principle of compositionality. This is an important point because many accounts of direct quotation in natural language take it virtually as an axiom that quotation be non-compositional. Cappelen and Lepore (2009), e.g., identify the first ‘‘basic quotational [feature] of particular importance” with the principle that ‘‘in quotation you cannot substitute co-referential or synonymous terms salva veritate.” If you cannot substitute synonymous terms without a change in the truth-conditions of a sentence, you cannot substitute synonymous terms in a sentence without a change in meaning. Now, the latter, the substitutability of synonyms salva significatione is logically equivalent to compositionality (as proven by Hodges, 2001). Cappelen and Lepore’s view of quotation hence logically entails that direct quotation be non-compositional. The widespread conviction that direct quotation should be treated in a non-compositional way – quasi as an exception to the principle of compositionality – is probably the reason why so few explicitly compositional treatments of direct quotation can be found in the literature. To my knowledge only three papers, Potts (1997), Pagin and Westerståhl (in press), and Werning (2005), have ever tried so. Due to space and time limits, I will not discuss the first two. A closer investigation, however, would reveal that the analyses Potts and Pagin and Westerståhl provide for direct quotation in natural language cannot be transferred to the domain of thought because they are either not properly compositional or they essentially make use of extralinguistic contextual arguments. 5. Direct quotation and compositionality To illustrate what kind of questions the compositionality requirement raises for the treatment of direct quotation, I will first turn to the most natural and still very widespread, so-called disquotational analysis of quotation. To make this analysis explicit, we first need a rudimentary formal framework. According to the standard view, syntactic operations are not operative on the set E of expressions of the language directly, but on a set T of underlying grammatical terms. The distinction between expressions and terms is useful to deal with lexically or syntactically ambiguous expressions such as bank or The boy watches the girl with the telescope. By convention, expressions are set in italics. An expressing function e with
e:T!E
ð3Þ
maps the set of terms onto the set of expressions and a meaning function l with
l:T!M
ð4Þ
maps the set of terms onto the set of meanings. The grammar of a language is a pair hT, Si where S is the set of syntactic operations. Each syntactic operation is a partial function on some Cartesian product of T into T. As a useful convention that allows us to reduce notational complexity we set primitive or complex expressions in indexed angel brackets in order to disambiguate them and to denote the underlying grammatical terms, e.g., hbanki1 and hbanki2. If an expression is unambiguous the index, by convention, is left away. In the disquotational analysis a syntactic operation of quotation q is adopted into the grammar. It is a total function defined as follows:
q : T ! T such that eðqðsÞÞ ¼ ‘eðsÞ’ and
lðqðsÞÞ ¼ eðsÞ:
ð5Þ
The grammatical term s that underlies the quoted expression e(s) is the argument of the syntactic operation of quotation q. The expression of a quotation, i.e., e(q(s)), is the concatenation of the onset quotation mark, the quoted expression, and the offset quotation mark: ‘e(s)’ – as is common, the operation of concatenation is left implicit to simplify notation. The meaning of a quotation is the quoted expression. This account of quotation might be called holophrastic because it takes the quoted expression as a whole and sets it in quotation marks. It can be shown that the disquotational analysis violates compositionality, provided the language to be considered contains synonymous expressions, e.g., the pair gorse and furze:
lðhgorseiÞ ¼ lðhfurzeiÞ
ð6Þ
Although the two expressions are synonymous, they are not identical:
gorse – furze:
ð7Þ
From our definition of q we derive the following:
lðqðhgorseiÞÞ ¼ gorse:
ð8Þ
lðqðhfurzeiÞ ¼ furze:
ð9Þ
M. Werning / Consciousness and Cognition 19 (2010) 751–761
757
From (7), (8), and (9) we may infer:
lðqðhgorseiÞÞ – lðqðhfurzeiÞ:
ð10Þ
If we, for the sake of the argument, assume compositionality, the meaning of a syntactically complex term q(s) should be some function fq of the meanings of its syntactic part s:
lðqðhgorseiÞÞ ¼ fq ðlðhgorseiÞÞ:
ð11Þ
Substitution of identicals according to (6) yields:
lðqðhgorseiÞÞ ¼ fq ðlðhfurzeiÞÞ:
ð12Þ
Due to compositionality it also holds that:
lðqðhfurzeiÞÞ ¼ fq ðlðhfurzeiÞÞ
ð13Þ
From (12) and (13), we hence get:
lðqðhgorseiÞÞ ¼ lðqðhfurzeiÞÞ:
ð14Þ
We have derived a contradiction with (10). The hypothetical assumption that the language be compositional must consequently be rejected. Similar arguments can be made for all holophrastic analyses of quotation provided they treat the contents of expressions in a context-invariant unambiguous way. To see this, one may adapt the argument to Davidson’s (1984) demonstrative analysis of direct quotation, which is another case of holophrastic quotation. As in any such analysis, quotation is taken as a syntactic operation q: T ? T such that e(q(s)) = ‘e(s)’. In the demonstrative analysis, for any sentential term u(n) in which a quotation may occur at position n, the meaning is given as the meaning of a discourse with backward demonstrative reference:8
lðuðqðsÞjnÞÞ ¼ lðs uðhthatijnÞÞ:
ð15Þ
Example: The meaning of
John said; ‘Furze is beautiful’
ð16Þ
is given as the meaning of the two-sentence discourse
Furze is beautiful: John said that:
ð17Þ
If we were licensed to assume compositionality, we should be allowed to replace synonyms in the discourse without changing its meaning. (17) should hence be synonymous to
Gorse is beautiful: John said that:
ð18Þ
However, this discourse according to (15) would have the same meaning as
John said; ‘Gorse is beautiful’:
ð19Þ
We have again derived a contradiction since (16) and (19) obviously have different truth-conditions. Compositionality must not be presupposed in the demonstrative analysis either. 6. Phonological quotation In two representative examples we have shown that any holophrastic treatment of direct quotation leads to a violation of the principle of compositionality unless one is ready to allow for extralinguistic context arguments.9 Permitting the latter would, however, block any transfer of the analysis from language to thought. Any analysis of direct quotation applicable to the domain of thought and thus possibly accounting for introspective self-awareness, must hence be non-holophrastic. It must refer to features of the quoted expression that are below the level of the whole expression. Such an analysis is justly called subsymbolic. To show that quotation in natural language can be analyzed in a compositional way without extralinguistic context arguments, I will here introduce the method of phonological quotation as an example of a subsymbolic analysis of direct quotation. The idea partly builds on Geach’s (1970) descriptive analysis of quotation. What’s new in the phonological analysis is that the means of reference to expressions is description by imitation. An expression is described by imitating its phonological structure. The phonological analysis is on par with alternative subsymbolic analyses that refer to expressions by imitating and describing graphical features of inscription or gestures of sign languages.
8 The substitution operator | in a term u(s|n) reads: replace the variable n – i.e. a placeholder that indicates a certain position – in the term u(n) with the term s. The syntactic full-stop operator . builds discourses from its left and right argument terms. 9 The formal proof is immediate and left to the reader.
758
M. Werning / Consciousness and Cognition 19 (2010) 751–761
Let us assume that each expression of our language be a temporal sequence of primitive phonological parts, e.g., phonemes, from a finite set. The expression dog, e.g., is a sequence of the plosive voiced dental consonant d, the closed-mid back vowel o, and the plosive voiced velar consonant g. To enable reference to the expressions of the language within the language, the set of terms T of the language has to include as a subset a set of atomic terms P that denote the elementary phonological parts necessary to generate each and every expression of the language. Neglecting the details of phonology, a crude approximation might be the following (capital letters – except the labels of sets – are used to refer to object language terms that are names of phonemes, whereas lower-case characters in our meta-language are used to refer to phonemes when they are not used as variables. The space bar _ denotes a distinctive length of silence):
P ¼ fA; B; . . . ; Z; g
ð20Þ
lðAÞ ¼ a; lðBÞ ¼ b; . . . ; lðZÞ ¼ z; lð Þ ¼ x milliseconds of silence:
ð21Þ
The phonological analysis of quotation to be developed here refers to phonemes and sequences of phonemes by imitation. Formally, the idea of imitation is captured by the fact that the phonemic terms are themselves expressed by the phonemes they denote:
eðAÞ ¼ a; eðBÞ ¼ b; . . . ; eðZÞ ¼ z; eð Þ ¼ x milliseconds of silence:
ð22Þ
To achieve reference to complex phonological sequences, a binary syntactic operation of concatenation is adopted into the grammar. The set of phonological descriptions is the closure P* of the set P with respect to concatenation.10 The syntactic operation of concatenation has a unique semantic counterpart operation: the function fs that maps pairs of phonological sequences to the sequence of them. Compositionality is warranted by the following homomorphism, with the variables s and t in P*: _
lðs_ tÞ ¼ fs ðlðsÞ; lðtÞÞ:
ð23Þ
To enable description by imitation also for complex expressions, the expressing function is to satisfy a similar constraint:
eðs_ tÞ ¼ fs ðeðsÞ; eðtÞÞ:
ð24Þ
Notice that all expressions of the language can be described by phonological descriptions, but usually not all concatenations do actually describe some word or some sequence of words of the language: The concatenation O_ G denotes the phonological sequences og, which is not a meaningful English word, while D_ ðO_ GÞ denotes the English word dog. The meaning of the quotation is derived compositionally:
lðD_ ðO_ GÞÞ ¼ fs ðlðDÞ; lðO_ GÞÞ ¼ fs ðD; fs ðlðOÞ; lðGÞÞÞ ¼ dog:
ð25Þ
Likewise the expression of the quotation is generated stepwise:
eðD_ ðO_ GÞÞ ¼ fs ðeðDÞ; eðO_ GÞÞ ¼ fs ðD; fs ðeðOÞ; eðGÞÞÞ ¼ dog:
ð26Þ
Quotation is hence expressed by an imitation of the phonological structure of the quoted expression. Be aware, however, that the expression of the quotation and the quoted expression are given different syntactic analyses, even though they exhibit the same phonological structure. In (25) the quoted expression dog is regarded as syntactically unstructured. It is the expression of the primitive, lexical term hdogi1:
dog ¼ eðhdogi1 Þ:
ð27Þ _
_
In (26) dog is regarded as the expression of the quotation D (O G):
dog ¼ eðD_ ðO_ GÞÞ:
ð28Þ
Keep in mind that in spoken language there is no phonological expression of the quotation marks we use in writing. The written expressions ‘dog’ and dog correspond to one and the same phonological sequence, and hence to one and the same expression in spoken language (however, see option three in the passage on higher-order quotation below). Notice that phonology is all there is to the identity of expressions in spoken language. The phonological analysis of quotation is here developed for spoken language and, thus, systematically introduces ambiguities on the level of expressions. These ambiguities, however, are syntactically resolved and do not pose any threat to a strict application of the principle of compositionality. For, compositionality regards the relation between terms and meanings, rather than the one between expressions and meanings. We may sum up that unlike holophrastic analyses, the phonological treatment provides a compositional analysis of direct quotation and avoids context arguments and syntactically unresolved ambiguities. I am aware that the phonological analysis is not a silver bullet against all sorts of difficulties quotation brings about in natural language. As with other analyses of quotation the treatment of mixed quotation is problematic (see Reimer,
10 To guarantee that terms are uniquely identified by a nested structure of syntactic operations, we adopt the convention that phonological sequences are described from the end. The innermost concatenation refers to the sequence of the final two phonemes of an expression. Thus D_ (O_ G) is a well formed term while ðD_ OÞ_ G is not.
M. Werning / Consciousness and Cognition 19 (2010) 751–761
759
2003). Here the quoted expression is used and mentioned at the same time (as in: Quine said that quotation ‘has a certain anomalous feature’). However, many other problems often brought forward against the phonological theory are less severe.11 Aside from being compositional, the phonological analysis has considerable further advantages over holophrastic analyses. Whereas holophrastic analyses only allow for the quotation of expressions with an existing underlying term, the phonological analysis immediately explains why non-words can be quoted, as in:
Hans always says ‘shnoff ’ when he speaks of snow:
ð29Þ
Holophrastic analyses furthermore have problems with explaining the quotation of syntactically ambiguous expressions. For, the syntactic operation of quotation takes as arguments only syntactically disambiguated terms. The phonological analysis has no problems here because it just recurs to the phonological surface of an expression. In a holophrastic framework there is, e.g., no syntactic analysis of the following sentence that renders the sentence true:
‘The boy watches the girl with the telescope’ is ambiguous:
ð30Þ
Of particular interest is the question whether phonological quotation as a way of description by imitation is consistent with the assumption that quotations can be recursively embedded, as in:
\dog" refers to ‘dog’; which refers to dogs:
ð31Þ
At first glance there seems to be no difference between the imitation of the imitation of the phonological structure of an expression and the simple imitation of the phonological structure of the expression. Both sound the same and, more importantly, have the same underlying syntactic structure. One option is to bite the bullet and argue that higher order quotation is in fact a notorious problem in spoken language. Only auxiliary conventions like reading the quotation marks aloud – quote dog unquote vs. quote quote dog unquote unquote – might be a means to circumvent the problem. A second option is to appeal to speech act theory and argue that, even though the phonological and syntactic structure of the expressions of a second and a first order quotation are the same, the illocutionary force and consequently the intentions of speaker and hearer are different. The quotation marks in writing could then be interpreted as signs of an imitative illocutionary act. Their role would thus be somewhat analogous to punctuation marks like the question mark or the exclamation mark. In many languages question marks and exclamation marks can, e.g., be placed after sentences that are otherwise indistinguishable from assertions. In those cases the punctuation marks indicate nothing but the change of illocutionary force. A third, rather technical option could be the following: We stipulate as a rule that each imitation start and end with a certain period of silence – or some other phonological marker: _dog_ would be the expression of the first order quotation D_ (O_ G); __dog__ the expression of the second order quotation _ (D_ (O_ (G_ _))); and so on. Notice that higher order quotations have to make explicit reference to the periods of silence in the expressions of lower-level quotations. This is because a full description of the lowerlevel expressions from the start to the end of the imitation has to be given. In this approach first and second order quotations now differ phonologically and syntactically. The quotation marks we use in writing could be interpreted as expressions of those periods of silence or simply as analogous tools. Regardless of whether the phonological analysis is the optimal treatment of direct quotation in natural language, what matters in the context of this paper is that it is the only strictly compositional analysis and as such the only one that may be transferred to the domain of thought – on a par with other comparable subsymbolic analyses. If self-awareness is regarded as self-reporting second-order thought that achieves access to an inner self, it has to be treated in analogy to direct quotation. Introspective thought is a form of direct mental quotation. However, if phonological or subsymbolic quotation is the only viable analysis of quotation that may be applied to the domain of thought, we may infer that awareness of an inner self taken as a second-order thought must be a form of phonological quotation or closely analogous to it. 7. Introspective self-awareness as the awareness of inner speech The conclusion that awareness of an inner self must be akin to phonological or otherwise subsymbolic quotation – think of representing or simulating the gestures of sign languages – has far reaching consequences. The most immediate consequence is that thoughts reported by introspective thoughts must have an accessible phonological or quasi-phonological structure (e.g., a subsymbolic gestural structure). The best and as far as I can see only candidate here is inner speech: the mental simulation of phonologically (or gesturally) structured speech. Inner speech seems to be a prerequisite for the awareness of an inner self. A second consequence is that thoughts cannot be introspectively reported if their constituents fail to 11 It is true, for instance, that in the phonological analysis the relation between the quotation and the quoted expression is less close than in holophrastic analyses – an often heard objection (Cappelen & Lepore, 2009) – because the quoted expression is not a syntactic constituent of the quotation, but merely referred to. I take this more as a strength than as a weakness. A sentence like ‘Dog’ refers to dogs, e.g., can no longer be regarded as a priori. Recall that the alleged aprioricity of such sentences generates various philosophical paradoxes. Take, e.g., Putnam’s (1981) famous brain-in-a-vat argument in the reconstruction of Wright (1992), shortest version: In my language ‘dog’ refers to dogs. In the language of a brain in a vat ‘dog’ does not refer to dogs. Therefore, I am not a brain in a vat. The problem with this argument is not its validity, but the alleged aprioricity of its two premises – the first justified by disquotation, the second by the causal theory of reference. Isn’t it paradoxical that I should be able to know a priori that I am not a brain in a vat? According to the phonological analysis, the first premise would not be a priori anymore, but boils down to something like the empirical statement A word referring to dogs in my language is the sequence of the plosive voiced dental consonant, the closed-mid back vowel, and the plosive voiced velar consonant. This sentence after some phonological scrutiny may eventually even turn out false.
760
M. Werning / Consciousness and Cognition 19 (2010) 751–761
have a phonological or otherwise subsymbolic accessible surface. This implies that the inner self is accessible only insofar as it is constituted by inner speech. If one furthermore makes the assumption that epistemic accessibility is an essential property of the inner self, as Descartes apparently did, the inner self reduces to that part of our mind that simulates phonologically or otherwise subsymbolically structured speech. One may thirdly infer that some competence of natural language is necessary to have introspective self-awareness. It is needless to say that this has numerous ontogenetic and phylogenic consequences. A fourth implication is that any impairment of our capacity to simulate inner speech should lead to an impairment regarding the inner self. There is indeed ample evidence for such a correlation (see Morin, 2005, for review). Such dysfunctions might be due to a stroke in language-related cortical areas, be caused by developmental retardation, or be grounded in psychiatric pathologies.12 Before we conclude, let me briefly summarize the overall argumentation of the paper: (P1) Self-awareness is a second-order state either in the domain of experience or in the domain of thought. (P2) If self-awareness is a second-order experience, and experience is phenomenally transparent, then self-awareness fails to achieve awareness of an inner self and is as such empty. (P3) If self-awareness is a second-order thought, then it is either (a) identified with a thought that is analogous to indirect quotation or (b) with one that is analogous to direct quotation. (P4) In the case of (a) self-awareness is an instance of extrospection, fails to achieve awareness of an inner self and is as such empty. (P5) Direct quotation is either analyzed holophrastically or phonologically (or otherwise subsymbolically). (P6) Holophrastic analyses of direct quotation fail to be strictly compositional whereas the phonological (or any likewise subsymbolic) analysis of quotation is strictly compositional. (P7) Thought is strictly compositional. (P8) Experience is phenomenally transparent. (C1) Hence: self-awareness is either empty as awareness of an inner self or analogous to phonological (or otherwise subsymbolic) quotation. (C2) Corollary: the inner self is only insofar accessible to awareness as it has an accessible phonological (or otherwise subsymbolic) structure, as apparently only inner speech does. (C3) Corollary: if being the object of self-awareness is essential to the inner self, then it apparently reduces to nothing but a stream of inner speech. I am aware that the argument of the paper explores only some portion of the relevant logical space. One might, e.g., question the exhaustiveness of the dichotomy in P1. Are experiences and thoughts – both understood in a broad sense – the only major categories of mental states? What about emotions?13 Even though there is no room for a closer discussion here, I would suggest distinguishing at least two groups: We have the group of low-level emotions like hunger and pain, which qualify as proprioceptive experiences and might be regarded as phenomenally transparent (see the foregoing treatment of pain). We also have the group of high-level emotions like hope and disappointment. These most likely are predominantly composed of thought-like states that involve beliefs, desires, and expectations. Here the compositionality requirement applies. In the context of introspective self-awareness and in the light of the argument in this paper, it should be of major concern whether the two groups exhaust the class of emotions or whether there are emotions that are neither experience-like nor thought-like. One may of course also raise more principled objections and question the compositionality of thought or the transparency of experience. It is a remarkable fact, though, that in the argument the property of phenomenal transparency does a similar job for experience as the property of compositionality does for thought. Despite potential points of criticism, each of the premises P1 to P8 seems to be rather well justified to me in its own light. This is especially important since the conclusions strike me as rather far-reaching. What about Descartes’s idea of an inner self? We have largely, however, not completely discarded this idea. Much of what Descartes suggested as belonging to the inner self can no longer be regarded as a part of it if being the object of introspective awareness is essential to the inner self. Still, there is a remainder: linguistically structured inner speech. This conclusion, of course, kicks off a whole cascade of follow-up questions: Do we have to postulate an author behind inner speech that remains unobserved in respects other than the phonological/subsymbolic structure we are introspectively aware of? Is there an issue of ownership? Does the phonological/subsymbolic structure of which one becomes aware have to be attributed to oneself in order to constitute the inner self of one’s own? May this attribution of ownership eventually even go wrong so that the speech-like structure is attributed to nobody or to somebody else? Do features internal to inner speech suffice to establish the unity of an inner self? Speech-internal features that indicate unity might be semantic such as coherence, syntactic as, e.g., anaphora and other means to interweave sentences into a discourse, and finally phonological and phonetic such as accent or voice. Similar questions arise for the uniqueness of the inner self. I am optimistic that the issues of authorship, ownership, unity, and uniqueness can in principle be dealt with in an inner speech account of the self. We may hence hope to 12 A good candidate for a correlation between inner speech deficits and an impaired inner self in psychiatry are some forms of schizophrenia as pointed out by Hoffman (1986). See Stephens and Graham (2000) for discussion. 13 For a discussion of the phenomenal transparency of emotions see Metzinger (2003). Imaginations also constitute a possible exception to transparency, for discussion see Werning (2004a).
M. Werning / Consciousness and Cognition 19 (2010) 751–761
761
save at least some fragment of Descartes’s idea of an inner self without falling back into the Cartesian metaphysics of a thinking substance and related conceptions of the ‘‘I”. References Beckermann, A. (2009). Es gibt kein Ich, doch es gibt mich. In M. Fürst, W. Gombocz, & C. Hiebaum (Eds.), Gehirne und Personen–Beiträge zum 8. Internationalen Kongress der Österreichischen Gesellschaft für Philosophie in Graz (pp. 1–17). Frankfurt, Germany: Ontos. Botvinick, M., & Cohen, J. (1998). Rubber hands ‘feel’ touch that eyes see. Nature, 391, 756. Brown, J. W. (2004). The illusory and the real. Mind and Matter, 2, 37–59. Cappelen, H., & Lepore, E. (2009). Quotation. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. Stanford, CA. Davidson, D. (1984). Quotation. In Truth and interpretation (pp. 79–92). Oxford: Clarendon Press. Dennett, D. C. (1992). The self as a center of narrative gravity. In F. Kessel, P. Cole, & D. Johnson (Eds.), Self and consciousness: Multiple perspectives. Hillsdale, NJ: Erlbaum. Descartes, R. (1904). Meditationes de prima philosophia. In C. Adam & P. Tannery (Eds.), Oeuvres de Descartes (Vol. VII, pp. 1–90). Paris (Year of original: 1641). Dretske, F. (2003a). How do you know you are not a zombie? In B. Gertler (Ed.), Privileged access. Philosophical accounts of self-knowledge (pp. 1–14). Burlington: Ashgate Publishing. Dretske, F. (2003b). Sensation and perception. In Y. H. Gunther (Ed.), Essays on nonconceptual content. Cambridge, MA: MIT Press (Year of original: 1981). Fernández, J. (2003). Privileged access naturalized. The Philosophical Quarterly, 53, 352–372. Fernández, J. (2005). Privileged access revisited. The Philosophical Quarterly, 55, 102–105. Fodor, J. A. (2001). Language, thought and compositionality. Mind & Language, 16(1), 1–15. Gallagher, S. (2005). How the body shapes the mind. Oxford: Oxford University Press. Geach, P. (1970). Quotation and quantification. In Logic matters (pp. 205–208). Oxford: Basil Blackwell. Greenberg, M. (2005). A new map of theories of mental content. Nous, 39, 299–320. Harman, G. (1990). The intrinsic quality of experience. Philosophical Perspectives, 4, 31–51. Heidegger, M. (1977). The origin of the work of art. In Basic writings (pp. 139–211). New York: Harper and Row (Year of original: 1935). Hodges, W. (2001). Formal features of compositionality. Journal of Logic, Language and Information, 10, 7–28. Hoffman, R. E. (1986). Verbal hallucinations and language production processes in schizophrenia. Behavioral and Brain Sciences, 9, 503–517. Kaplan, D. (1989). Demonstratives. In Themes from Kaplan (pp. 481–563). Oxford: Oxford University Press. Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317, 1096–1099. McDowell, J. (1996). Mind and world. Cambridge, MA: Harvard University Press. Metzinger, T. (2003). Phenomenal transparency and cognitive self-reference. Phenomenology and the Cognitive Sciences, 2, 353–393. Moore, G. E. (1903). The refutation of idealism. Mind, 12, 433–453. Morin, A. (2005). Possible links between self-awareness and inner speech: Theoretical background, underlying mechanisms, and empirical evidence. Journal of Consciousness Studies, 12, 115–134. Pagin, P., & Westerståhl, D. (in press). Compositionality. In K. von Heusinger, C. Maienborn, & P. Portner (Eds.), Semantics. An international handbook of natural language meaning. Berlin: Mouton de Gruyter. Potts, C. (1997). The dimensions of quotation. In C. Barker & P. Jacobsen (Eds.), Direct compositionality (pp. 405–431). Oxford: Oxford University Press. Putnam, H. (1981). Brains in a vat. In Reason, truth, and history (pp. 1–21). Cambridge: Cambridge University Press. Reimer, M. (2003). Too counter-intuitive to believe? Pragmatic accounts of mixed quotation. Belgian Journal of Linguistics, 17, 167–186. Ricoeur, P. (1992). Oneself as another. Chicago: University of Chicago Press. Stephens, G. L., & Graham, G. (2000). When self-consciousness breaks: Alien voices and inserted thoughts. Cambridge, MA: MIT Press. Werning, M. (2004a). Self-awareness and imagination. In J. Sàágua (Ed.), The explanation of human interpretation (pp. 369–377). Lisbon: Colibri. Werning, M. (2004b). Compositionality, context, categories and the indeterminacy of translation. Erkenntnis, 60, 145–178. Werning, M. (2005). Right and wrong reasons for compositionality. In M. Werning, E. Machery, & G. Schurz (Eds.), The compositionality of meaning and content (Vol. I: Foundational Issues, (pp. 285–309). Frankfurt: Ontos Verlag. Wright, C. (1992). On Putnam’s proof that we are not brains-in-a-vat. Proceedings of the Aristotelian Society, 92, 67–94.
Consciousness and Cognition 19 (2010) 762–777
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
The others: Universals and cultural specificities in the perception of status and dominance from nonverbal behavior q Gary Bente a,*, Haug Leuschner a, Ahmad Al Issa b, James J. Blascovich c a
Department of Psychology, University of Cologne, Germany Department of English, American University of Sharjah, United Arab Emirates c Department of Psychology, University of California, Santa Barbara, United States b
a r t i c l e
i n f o
Article history: Available online 13 July 2010 Keywords: Culture Status Dominance Evaluation Nonverbal behavior Computer-animation
a b s t r a c t The current study analyzes trans-cultural universalities and specificities in the recognition of status roles, dominance perception and social evaluation based on nonverbal cues. Using a novel methodology, which allowed to mask clues to ethnicity and cultural background of the agents, we compared impression of Germans, Americans and Arabs observing computer-animated interactions from the three countries. Only in the German stimulus sample the status roles (employee vs. supervisor) could be recognized above chance level. However we found significant correlations in dominance perception across all countries. Significant correlations were only found for evaluation between German observers and observers from the other two countries. Perceived dominance uniformly predicted the assignment of status-roles in all cultures. Microanalysis of movement behavior further revealed predictive value of specific nonverbal cues for dominance ratings. Results support the hypothesis of universalities in the processing of dominance cues and point to cultural specificities in evaluative responses to nonverbal behavior. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Man is a ‘‘social animal” (Aronson, 1972) equipped with a unique capacity to process and to adapt to complex affordances created by our social environment (Freeman, Rule, & Ambady, 2009). Culture has been identified as a core factor influencing this capacity. With regard to social information processing it has been conceptualized as a human universal as well as a cause of diversity (see Chiao & Ambady, 2007), either applying a general or a particular concept (Vogeley & Roepstorff, 2009). With respect to the biological foundations of social cognition humans can be considered as equal and as distinct from all other living beings. Given that evolution created the biological basis for symbolic interaction, mutual understanding, and social organization, culture describes a general human achievement emerging from, and at the same time driving the particular phylogeny of the human brain (Nettle, 2009; Tomasello, 1999; Tomasello, Carpenter, Call, Behne, & Moll, 2005). Culture in this perspective is conceptualized as a general or universal characteristic of humans’ cognitive capacities enabling us to construe a system of social structures, rules and rituals to cope with environmental and social challenges and to buffer collective and individual needs and urges. Different cultures however seemingly have found their own way to adapt to these life affordances, and furthermore also created different environments and communication systems to which homo-sapiens had to adapt (Chiao & Ambady, 2007; Rohner, 1984). Viewed this way culture no longer comes as a universal but as a cause of diversity, instantiated in self construal, cognitive styles, perceptual schemata and communication patterns which serve to q
This article is part of a special issue of this journal on Self, Other and Memory. * Corresponding author. E-mail address:
[email protected] (G. Bente).
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.006
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
763
assign meaning to the physical and social world (Markus & Kitayama, 1991; Kitayama, Duffy, Kawamura, & Larsen, 2003; Nisbett, Peng, Choi, & Norenzayan, 2001). Consequently the question has repeatedly been raised regarding to which degree and under which conditions universals in social information processing override cultural specificities or, conversely, whether different cultures imply distinct cognitive implementations, which influence the way we perceive ourselves and others (Markus & Kitayama, 1991; Matsumoto, 2003, 2006; Zebrowitz-McArthur, 1988) and which might even affect the way our brain works (Hedden, Ketay, Aron, Markus, & Gabrieli, 2008; Chiao et al., 2010; White, Lehman, & Cohen, 2006). We consider expressions and perceptions of power and dominance as a paradigmatic case to approach this question. Social hierarchies are ubiquitous in human societies and successful navigation of the social world implies particular skills to produce and process social cues which reflect, construct or consolidate the vertical dimension of social systems. There is ample evidence that dominance and power are rarely communicated explicitly but mainly expressed through subtle nonverbal cues (Aguinis, Simonsen, & Pierce, 1998; Argyle, Salter, Nicholson, Williams, & Burgess, 1970; Burgoon, 1994; Carli, Martin, Leatham, Lyons, & Tse, 1993; Dovidio & Ellyson, 1982; Dovidio, Ellyson, Keating, Heltman, & Brown, 1988; Edinger & Patterson, 1983; Lee, Matsumoto, Kobayashi, Krupp, & Maniatis, 1992; Mehrabian, 1969, 1970; Mignault & Chaudhuri, 2003; Remland, 1982). It could be shown that culture is influential in molding these nonverbal expressions as well as their perception and cognitive processing (Kowner & Wiseman, 2003; Matsumoto, 2006; Sussman & Rosenfeld, 1982). What poses a particular problem for our understanding of the subtle dynamics of nonverbal behavior is the fact that it is largely produced and processed automatically and without conscious awareness (see Burgoon, Berger, & Waldron, 2000; Choi, Gray, & Ambady, 2005; Newman & Uleman, 1989; Uleman & Bargh, 1989). Andersen (1999) commented on the potential implications for cross-cultural communication: ‘‘Because we are usually not aware of our own nonverbal behavior, it becomes extremely difficult to identify and master the nonverbal behavior of another culture” (p. 258). Against this background the current study aimed to identify universals and cultural specificities in the perception of nonverbal behavior and to answer the question whether people from different cultures are equally able to identify dominance and underlying status roles in nonverbal interactions shown in their own or in foreign cultures.
2. Background Cultures have been shown to differ with regard to the value they assign to the vertical dimension of social relations (Hofstede, 1980; Triandis & Gelfand, 1998), either accepting that inequalities in power and status are natural or existential or seeing them as man-made and largely artificial (see Naylor, 2009). Against this background it doesn’t come as a surprise that cultures are regarded as a major source of diversity in the manifestation of power and consequently as a potential cause of misunderstanding when it comes to the perception and interpretation of dominance cues in intercultural communication (Aguinis et al., 1998; Gudykunst & Ting-Toomey, 1988; Gudykunst, Ting-Toomey, & Nishida, 1996). However, data on cultural diversity in this regard is elusive baring some evidence that nonverbal expressions of dominance, although based on culture-specific attitudes towards status and power, are recognized and processed as trans-cultural universals (Burgoon, Buller, & Woodall, 1996; Guerrero & Floyd, 2005; Keating, 1985; Kowner & Wiseman, 2003). A clarification of the conflating concepts involved in this realm might be helpful to approach the partly contradictory results (see Burgoon, Johnson, & Koch, 1998). According to Dunbar and Burgoon (2005) the terms status, power and dominance should be treated as interrelated but nevertheless ‘‘separate constructs” (p. 208). With regard to dyadic power theory (Dunbar, 2004; Rollins & Bahr, 1976) power can be conceptualized as a latent variable (Komter, 1989), i.e. having the potential to influence others, which can rely on different power bases (French & Raven, 1959), such as explicit rules (legitimate power), means control (coercive and reward power), quality of relation (referent power) or skills and knowledge (expert power). Status thus can be understood as a structural basis of (legitimate) power, derived from asymmetric role assignments in social systems (superior vs. subordinate). In contrast, dominance describes an overt phenomenon which is manifest in the interaction patterns, i.e., a set of ‘‘. . .expressive, relationally based communicative acts by which power is exerted and influence achieved” (Dunbar & Burgoon, 2005, p. 209; see also Burgoon & Dunbar, 2000; Burgoon et al., 1998). According to Kowner and Wiseman (2003) cultural values influence different instantiations of status and dominance on different levels: ‘‘Culture, we argue, determines the values attached to status, but it also affects the magnitude at which a person’s status is manifested through behavior in a given society” (p. 206). In a cross-cultural business negotiation, for example, cultural display rules might reveal different expressions of status in different groups; nevertheless observers could agree upon who is dominant and who is submissive. Moreover, inferences about underlying status relations could converge independently from what a particular culture would consider appropriate or successful in asymmetric interactions. Thus we can assume different layers of dominance perception: a descriptive layer, potentially relying on universal perceptual mechanisms, and an evaluative layer relying on culture-specific comparisons to the self and experiences with other members of the in-group. Kowner and Wiseman (2003) found that for the US and Japan although the descriptions of typical status-related behavior were ‘‘far from identical” (p. 207), the named behaviors were unanimously interpreted as either dominant or submissive across the cultures. These results support the hypothesis of culture-specific expressions of status and power (see also Triandis & Gelfand, 1998; Triandis & Suh, 2002), and also show that status-relevant cues can be identified by members of other cultures, even though not part of their own repertoire. In a recent brain imaging study Freeman, Rule, Adams, and Ambady (2009) provided additional evidence that members of different cultures (in this case Americans and Japanese) are equally able to differentiate dominant and submissive postures, displayed on digitally edited photographs (contour
764
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
images), but that both groups show distinct behavioral as well as neural responses to both classes of stimuli, which indicated different evaluative processes. Following different approaches both studies thus point to universals in the description and identification of dominance cues across cultures, but also suggest specificities in the evaluative component of person perception. Moreover as shown by Matsumoto and Kudoh (1987) different cultures also might give different priority to both impression dimensions when judging other people. While judgments of Japanese observing specific body postures were mainly concerned with status and power issues Americans focused on evaluative aspects (like or dislike). It seems promising against this background to more systematically differentiate dominance perception respectively the recognition of status roles from evaluative responses reflecting the perceived appropriateness and likability of the observed behavior. Beyond the unclear distinction between descriptive and evaluative aspects in impression formation the studies of Kowner and Wiseman (2003), Freeman, Rule, and Ambady (2009) and Freeman, Rule, Adams, et al. (2009) also bare a common methodological problem. They are based on preselected, prototypical dominance cues and widely neglect the subtle dynamics and inter-individual variance of nonverbal behavior, which to some degree limits ecological validity of the studies. In principle the use of video recordings of social interactions could help to solve this problem. However video samples carry category relevant information (e.g., clues to ethnicity and nationality) attached to the outward appearance of the actors (e.g. skin color, facial features, clothing) and thus are likely to activate particular stereotypes (see Bargh, Chen, & Burrows, 1996; Chen & Bargh, 1997) which could contaminate the effects of nonverbal cues. To escape from this dilemma we have suggested to employ computer-animations of neutral artificial characters instead of videos (Bente, 1989; Bente, Feist, & Elder, 1996). As revealed through earlier studies, character animations are able to accurately reproduce video recorded nonverbal behavior in its spatial details and subtle dynamics (Bente, Krämer, & Petersen, 2002; Bente, Krämer, Petersen, & de Ruiter, 2001; Bente, Petersen, Krämer, & de Ruiter, 2001) and to induce social impressions comparable to the original video sequences they are based on (Bente, Krämer, et al., 2001). This methodology has recently been introduced in cross-cultural research (Bente, Senokozlieva, Pennig, Al-Issa, & Fischer, 2008) and has been used in the current study to identify universalities and cultural specificities in the recognition of status roles, in the perception of dominance and in social evaluation based on nonverbal behavior. In particular we asked whether status-roles assigned to interlocutors can be identified independently from the country of origin of actors and observers or whether there is an in-group advantage for the recognition of power status comparable to the one described for the recognition of emotions from facial expressions (Elfenbein & Ambady, 2002a, 2002b; Elfenbein & Ambady, 2003; Matsumoto, 2002; Matsumoto, Olide, & Willingham, 2009). Further we aimed to explore the influence of culture on the production and perception of social dominance cues and liking and to determine how these two impression dimensions covary within and across cultures. The last step of our analyses should answer the question how selected nonverbal cues influence the perceptions of dominance and social evaluation and how perceived dominance leads to status-role ascriptions in different cultures. 3. Method 3.1. Selection of cultures The selection of cultures for the current study was guided by theoretical considerations and previous data on cultural differences as well as considerations of societal relevance. We selected three cultures described as either similar or dissimilar with regard to three basic cultural value dimensions ‘‘Power Distance (PDI)”, ‘‘Individualism/Collectivism (IDV)” and ‘‘Uncertainty Avoidance (UAI) (see Hofstede, 1980, 2003; Triandis, 1995; Triandis, Bontempo, Villareal, Asai, & Lucca, 1998). With regard to PDI Germany and the USA are described to be more similar and both different from UAE (PDI-scores: USA = 40; GER = 35; UAE = 80), Germany is placed on the IDV-rankings between the two other countries (IDV-scores: USA = 91; GER = 67; UAE = 38) and is placed on the UAI-ranking closer to the UAE (UAI-scores: USA = 46; GER = 65; UAE = 68) (see Clearly Cultural, 2009). As all three dimensions have to be considered as influential with regard to behavioral manifestations of power and dominance in particular in interpersonal conflict situations (Ting-Toomey & Oetzel, 2001, 2002) we overall assumed USA and UAE would provide the most different cultural background and Germany would range somewhere in-between the two. 3.2. Stimulus material The stimulus material consisted of 30 one-minute dyadic interaction sequences (10 from each country: Germany, USA and UAE). Voluntary male student participants were recruited for role play interactions at the University of Cologne (Germany), the University of California, Santa Barbara (USA), and the American University of Sharjah (UAE). Only applicants who were born and grew up in the respective country were included in the sample. The participants were informed that they could stop participation at any time and could ask for deletion and non-use of the video recordings. Furthermore they were informed that the video recordings would not be used in the subsequent perception study, but instead the study would utilize reconstructed computer animations, thus guaranteeing anonymity. Role players were instructed to solve a conflict in a managerial interaction either by taking the role of the employee or the supervisor. Participants were randomly assigned to the dyads as well as to the different roles (employee vs. supervisor). The task implied an interaction between a supervisor and an important, but recently unreliable employee. To increase self involvement the participants were told that this task is
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
765
Fig. 1. Snapshots from the original videos and corresponding screenshots from computer-animations (from: Bente et al., 2008).
often utilized as a diagnostic tool with respect to the capability for professional conflict resolution. During debriefing the participants were asked whether they encountered any problems in identifying with the assigned role and in performing the task, which was not the case. Dyadic interactions were video recorded at the three institutions using a standardized central camera perspective. Oneminute sequences were selected from each of the recordings beginning after the actors had taken their places and completed greetings. Movement behavior was transcribed from these video sequences using a specially developed computer-assisted coding software as described in Bente et al. (2008). Computer animations of the dyadic interactions were rendered from these protocols with 25 frames per second using culture fair computer characters (avatars) appearing as wooden manikins. Fig. 1 shows three screen capture images of the original videos and the corresponding frames from the computer animation sequences. The resulting computer animation sequences accurately featured all details and dynamics of movement behavior visible in the video recordings, excluding facial activity. 3.3. Design of the perception study The computer animated stimuli were shown to observers in Germany, the USA and the UAE in a 3 3 2 2 experimental design, including one within-subject factor: stimulus origin (Germany, USA and UAE) and three between-subjects factors: observer nationality (Germany, USA and UAE), stereotype activation (with and without previous information about the origin of the stimuli) and role of the stimulus person (employee vs. supervisor). Pretests in Germany revealed that a maximum of 15 clips could be shown to each observer. Thus we decided to split the 3 10 stimulus sequences into two batches, to be shown to two independent observer groups in each country (five clips for each stimulus country). While this worked out for the observer samples in Germany and the USA the batch size for the UAE observers had to be cut down to 10 interaction sequences per observer instead of 15 after a first test, due to organizational time constraints (lecture pauses). Thus the first stimulus batch in the UAE contained four interactions from Germany, three interactions from the US and four interactions from the UAE and the second batch contained three interactions from Germany, three interactions from the US and four interactions from the UAE. 3.4. Participants Overall a total of 570 (USA: 187; GER: 189; UAE: 194) student participants took part in the observation study. The average age of the participants was 21.4 years, SD = 3.93 (USA: 20.1, SD = 2.85; GER: 23.8, SD = 5.15; UAE: 20.4, SD = 1.88). The sample consisted of 283 male and 287 female participants (USA: 94/93; GER: 91/98; UAE: 98/96), who were randomly assigned to the experimental conditions (stereotype activation: information vs. no information about the nationality of the stimulus dyads) in each country. In each condition, approximately half of the subjects were instructed to evaluate the person in the animation video sitting on the left side (employee) respectively on the right side (supervisor) of the screen from their point of view. Groups were balanced for gender. Cell frequencies combined for male and female observers varied between 45 and 50.
766
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
3.5. Procedure Three rotations were applied to stimulus presentations in each experimental condition to avoid serial effects. Participants completed the experiment in groups of up to 16 participants (maximum number for one stimulus rotation). For each condition (with and without pre-information about the stimuli’s country of origin) and for each of the three clip orders, participants took part in separate sessions. Before viewing the stimulus material the participants filled out a socio-demographic questionnaire. The participants were informed that the research objective was to learn more about the effects of nonverbal behavior on impression formation. Instructions were read to the participants in a standardized form indicating which person on the screen to focus and how to fill out the questionnaires. To analyze the effects of stereotype activation half of the subjects were informed about the country of origin of the dyads prior to each animation sequence (e.g., ‘‘the next dyad is coming from the US”). After the first sequence participants were asked about difficulties following the instruction and in the few cases were questions were raised (two groups in UAE) the instruction was repeated and if necessary explained in more detail. The animation clips (AVI-files) were then successively displayed via LCD wall projection. Following each animation sequence, the participants were prompted to fill out the questionnaire containing impression ratings (21 items), and a question about the assumed role of the interlocutors. After completion of the questionnaire, the next sequence was played. This procedure was repeated until all animations had been presented. The participants were then debriefed and received an incentive of 15 Euro (Germany), the corresponding sum in Dollars (USA) or a lunch voucher of the same value (UAE). 3.6. Dependent variables Status-role recognition was assessed by a single forced choice item asking which role combination was presented in the animation sequence viewed from the perspective of the target person (either right or left person on screen), thus achieving status-role ascriptions for both actors from all participants. Possible selections were: supervisor/employee, employee/supervisor, employee/employee, supervisor/supervisor. Seven-point bipolar adjective pairs were used to assess the observers’ impressions formed on the basis of the computeranimated nonverbal interactions. Overall 21 items were compiled covering basic dimensions of person perception as described in the literature (Andersen & Andersen, 2005; Andersen, Andersen, & Jensen, 1979; Dunbar & Burgoon, 2005; Kudoh & Matsumoto, 1985; Matsumoto & Kudoh, 1987; McCroskey & Jenson, 1975; Mehrabian, 1970; Osgood, 1966): In particular the list contained four items for Dominance (dominant–submissive, weak–strong, respectful–disrespectful and confident– unconfident) and four items for Evaluation (unfriendly–friendly, believable–unbelievable, likable–dislikeable and cold–warm-hearted). Further four items for Competence (incompetent–competent, intelligent–unintelligent, uninformed– informed and expert–inexpert) were selected from McCroskey and Jenson’s (1975) Source Credibility Scale (four items with highest loadings >.71) assuming particular correlations with the Dominance items. Beyond these items relevant to the current research questions the list further contained four items for Activity (rigid–flexible, dynamic–static, passive–active and lively–still), four items for Composure (anxious–calm, tense–relaxed, poised–nervous and composed–excitable) and one item for Masculinity (masculine–feminine; see Hofstede, 1980). The latter dimensions were included in exploratory factor analysis to ensure differential construct validity of the target dimensions but not treated as dependent variables in further analysis at this stage. To ascertain semantic equivalence of the items across cultures, all adjectives originally formulated in English were translated into German and Arabic by bilingual translators, and were then translated back to English by another translator. In the very few cases of resulting discrepancy both translators had to agree upon the most appropriate translation. 4. Results 4.1. Recognition of status roles First data inspection indicated a strong tendency for all observers to categorize the observed posers as employees. To correct for this response bias we calculated the ‘‘unbiased hit rate” (Wagner, 1993, 1997), which relates hit rates to false alarms (see also Elfenbein, Mandal, Ambady, Harizuka, & Kumar, 2002). So-called confusion matrices, including unbiased hit rates as well as unbiased chance values were calculated for each stimulus person. To test unbiased hit rates against chance value separate t-tests were conducted for each stimulus country separately. Significant effects above chance level were only found for German posers (employees: Mu = .357, SDu = .188; Mc = .278, SDc = .071), t(53) = 3.565, p = .000, d = .556; supervisors: Mu = .299, SDu = .198; Mc = .215, SDc = .072), t(53) = 3.587, p = .000, d = .564. No significant hit rates above chance level were found for USA and UAE posers, indicating that only in the German dyads the assigned status roles were recognizable from nonverbal behavior. To examine potential in-group advantage in status-role recognition we further conducted six separate ANOVAS for each combination of country of stimulus (USA, GER, and UAE) and role (supervisor and employee) with N = 60 stimuli and unbiased hit rate as dependent variables. No significant results were found to support the hypothesis of an in-group advantage in status-role recognition. Observers were no better in recognizing the status roles in the interactions from their own culture than from the other countries.
767
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777 Table 1 Pearson-correlation coefficients for status-role ascriptions. Ascribed role
Country of observers
Employee
USA
GER
Supervisor
USA
GER
* **
GER
UAE
r p N r p N
.759** .000 60
.699** .000 40 .679** .000 40
r p N r p N
.736** .000 60
.702** .000 40 .653** .000 40
Significant at the 0.05 level (2-tailed). Significant at the 0.01 level (2-tailed).
Two possible reasons might account for the low hit rates in particular for the US and the UAE sample. Either no meaningful status cues appeared in the interactions thus leaving the status-role recognition open for random guesses, or nonverbal status cues did not systematically covary with the roles as expected, i.e. there were employees showing high status behavior and bosses behaving in a low status manner. If the latter explication holds true we should be able to find high correlations in status role ascription across stimuli independently from the actual role assigned to the actors. This was in fact the case. Table 1 depicts the results of the corresponding analysis showing highly significant correlations across all countries in the ascription of both status roles. This result not only suggests a systematic influence of nonverbal behavior on status role ascription but also points to trans-cultural universals in the perception of the status-relevant cues. 4.2. Dimensions of impression formation As a basis for the comparative analysis of the impression ratings construct validity of the target dimensions Dominance and Evaluation as well as measurement invariance across cultures for these latent dimensions had to be established (Meredith, 1993; Wu, Li, & Zumbo, 2007). To determine the factor structure of the impression ratings we first analyzed the complete item set with Exploratory Factor Analysis (EFA) and Reliability Analysis (RA). Based on these results we set up a factor model according to our hypothetical model and used Multiple Group Confirmatory Factor Analysis (MG-CFA) to refine the factor model and examine the equivalence of measurement across all three countries. Rational and detailed results of EFA, RA and MG-CFA are provided as Complementary Material.1 EFA and RA lead to a reduced set of seven items revealing a two factor solution representing the dimensions Dominance (confident–unconfident, weak–strong, dominant–submissive, expert–inexpert) and Evaluation (unfriendly–friendly, likable– dislikeable, respectful–disrespectful). A few deviations from the item groupings found in the literature occurred in EFA. The item ‘‘expert–inexpert”, expected to load on an independent Competence factor loaded highly on the Dominance dimension. Further the item ‘‘respectful–disrespectful”, expected to load on the Dominance dimension showed high loadings on the Evaluation dimension (see Complementary Material). The model test for the two latent judgment dimensions Dominance and Evaluation by means of MG-CFA revealed strict factorial invariance across the countries (USA vs. GER vs. UAE) in all relevant fit criteria (SRMR = .039, RMSEA = .063, CFI = .903, see Complementary Material) and therefore the resulting factor scores could be used for ANOVA group comparisons (Meredith, 1993; Vandenberg & Lance, 2000; Wu et al., 2007). Factor score weights from the multiple-group model were then used to calculate individual values for the latent factors Dominance and Evaluation. The Pearson-correlation coefficient between the resulting scores was r = .012, p = .283 suggesting two largely independent latent impression dimensions. 4.3. Influence of independent variables on impression formation The influence of the independent variables on Dominance and Evaluation was analyzed by four-way (3 3 2 2) repeated measure ANOVA with country of stimulus CS: (USA, GER or UAE) as within-subjects factor and country of observer (CO: USA, GER or UAE), role (Ro: supervisor or employee) and advance information about the dyad’s country of origin (AI: information available or not available) as the between-subjects factors with N = 570 participants. In addition, pair wise comparisons with Bonferroni adjustments were conducted to determine simple main effects.
1
Complementary Material can be downloaded from the journal’s website.
768
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
Table 2 Means and standard deviations of dominance and evaluation referred to country of stimulus and role. Role
Country of stimulus USA
GER
UAE
Total
M SD
3.946 .802
3.821 .743
4.094 .646
3.953 .731
Supervisor
M SD
3.939 .771
4.185 .675
4.186 .751
4.103 .732
Total
M SD
3.942 .786
4.000 .733
4.139 .701
4.027 .740
M SD
3.669 .608
3.688 .565
3.762 .551
3.707 .575
Supervisor
M SD
3.573 .612
3.726 .539
3.795 .607
3.698 .586
Total
M SD
3.622 .611
3.707 .552
3.778 .579
3.702 .581
Dominance Employee
Evaluation Employee
Table 3 ANOVA tables of four-way repeated measures ANOVA for dominance and evaluation with the factors country of stimulus. Country of observer, role, and advance information. F
p
g2
6.113 .547 5.408 .103 2.759 .619 .248 .495 .556
11.002 .984 9.732 .186 4.966 1.115 .446 .891
.000 .413 .000 .822 .001 .347 .633 .465
0.012 0.002 0.011 0.000 0.011 0.003 0.001 0.002
2 1 1 2 2 1 2 557
.274 9.494 .202 .267 .536 .000 .337 .537
.509 17.680 .377 .498 .998 .001 .627
.601 .000 .540 .608 .369 .979 .535
0.001 0.010 0.000 0.001 0.001 0.000 0.001
7.102 3.970 1.720 .003 1.245 2.627 .156 1.096 261.480
2 4 2 2 4 4 2 4 1114
3.551 .993 .860 .001 .311 .657 .078 .274 .235
15.128 4.229 3.663 .006 1.326 2.798 .333 1.168
.000 .002 .026 .994 .258 .025 .717 .323
0.012 0.007 0.003 0.000 0.002 0.005 0.000 0.002
34.468 .030 .229 .154 .184 2.229 .381 266.004
2 1 1 2 2 1 2 557
17.234 .030 .229 .077 .092 2.229 .191 .478
36.088 .064 .479 .161 .193 4.668 .399
.000 .801 .489 .851 .824 .031 .671
0.059 0.000 0.000 0.000 0.000 0.004 0.001
Source of variance
SS
df
Dominance: within-subjects effects Country of stimulus (CS) CS CO CS Role CS Info CS CO Role CS CO Info CS Role Info CS CO Role Info Error(CS)
11.760 2.104 10.403 .199 10.616 2.383 .476 1.906 595.394
1.924 3.848 1.924 1.924 3.848 3.848 1.924 3.848 1071.534
Dominance: between-subjects effects Country of observer (CO) Role Info CO Role CO Info Role Info CO Role Info Error
.547 9.494 .202 .535 1.072 .000 .673 299.108
Evaluation: within-subjects effects Country of stimulus (CS) CS CO CS Role CS Info CS CO Role CS CO Info CS Role Info CS CO Role Info Error(CS) Evaluation: between-subjects effects Country of observer (CO) Role Info CO Role CO Info Role Info CO Role Info Error
MS
Note: The ANOVA for dominance is greenhouse-Geisser corrected; effect size is calculated by g2 = SSbetween/SStotal.
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
769
Fig. 2. Interaction between Role and stimuli’s Country of Origin in Dominance ratings (scale ranges from 1 to 7 with 4 as the scale mean).
The specified independent variables only accounted for 5.5% of variance in the dominance ratings and for 9.5% in the evaluative ratings. Due to the sample size of 570 participants, the test power of the ANOVAs was sufficiently high to confirm even smallest effects. In fact, group means deviated only minimally from the scale means (scale means = 4; see Table 2) and all significant results showed very weak effect sizes (see Table 3). Thus we refrain from far going interpretations and only point out two results relevant to the further analysis. First it is remarkable that advance information about the stimuli’s country of origin showed no significant main effect, which puts some emphasis on the predominant role of nonverbal behavior in impression formation. However it cannot be excluded that the stereotype induction intended was to weak or there were no preconceptions in the minds of the observers, which could have exerted a systematic influence on their judgments. Secondly a significant main effect of status role and an interaction effect between culture and status role could be found. The interaction is depicted in Fig. 2. Separate paired t-tests for the different stimulus countries revealed that the effects were due to differences in the German stimulus sample only, showing that across all observes German supervisors were perceived as more dominant as the employees (t = 6.119, p < .000, d = .513), which was not the case for the other countries (USA: t = .121, n.s.; UAE: t = 1.621, n.s.). Although unsystematic with regard to our independent variables the impression variance across the stimuli revealed a unique possibility to analyze cross-cultural correlations in the perception of the stimulus person independently from nationality and role of the actors. Further analyses therefore were based on stimuli as cases using averages of impression ratings across the observers from each culture as dependent measures. 4.4. Correlations of dominance and evaluation within and across countries To analyze within- and between-culture correlations of the impression ratings the factor scores of Dominance and Evaluation were aggregated for each stimulus person in the dyadic interactions and Pearson-correlation coefficients were calculated on this basis with N = 60 stimulus persons (USA and GER) respectively N = 40 stimulus persons (UAE). Results are shown in Table 4. Regardless of knowing the stimuli’s country of origin, observers of all three countries showed significant correlations in their dominance ratings (see also Fig. 3). Within-country correlations were even higher than cross-country correlations indicating a stable impression formation based on nonverbal cues and being immune towards advance information about the stimuli’s country of origin. The results suggest trans-cultural universals in dominance perception, which are independent from knowledge about the poser’s country of origin. For the Evaluation dimension within-country correlations were significant but except from Germany only of a medium size, indicating a potential influence of source information on the US and UAE judgments. Cross-country correlations were only significant for German and US observers and for German and UAE observers but not for US and UAE observers supporting the hypothesis that similarities in the basic cultural values might be influential with regard to the evaluation of nonverbal behavior (Freeman, Rule, & Ambady, 2009; Freeman, Rule, Adams, et al., 2009; Matsumoto, 2002). Again the correlation pattern was independent from advance information about the stimuli’s country of origin. Correlations between judgments of Dominance and Evaluation revealed further insights into cultural specificities in impression formation. Results are shown in Table 5. No significant correlations were found between cultures. However
770
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
Table 4 Pearson-correlation coefficients for dominance and evaluation between country of observers with regarding of advance information about country of stimulus. Country
Dominance Without advance info
USA
GER
UAE
With advance info
USA
GER
Evaluation Without advance info
USA
GER
UAE
With advance info
USA
GER
r p N r p N r p N
Without advance info
With advance info
GER
UAE
USA
.766** .000 60
.720** .000 40 .781** .000 40
.793** .000 60
UAE
.822** .000 60 .884** .000 40 .790** .000 60
r p N r p N
r p N r p N r p N
GER
.486** .000 60
r p N r p N
.085 .603 40 .441** .004 40
.689** .000 40 .816** .000 40
.509** .000 60 .694** .000 60 .412** .008 40 .429** .001 60
.303 .057 40 .529** .000 40
*
Significant at the .05 level (2-tailed). Significant at the .01 level (2-tailed).
**
within cultures we found a significant positive correlation between Dominance and Evaluation for the US observers, while for German observers we found a zero correlation and for Arabs we found a significant negative correlation. This result is in line with the findings of Freeman, Rule, and Ambady (2009) and Freeman, Rule, Adams, et al. (2009) showing that the expression of dominance is more favorable in a typical western culture (USA) than in a typical eastern culture, such as Japan. 4.5. Predicting the status-role assignment by dominance perception So far we found that the recognition of roles was only possible for the German stimulus sample (see also results from ANOVA), but that despite the failure to identify underlying status roles in the other countries, the Dominance ratings across all observer groups and all stimuli as well as the ascription of the status role – although wrong – correlated significantly. Consequently we asked the question whether role ascription could be predicted by Dominance ratings, independently from the actual role assigned to the actors. To determine to which degree the assignment of status-roles in the three countries was influenced by the perception of Dominance regression analyses were conducted across all observers and for each country separately with frequency of perceiving the poser as supervisor (M = 52.35, SD = 23.421) as response variable and Dominance (M = 4.034, SD = .602) as predictor variable. Regression analysis across countries explained about 65% of the total variance (r2 = .649; bstd = .806, p = .000). Separate analyses for the three countries confirmed the predictive value of dominance perception for status-role assignment explaining between 58% and 78% of the variance (USA: r2 = .583; bstd = .763, p = .000, GER: r2 = .774; bstd = .880, p = .000, UAE: r2 = .783; bstd = .885, p = .000). Fig. 3 visualizes the cross-cultural correlations in Dominance impressions and also shows that the observers from all countries most uniformly based their guesses of the underlying status-roles on this impression dimension. To determine the degree of accuracy with which the predictor variable Dominance was able to discriminate between the status-role ascriptions (supervisor or employee) we further conducted a discriminant analysis with 7331 valid individual judgments of all 570 participants The mean of Dominance (M = 4.034) as decision criterion was able to classify 70.5% of all cases correctly as mentioned supervisor or as mentioned employee: if the factor score of Dominance was greater than
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
771
Fig. 3. Scatter-plots for pair-wise cross-country correlations of Dominance ratings and status-role ascriptions (predominant role ascriptions determined via median split: j = supervisor; } = employee).
4.034, the observer identified the role as supervisor in 75.6% of all cases, and if the factor score of Dominance was less than 4.034, the observer identified the role as employee in 66.9% of all cases. Overall the results suggest that perceived dominance predicts ascription of status-roles in all cultures under investigation.
772
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777 Table 5 Pearson-correlation coefficients between dominance and evaluation. Evaluation Country Dominance
* **
USA
GER
UAE
USA
r p N
.352** .006 60
.023 .862 60
.273 .088 40
GER
r p N
.176 .179 60
.004 .978 60
.287 .073 40
UAE
r p N
.088 .589 40
.008 .960 40
.349* .027 40
Significant at the 0.05 level (2-tailed). Significant at the 0.01 level (2-tailed).
4.6. Predicting dominance and evaluation by nonverbal behavior The last step of data analysis should answer the question which nonverbal cues account for the variance in Dominance ratings and consequently the status role ascription and for the variance in the Evaluation of the actors as well. To answer this question, movement and postural parameters, described as potentially relevant to the impression dimensions under investigation were extracted from the data protocols underlying our animations (see Dovidio & Ellyson, 1982; Dovidio et al., 1988; Fehr & Exline, 1987; Hall, Coates, & LeBeau, 2005; Harrigan, Oxman, & Rosenthal, 1985; Matsumoto & Kudoh, 1987; Schwartz, Tesser, & Powell, 1982) focusing on: (1) the orientation towards the partner (head rotation as mean angular deviation from direct eye line), (2) the vertical head position (mean angular up-down shift of the head), (3) the openness (mean distance between both hands and elbows as well as between both feet and knees), (4) the expansivity of extremities (mean distance of hands and elbows from chest plus distance between feet and knees from hips) and (5) the overall movement activity (time spent in motion as percentage of observation time any movement activity was observed). As shown in Table 6 relevant proportions of variance of the Dominance ratings could be explained by the chosen nonverbal parameters (USA: 33.2%, GER: 41.4%, UAE: 44.9%). Uniformly dominance was based on the vertical head posture as well as on the opening of upper and lower extremities. For all cultures a lifted head and an open posture were perceived as more dominant. German observers as well as American observers further based their judgments on the orientation towards the partner (rotational head orientation). A more direct orientation was perceived as more dominant. Two particularities further appeared for the German observers. Less movement activity and stretching of the lower extremities was perceived as more dominant here. Overall the exploratory analysis revealed a strong influence of nonverbal cues on Dominance perception and a remarkable overlap in the use of nonverbal cues across the cultures. Table 7 summarizes the results of the regression analysis for the Evaluation dimension. Explained variance was lower than for Dominance (USA: 25.8%, GER: 22.5%, UAE: 22.4%) and the cultural overlaps with regard to the predictive value of various nonverbal cues were smaller and more specific. US observers and UAE observers did not show any overlaps, while German observers showed overlaps with regard to one parameter with USA and one with UAE, which is consistent with the intercorrelations in the Evaluation ratings reported above. In USA and GER it was essential for higher Evaluation scores that the lower extremities were not outstretched. In UAE and GER higher Evaluation scores were predicted by closure of the lower extremities. A lifted head position only predicted evaluation for the US observers, showing the same direction as for Dominance, possibly explaining the significant correlation between evaluation and dominance ratings in the US as reported above. A peculiarity appeared for the US with regard to global movement activity indicating that higher activity levels were
Table 6 Beta coefficients of analysis of regression for explaining dominance. USA R2 = .332
GER R2 = .414
UAE R2 = .449
Explanatory variable
std. Beta
t
p
std. Beta
t
p
std. Beta
t
p
Vertical head posture Rotational head orientation Opening upper extremities Opening lower extremities Expansivity up. extremities Expansivity low extremities Time spent in motion
.328 .271 .320 .190 .026 .088 .069
2.702 2.266 2.720 1.607 .217 .683 .534
.005 .014 .004 .057 .415 .249 .298
.351 .181 .368 .223 .068 .215 .259
3.087 1.614 3.340 2.014 .612 1.791 2.151
.002 .050 .001 .025 .272 .040 .018
.330 .125 .392 .340 .081 .050 .176
2.297 .908 2.800 2.446 .558 .334 1.101
.014 .185 .004 .010 .291 .370 .139
Note: All significances are 1-tailed.
773
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777 Table 7 Beta coefficients of analysis of regression for explaining evaluation. USA R2 = .258
GER R2 = .225
UAE R2 = .224
Explanatory variable
std. Beta
t
p
std. Beta
t
p
std. Beta
t
p
Vertical head posture Rotational head orientation Opening upper extremities Opening lower extremities Expansivity up. extremities Expansivity low. extremities Time spent in motion
.250 .210 .044 .068 .140 .260 .356
1.954 1.665 .354 .547 1.117 1.923 2.627
.028 .051 .362 .293 .135 .030 .006
.091 .238 .152 .260 .150 .313 .022
.699 1.848 1.200 2.043 1.172 2.269 .158
.244 .035 .118 .023 .123 .014 .438
.103 .171 .194 .365 .072 .163 .172
.603 1.043 1.169 2.214 .417 .919 .909
.275 .152 .126 .017 .340 .183 .185
Note: All significances are 1-tailed.
evaluated more positively. It is noteworthy that directed head orientation as an affiliation cue and an indicator of visual attention only accounted for positive evaluations in the German observer sample. Seemingly the interpretation of this strong social cue underlies particular cultural variance which might constitute a basis for misunderstanding and conflict in crosscultural encounters (Argyle & Cook, 1973; Ellsworth & Ludwig, 1972). 5. Discussion Applying a novel computer-animation methodology (see Bente et al., 2008), we investigated the effects of nonverbal behavior from Germany, the USA, and the UAE on status-role recognition and on the impression formation of observers in the three countries focusing on the dimensions Dominance and Evaluation. The study demonstrated the usefulness of the methodology which permitted the use of culture fair characters showing nonverbal behavior transcribed from real-life interactions and thus to avoid stereotype relevant inferences from the outward appearance of the original actors. Furthermore the animation methodology provided access to the quantification of movement patterns relevant to the impression dimensions under investigation and allowed for the analysis of differential effects of specific nonverbal cues on impression formation. Regarding the recognition of status roles we found hit rates above chance only for the German stimuli but nor for the US and UAE stimuli. No in-group advantage was found for status-role recognition in any of the countries comparable the one reported for the recognition of emotions from facial expressions (Elfenbein & Ambady, 2002a, 2003; Matsumoto, 2002; Matsumoto et al., 2009). The fact that observers were not able to correctly identify status roles in the US and the UAE sample even if observer and stimulus country matched might cast some doubt on the effectiveness of our status role inductions. In principle the unexpected distribution of perceived Dominance across the roles might be due to the nature of the role-play situation. Comments during debriefing revealed that most role players took the task quite seriously, however they also suggested another explanation. The conflict imposed on the role play interaction in fact assigned status-power to the supervisor. But in addition it also assigned expertise and potential coercive power to the employee. Thus we might have created a power equilibrium based on different power sources (legitimate vs. expert/coercive; see French & Raven, 1959) which left room for control attempts on both sides. It remains unclear whether the induction of a more pronounced power asymmetry would have led to better recognition rates. Future studies will have to address this issue, e.g. by combining status role inductions with an organizational problem-solving task instead of a power conflict. By means of EFA and CFA we were able to provide a factorial model of social impressions distinguishing the two factors Dominance and Evaluation. Separate ANOVAs for both dimensions revealed unsystematic and weak effects of the independent variables (country of stimuli, country of observer, status-role and advance information about the stimuli’s country of origin). A significant main effect of role on Dominance ratings occurred, which was largely relying on a pronounced difference between employees and supervisors in the German stimulus sample only. No main effect of advance information about the stimuli’s country of origin was found, suggesting a strong effect of observed nonverbal behavior as compared to advance information. However, overall effect sizes were very small and all group means were close to the scale means. As a major cause we assume the large inter-individual variance in the behavior of the stimulus persons across and within cultures, which suppressed the influence of culture of origin as well as the status-role assigned to the posers. This assumption is supported by the analysis of the hit rates in the role recognition task and might also have been caused by the specific setting. The fact, however, that we found status-role recognition above chance for the German stimuli as well as an interaction effect between status-role and stimulus country for the Dominance ratings, pointing in the same direction, might also suggest another explanation referring to actual cultural differences in the expression of power status and dominance. Although described as a high power distance (PDI) culture the collectivistic orientation in the UAE could inhibit the open expression of status differences through nonverbal behavior (see Ting-Toomey & Oetzel, 2001, 2002). Americans on the other hand, though characterized by an individualistic (IDV) orientation, could be expected, because of low power distance, to emphasize equality. Two different reasons thus could account for the same effect, namely the lack of significant differences in nonverbal dominance expression in status–asymmetric interactions. PDI and IDV however would not explain then the differences in
774
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
perceived dominance between employees and supervisors in the German sample. It can be speculated whether the third dimension ‘‘uncertainty avoidance” (UAI) where Germany scores are approximately as high as UAE can explain this phenomenon. Not being inhibited by a collectivistic attitude, as the UAE people are, we could suppose that the indication of status roles through nonverbal behavior would be more likely then in the German culture. Overall this data makes a strong argument to avoid an isolated analysis of singular cultural value dimensions when it comes to the prediction of concrete behavior. More studies in fact are needed to disentangle the complex interplay between the various cultural value dimensions and social behavior in the future. Despite the still unclear influence of culture and status role on the expression of nonverbal dominance cues, the behavioral variance across the stimulus persons provided a solid basis for the analyses of cultural specificities and communalities on the perceptual level. In fact, our analysis based on the factor values revealed significant positive correlations for Dominance ratings between all three countries, supporting the hypothesis of cultural universalities in the perception of Dominance from nonverbal behavior. This result is in line with data reported by Keating et al. (1981), which revealed universals in Dominance perception from facial expressions and with those of Kowner and Wiseman (2003) who could show that nonverbal status cues could be identified as such even if not part of the own cultural repertoire. With regard to the Evaluation dimension of person perception we only found a significant correlation between German and US observers and between German and UAE observers but not between US and UAE observers. This result supports the hypothesis that culture has a more pronounced influence on Evaluation than on Dominance perception and that similarity in values breeds similarities in Evaluation (Swift, 1999). A similarity between Germany and the US has been found for PDI, while with regard to IDV-scores Germany lies between the two other cultures and with regard to UAI even is more similar to the UAE, thus positioning Germany in-between USA and UAE, when taking into account all three dimensions (see Clearly Cultural, 2009). This view is in fact consistent with the correlations found and holds that cultural differences can hardly be explained through a singular cultural value dimension, when it comes to understanding self-construal, other perception and communication in real-life encounters (Hofstede, 2001; Triandis & Gelfand, 1998). As for Dominance prior information about the stimuli’s country of origin did not affect the pattern of inter-correlations in Evaluation. Exploratory analysis of nonverbal cues relevant to Dominance perception and Evaluation revealed tentative explanations for the correlation patterns found. While we found significant variance explanation and large cross-cultural overlaps in the predictive value of postural dynamics (in particular head lift, openness of extremities and head orientation to the partner) for Dominance, regression analysis for Evaluation showed a weaker predictive value of the selected nonverbal parameters. A few overlaps were only found between Germany and the US and Germany and the UAE, which matches the correlation patterns found for the Evaluation dimension. Overall results from correlation analysis strongly support the hypothesis of cultural universalities in the perception of Dominance and of cultural specificities in the Evaluation dimension and thus aligns with previous results based on different paradigms (Freeman, Rule, & Ambady, 2009; Freeman, Rule, Adams, et al., 2009; Kowner & Wiseman, 2003). The fact that disclosure of the stimuli’s country of origin did not influence correlation patterns in either of the dimensions also underpins the strong impact of nonverbal behavior in cross-cultural person perception, which is largely independent from stereotypes or predetermined evaluations of the other cultures as possibly caused by stereotypes or hearsay. It is noteworthy that the observers from the three cultures not only converged in their Dominance judgments but also unanimously based their guesses about the underlying status-roles on these impressions. Although hit rates mostly failed to surmount the chance level, the assignments of the roles (supervisor vs. employee), whether correct or incorrect, were predictable by the Dominance ratings as shown via regression- and discriminant-analyses. Thus it can be supposed that in view of a lack of additional context information, low level inferences from nonverbal behavior to Dominance also affected higher level inferences to underlying status roles. This observation might be of particular relevance in cross-cultural business encounters, where status differences are not known or obscured. In sum the data presented here suggests a multi-layer model of person perception which holds particular explanatory value for understanding cultural diversity in social cognition. While the perception of Dominance cues seems to rely on universals with regard to the extracted cues as well as to the inferential mechanisms they invoke, Evaluation as basic criterion to approach or avoid another person seems to be more affected by cultural standards. This is in accordance with the findings of Freeman, Rule, and Ambady (2009) and Freeman, Rule, Adams, et al. (2009), who showed that viewing photographs with dominant or submissive postures resulted in differential activities in areas related to evaluative processing of social cues, when seen either by Westerners (US-Americans) or by Easterners (Japanese). In fact this result implies the ability of both groups to uniformly and adequately qualify the displayed cues as either dominant or submissive. However based on this, both groups showed distinct evaluative responses. Both our own and Freeman et al.’s data thus point to two possibly independent layers of person perception, one which is more basic and culturally invariant (Dominance) and the other which is learning dependent and culture-specific (Evaluation). Yet it can be asked whether even more independent layers are active in social impression formation. Data from our explorative factor analysis, which have not yet been modeled in CFA at this point, support the assumption that perceptions of activity on one hand and inferences about the other persons’ inner states (composure, relaxation) on the other hand might be candidates for such layers. More systematic research will be needed to disentangle the interplay of these components and to determine the particular influence of culture on each level of information processing. In this endeavor a significant contribution of cultural neuroscience can be expected, as it would allow identification of potentially distinct neural mechanisms underlying social information processing on the different levels (Chiao & Ambady, 2007; Chiao, Li, & Harada, 2008; Chiao et al., 2010; Han & Northoff, 2008). Although the work reported here
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
775
emphasized the implementation of behavioral paradigms, the experimental techniques applied possess particular value for cultural neuroscience studies as well, as they allow for unprecedented degrees of stimulus control In particular the computer animation approach can go beyond the use of verbal reports of prototypical behaviors (see Kowner & Wiseman, 2003) or static images of postures and gestures and include subtle variations of nonverbal behavior while preserving its natural dynamics. Acknowledgments The reported research has been funded by the German Science Foundation (DFG, Deutsche Forschungsgemeinschaft) within the Special Research Program ‘‘Media and Cultural Communication” (SFB/FK 427). We wish to thank Sybille Pennig and Maria Senokozlieva for their valuable support in data collection at the different institutions in Germany, the USA and the UAE, which posed a particular challenge in this cross-cultural endeavor. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.concog.2010.06.006. References Aguinis, H., Simonsen, M. M., & Pierce, C. A. (1998). Effects of nonverbal behavior on perceptions of power bases. Journal of Social Psychology, 138(4), 455–469. Andersen, P. A. (1999). Cues of culture: The basis of intercultural differences in nonverbal communication. In L. A. Samovar & R. E. Porter (Eds.), Intercultural communication. A reader (9th ed., pp. 258–269). Belmont, CA: Wadsworth Publishing Company. Andersen, P. A., & Andersen, J. F. (2005). Measurements of perceived nonverbal immediacy. In V. Manusov (Ed.), The sourcebook of nonverbal measures (pp. 113–126). Mahwah: New Jersey LEA. Andersen, J. F., Andersen, P. A., & Jensen, A. D. (1979). The measurement of nonverbal immediacy. Journal of Applied Communication Research, 7, 153–180. Argyle, M., & Cook, M. (1973). Gaze and mutual gaze. Cambridge University Press. Argyle, M., Salter, V., Nicholson, H., Williams, M., & Burgess, P. (1970). The communication of inferior and superior attitudes by verbal and non-verbal signals. British Journal of Social and Clinical Psychology, 9, 222–231. Aronson, E. (1972). The social animal. San Francisco: Freeman. 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 and Social Psychology, 71(2), 230–244. Bente, G. (1989). Facilities for the graphical computer simulation of head and body movements. Behavior Research Methods, Instruments, & Computers, 21(4), 455–462. Bente, G., Feist, A., & Elder, S. (1996). Person perception effects of computer simulated male and female head movement. Journal of Nonverbal Behavior, 20, 213–228. Bente, G., Krämer, N. C., & Petersen, A. (2002). Virtuelle Realität als Gegenstand und Methode in der Psychologie. In G. Bente, A. Petersen, & N. C. Krämer (Hrsg.) (Eds.), Virtuelle Realitäten (pp. 1–31). Göttingen: Hogrefe. Bente, G., Krämer, N. C., Petersen, A., & de Ruiter, J. P. (2001). Computer animated movement and person perception. Methodological advances in nonverbal behavior research. Journal of Nonverbal Behavior, 25(3), 151–166. Bente, G., Petersen, A., Krämer, N. C., & de Ruiter, J. P. (2001). Transcript-based computer animation of movement: Evaluating a new tool for nonverbal behavior research. Behavior Research Methods, Instruments, & Computers, 33(3), 303–310. Bente, G., Senokozlieva, M., Pennig, S., Al-Issa, A., & Fischer, O. (2008). Deciphering the secret code. A new methodology for the cross-cultural analysis of nonverbal behavior. Behavior Research Methods, 40(1), 269–277. Burgoon, J. K. (1994). Nonverbal signals. In M. L. Knapp & G. R. Miller (Eds.). Handbook of interpersonal communication (Vol. 2, pp. 344–390). Beverly Hills, CA: Sage. Burgoon, J. K., Berger, C. R., & Waldron, V. R. (2000). Mindfulness and interpersonal communication. Journal of Social Issues, 56, 105–127. Burgoon, J. K., Buller, D. B., & Woodall, W. G. (1996). Nonverbal communication: The unspoken dialogue. New York: Harper & Row. Burgoon, J. K., & Dunbar, N. E. (2000). Interpersonal dominance as a situationally, interactionally, and relationally contingent social skill. Communication Monographs, 67, 96–121. Burgoon, J. K., Johnson, M. L., & Koch, P. T. (1998). The nature and measurement of interpersonal dominance. Communication Monographs, 65, 309–335. Carli, L. L., Martin, C., Leatham, G., Lyons, K., & Tse, I. (1993). Perceptions of nonverbal behavior. Paper presented at the annual meeting of the American psychological organisation, Toronto, Canada. Chen, M., & Bargh, J. A. (1997). Nonconscious behavioral confirmation processes: The self-fulfilling nature of automatically-activated stereotypes. Journal of Experimental Social Psychology, 33, 541–560. Chiao, J. Y., & Ambady, N. (2007). Cultural neuroscience: Parsing universality and diversity across levels of analysis. In S. Kitayama & D. Cohen (Eds.), Handbook of cultural psychology (pp. 237–254). New York: Guilford Press. Chiao, J. Y., Harada, T., Komeda, H., Li, Z., Mano, Y., Saito, D., et al (2010). Dynamic cultural influences on neural representations of the self. Journal of Cognitive Neruroscience, 22(1), 1–11. Chiao, J. Y., Li, Z., & Harada, T. (2008). Cultural neuroscience of consciousness: From visual perception to self-awareness. Journal of Consciousness Studies, 10, 11. Choi, Y. S., Gray, H. M., & Ambady, N. (2005). The glimpsed world: Unintended communication and unintended perception. In R. R. Hassin, J. S. Uleman, & J. A. Bargh (Eds.), The new unconscious (pp. 309–333). New York: Oxford University Press. Clearly Cultural (2009). Making sense of cross-cultural communication.
Retrieved 07.12.09. Dovidio, J. F., & Ellyson, S. L. (1982). Decoding visual dominance – Attributions of power based on relative percentages of looking while speaking and looking while listening. Social Psychology Quarterly, 45(2), 106–113. Dovidio, J. F., Ellyson, S. L., Keating, C. F., Heltman, K., & Brown, C. E. (1988). The relationship of social power to visual displays of dominance between men and women. Journal of Personality and Social Psychology, 54, 233–242. Dunbar, N. E. (2004). Dyadic power theory: Constructing a communication-based theory of relational power. Journal of Family Communication, 4(3/4), 235–248. Dunbar, N. E., & Burgoon, J. K. (2005). Perceptions of power and interactional dominance in interpersonal relationships. Journal of Social and Personal Relationships, 22(2), 207–233. Edinger, J. A., & Patterson, M. L. (1983). Nonverbal involvement and social-control. Psychological Bulletin, 93(1), 30–56.
776
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
Elfenbein, H. A., & Ambady, N. (2002a). Is there an ingroup advantage in emotion recognition? Psychological Bulletin, 128, 243–249. Elfenbein, H. A., & Ambady, N. (2002b). On the universality and cultural specificity of emotion recognition: A meta-analysis. Psychological Bulletin, 128, 205–235. Elfenbein, H. A., & Ambady, N. (2003). When familiarity breeds accuracy: Cultural exposure and facial emotion recognition. Journal of Personality and Social Psychology, 85(2), 276–290. Elfenbein, H. A., Mandal, M. K., Ambady, N., Harizuka, S., & Kumar, S. (2002). Crosscultural patterns in emotion recognition: Highlighting design and analytical techniques. Emotion, 2, 75–84. Ellsworth, P. C., & Ludwig, L. M. (1972). Visual behavior in social interaction. Journal of Communication, 22, 375–403. Fehr, B. J., & Exline, R. V. (1987). Social visual interactions: A conceptual and literature review. In A. W. Siegman & S. Feldstein (Eds.). Nonverbal behavior and communication (Vol. 2, pp. 225–326). Hillsdale, NJ: Lawrence Erlbaum. Freeman, J. B., Rule, N. O., Adams, R. B., Jr., & Ambady, N. (2009). Culture shapes a mesolimbic response to signals of dominance and subordination that associates with behavior. NeuroImage, 47, 353–359. Freeman, J. B., Rule, N. O., & Ambady, N. (2009). The cultural neuroscience of person perception. In J. Y. Chiao (Ed.), Progress in brain research (pp. 191–201). Elsevier. French, J., & Raven, B. H. (1959). The bases of social power. In D. Cartwright (Ed.), Studies in social power (pp. 150–167). Ann Arbor, MI: Institute for Social Research. Gudykunst, W. B., & Ting-Toomey, S. (1988). Culture and affective communication. American Behavioral Scientist, 31(3), 384–400. Gudykunst, W. B., Ting-Toomey, S., & Nishida, T. (1996). Communication in personal relationships across cultures. Thousand Oaks, CA: Sage Publications, Inc. Guerrero, L. K., & Floyd, K. (2005). Nonverbal communication in close relationships. New York: Routledge. Hall, J. A., Coates, E. J., & LeBeau, L. S. (2005). Nonverbal behavior and the vertical dimension of social relations: A meta-analysis. Psychological Bulletin, 131, 898–924. Han, S., & Northoff, G. (2008). Culture-sensitive neural substrates of human cognition: A transcultural neuroimaging approach. Nature Reviews Neuroscience, 9, 646–654. Harrigan, J. A., Oxman, T. E., & Rosenthal, R. (1985). Rapport expressed through nonverbal behavior. Journal of Nonverbal Behavior, 9(2), 95–110. Hedden, T., Ketay, S., Aron, A., Markus, H. R., & Gabrieli, J. D. E. (2008). Cultural influences on neural substrates of attentional control. Psychological Science, 19, 12–17. Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Newbury Park, CA: Sage. Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions and organizations across nations. Thousand Oaks, CA: Sage. Hofstede, G. (2003). Culture’s consequences (2nd ed.). Newbury Park, CA: Sage. Keating, C. F. (1985). Gender and the physiognomy of dominance and attractiveness. Social Psychology Quarterly, 48, 61–70. Keating, C. E., Mazur, A., Segall, M. H., Cysneiros, R. G., Divale, W. T., Kilbride, J. E., et al (1981). Culture and the perception of social dominance from facial expression. Journal of Personality and Social Psychology, 40(4), 615–626. Kitayama, S., Duffy, S., Kawamura, T., & Larsen, J. T. (2003). Perceiving an object and its context in different cultures: A cultural look at New Look. Psychological Science, 14, 201–206. Komter, A. (1989). Hidden power in marriage. Gender and Society, 3, 187–216. Kowner, R., & Wiseman, R. (2003). Culture and status-related behavior: Japanese and American perceptions of interaction in asymmetric dyads. CrossCultural Research, 37(2), 178–210. Kudoh, T., & Matsumoto, D. (1985). Cross-cultural examination of the semantic dimensions of body postures. Journal of Personality and Social Psychology, 48(6), 1440–1446. Lee, M. E., Matsumoto, D., Kobayashi, M., Krupp, D., & Maniatis, E. F. (1992). Cultural influences on nonverbal behavior in applied settings. In R. S. Feldman (Ed.), Applications of nonverbal behavioral theories and research (pp. 239–261). Hillsdale, NJ: Lawrence Erlbaum Associates. Markus, H., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224–253. Matsumoto, D. (2002). Methodological requirements to test a possible in-group advantage in judging emotions across cultures: Comment on Elfenbein and Ambady (2002) and Evidence. Psychological Bulletin, 128, 236–242. Matsumoto, D. (2003). Cross-cultural research. In S. Davis (Ed.), The handbook of research methods in experimental psychology (pp. 189–208). Oxford, UK: Blackwell. Matsumoto, D. (2006). Culture and nonverbal behavior. In V. Manusov & M. Patterson (Eds.), Handbook of nonverbal communication (pp. 219–235). Thousand Oaks, CA: Sage. Matsumoto, D., & Kudoh, T. (1987). Cultural similarities and differences in the semantic dimensions of body postures. Journal of Nonverbal Behavior, 11(3), 166–179. Matsumoto, D., Olide, A., & Willingham, B. (2009). Is there an ingroup advantage in recognizing spontaneously expressed emotions? Journal of Nonverbal Behavior, 33(3), 181–191. McCroskey, J. C., & Jenson, T. A. (1975). Image of mass-media news sources. Journal of Broadcasting, 19(2), 169–180. Mehrabian, A. (1969). Significance of posture and position in the communication of attitude and status relationships. Psychological Bulletin, 71, 359–372. Mehrabian, A. (1970). A semantic space for nonverbal behavior. Journal of Consulting and Clinical Psychology, 35(2), 248–257. Meredith, W. (1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58, 525–543. Mignault, A., & Chaudhuri, A. (2003). The many faces of a neutral face: Head tilt and perception of dominance and emotion. Journal of Nonverbal Behavior, 27(2), 111–132. Naylor, M. (2009). Resolving intercultural tensions. Cross-cultural impact in the 21st century.
Retrieved 12.11.09. Nettle, D. (2009). Beyond nature versus culture: Cultural variation as an evolved characteristic. Journal of the Royal Anthropological Institute, 15, 223–240. Newman, L. S., & Uleman, J. S. (1989). Spontaneous trait inference. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 155–188). New York: Guilford. Nisbett, R. E., Peng, K., Choi, I., & Norenzayan, A. (2001). Culture and systems of thought: Holistic vs. analytic cognition. Psychological Review, 108, 291–310. Osgood, C. E. (1966). Dimensionality of the semantic space for communication via facial expressions. Scandinavian Journal of Psychology, 7, 1–30. Remland, M. (1982). The implicit ad hominem fallacy: Nonverbal displays of status in argumentative discourse. Journal of the American Forensics Association, 19, 79–86. Rohner, R. P. (1984). Toward a conception of culture for cross-cultural psychology. Journal of Cross-Cultural Psychology, 15, 111–138. Rollins, B. C., & Bahr, S. J. (1976). A theory of power relationships in marriage. Journal of Marriage and the Family, 38, 619–627. Schwartz, B., Tesser, A., & Powell, E. (1982). Dominance cues in nonverbal behavior. Social Psychology Quarterly, 45(2), 114–120. Sussman, N. M., & Rosenfeld, H. M. (1982). Influence of culture, language and sex on conversational distance. Journal of Personality and Social Psychology, 106, 215–225. Swift, J. S. (1999). Cultural closeness as a facet of cultural affinity: A contribution to the theory of psychic distance. International Marketing Review, 16(3), 182–201. Ting-Toomey, S., & Oetzel, J. (2001). Managing intercultural conflict effectively. Thousand Oaks, CA: Sage. Ting-Toomey, S., & Oetzel, J. (2002). Cross-cultural face concerns and conflict styles: Current status and future directions. In W. Gudykunst & B. Mody (Eds.), Handbook of international and intercultural communication (2nd ed., pp. 143–163). Thousand Oaks, CA: Sage. Tomasello, M. (1999). The cultural origins of human cognition. Cambridge: Harvard University Press. Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005). Understanding and sharing intentions: The origins of cultural cognition. Behavior and Brain Science, 28, 675–735.
G. Bente et al. / Consciousness and Cognition 19 (2010) 762–777
777
Triandis, H. (1995). Individualism and collectivism. New directions in social psychology. Boulder: Westview Press. Triandis, H. C., Bontempo, R., Villareal, M. J., Asai, M., & Lucca, N. (1998). Individualism and collectivism: Cross-cultural perspectives on self-ingroup relationships. Journal of Personality and Social Psychology, 54(2), 323–338. Triandis, H. C., & Gelfand, M. J. (1998). Converging measurements of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology, 74, 118–128. Uleman, J. S., & Bargh, J. A. (Eds.). (1989). Unintended thought. New York: Guilford. Triandis, H. C., & Suh, E. M. (2002). Cultural influences on personality. Annual Review of Psychology, 53, 133–160. Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3(1), 4–69. Vogeley, K., & Roepstorff, A. (2009). Contextualising culture and social cognition. Trends in Cognitive Science, 13(12), 511–516. Wagner, H. L. (1993). On measuring performance in category judgment studies of nonverbal behavior. Journal of Nonverbal Behavior, 17, 3–28. Wagner, H. L. (1997). Methods for the study of facial behavior. In J. A. Fernandez-Dols & J. M. Russell (Eds.), The psychology of facial expression (pp. 31–53). Cambridge: Cambridge University Press. White, K., Lehman, D. R., & Cohen, D. (2006). Culture, self-construal, and affective reactions to successful and unsuccessful others. Journal of Experimental Social Psychology, 42(5), 582–592. Wu, A. D., Li, Z., & Zumbo, B. D. (2007). Decoding the meaning of factorial invariance and updating the practice of multi-group confirmatory factor analysis. Practical Assessment Research & Evaluation, 12(3), 1–26. Zebrowitz-McArthur, L. (1988). Person perception in cross-cultural perspective. In M. H. Bond (Ed.), The cross-cultural challenge to social psychology (pp. 245–265). Newbury Park, CA: Sage.
Consciousness and Cognition 19 (2010) 778–801
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Psychogenic amnesia – A malady of the constricted self q Angelica Staniloiu a,*, Hans J. Markowitsch a,b, Matthias Brand c,d a
Physiological Psychology, University of Bielefeld, Bielefeld, Germany Alfried Krupp Institute for Advanced Study, Greifswald, Greifswald, Germany General Psychology: Cognition, University of Duisburg-Essen, Germany d Erwin L. Hahn Institute for Magnetic Resonance Imaging, Essen, Germany b c
a r t i c l e
i n f o
Article history: Available online 23 July 2010 Keywords: Dissociative amnesia Inferolateral prefrontal cortex Self-consciousness Emotion Autobiographical–episodic memory (AEM)
a b s t r a c t Autobiographical–episodic memory is the conjunction of subjective time, autonoetic consciousness and the experiencing self. Understanding the neural correlates of autobiographical–episodic memory might therefore be essential for shedding light on the neurobiology underlying the experience of being an autonoetic self. In this contribution we illustrate the intimate relationship between autobiographical–episodic memory and self by reviewing the clinical and neuropsychological features and brain functional imaging correlates of psychogenic amnesia – a condition that is usually characterized by severely impaired retrograde memory functioning, in absence of structural brain damage as detected by standard imaging. We demonstrate that in this disorder the autobiographical–episodic memory deficits do not exist in isolation, but occur with impairments of the autonoetic self-consciousness, emotional processing, and theory of mind or executive functions. Furthermore functional and metabolic brain alterations involving regions that are agreed upon to exert crucial roles in memory processes were frequently found to accompany the psychogenic memory ‘‘loss”. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction ‘‘Memory connects innumerable single phenomena into a whole, and just as the body would be scattered like dust in countless atoms if the attraction of matter did not hold it together so consciousness – without the connecting power of memory – would fall apart in as many fragments as it contains moments.” (Ewald Hering, 1870/1895, p. 12). ‘‘Memory requires more than mere dating of a fact in the past. It must be dated in my past. In other words, I must think that I directly experienced its occurrence.” (William James, 1890/1950, p. 650). Autobiographical–episodic memory (AEM) has a close connection to self and self understanding, which is concisely captured by Tulving’s (2005) last definition of AEM. According to this, AEM could be viewed as a threefold cord that results from the uniting of subjective time, autonoetic consciousness and the experiencing self (Tulving, 2005). As William James’ quotation suggests, the appearance of AEM on the one hand, reflects and occurs in the context of the attainment of a new stage of self understanding and self awareness (Nelson, 2003). On the other hand, the emergence of AEM supports further self development and the capacity – that is crucial to at least highly individualized societies – for maintaining a consistent feeling of identity and a coherent awareness of self‘s continuity over time (Nelson, 2003, 2005; Newen & Schlicht, 2009; Vogeley & Kupke, 2007; Welzer & Markowitsch, 2005).
q
This article is part of a special issue of this journal on Self, Other and Memory. * Corresponding author. Address: Physiological Psychology, University of Bielefeld, P.O.B. 100131, D-33501 Bielefeld, Germany. E-mail address: [email protected] (A. Staniloiu).
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.024
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
779
In the field of clinical psychiatry and psychoanalysis the role of AEM for one’s sense of self coherence and ‘‘trans-temporal sameness” (Metzinger, 2008), well-being or capacity to reconstruct oneself has been emphasized since Freud and ‘‘beyond”, despite differences in the working models of the self, which ranged from a vertically-organized self, with areas of conflict ‘‘buried” by repression to a horizontally aligned and sequenced view of self, with territories of incompatibility separated by dissociation (S.A. Mitchell & Black, 1995). Patients with psychogenic amnesia have been described for more than a century, providing evidence for the significant influence of social environment on AEM (cf. Markowitsch, 1992). In this review, after defining psychogenic amnesia, we outline the memory systems and the main brain structures that are involved in AEM and the processes necessary for building an episodic memory. We then illustrate the ways in and by which stress-related and psychological factors could lead to both a severe compromise of AEM and disturbances of several functions underlying the experience of being a self. 2. What and where (in the international classifications of diseases) is psychogenic amnesia? Borrowed from Greek (Pearce, 2007) the word amnesia describes the most severe form of memory impairment and refers to an inability to learn new information or recall previously learned information. The term is nowadays used to refer to a symptom of a disorder, a syndrome or a specific disease. In the latter case, its employment is congruent with the traditional view of amnesia as a memory impairment that occurs in an alert, responsive person in the absence of (or out of proportion in comparison to) other significant cognitive impairments, being subsequently restricted to specific disorders. These disorders are named amnesic (ICD-10, 1992) or amnestic (DSM-IV-TR, 2000) disorders. They have as a core feature a memory impairment that is not due to dementia, has functional relevance and represents a decline from a previously attained level of function. Based on their preponderant etiological link amnes(t)ic disorders are categorized as being due to a general medical condition (e.g. neurological event), direct effects of substances (e.g. alcohol) persisting beyond the period of intoxication or withdrawal or psychological factors. The term psychogenic amnesia has traditionally been used to describe episodes of retrograde and/or anterograde (AEM) memory loss, which are precipitated by psychological stresses and occur in the absence of identifiable brain damage. The memory impairment in psychogenic amnesia has classically been regarded to be reversible (Loewenstein, 1991), with older studies reporting a high percentage of recovery within a month from the onset of amnesia (Kanzer, 1939). Newer studies that employed extensive neuropsychological testing have however identified a prolonged course of the memory impairments in a substantial number of patients with psychogenic memory loss (Kritchevsky, Chang, & Squire, 2004; Markowitsch, Thiel, Kessler, von Stockhausen, & Heiss, 1997). Apart from psychogenic, other terminologies (such as hysterical, dissociative, functional or medically unexplained or mnestic block syndrome) have over the years been employed to capture the category of amnesic disorders without direct evidence of significant brain damage on conventional structural imaging techniques. In the current main official classifications of diseases (DSM-IV-TR, 2000; ICD-10, 1992) earlier diagnostic designations of hysterical or psychogenic amnesia are now preponderantly subsumed under the diagnostic categories of dissociative disorders in DSM-IV-TR and dissociative (conversion) disorders in ICD-10, but also under other diagnostic subcategories such as somatization disorder (in DSM-IV-TR only), post-traumatic stress disorder and acute stress disorder (DSM-IV-TR and ICD-10). Despite its gradual vanishing from the international nomenclatures of diseases, the terminology psychogenic amnesia has however survived in the scientific literature for a number of reasons. As opposed to the current ICD-10 and DSM-IV-TR diagnostic subcategory of dissociative amnesia, the term psychogenic amnesia is more comprehensive, being employed to encompass more than one diagnostic entity (see below). In addition, in contrast to the designation dissociative amnesia, which carries a specific theoretical load, the term psychogenic amnesia conveys the preponderant etiological link of the amnesic condition to psychological factors, without making a priori assumptions about the nature of the psychological mechanisms involved (McKay & Kopelman, 2009).This is of importance given that not only dissociation (defined by Janet [1907, p. 23] as ‘‘an inability of the personal self to bind together the various mental components in an integrated whole under its control”), but several other psychological mechanisms, such as hyper-suppression or cognitive avoidance (Fujiwara et al., 2008; Lemogne et al., 2009; Tramoni et al., 2009) might concurrently contribute to the psychogenic memory loss. While the term amnesia was used rather loosely until the 1980s, being equalized with ‘global amnesic syndrome’, studies of patients with amnesic disorders of neurological or psychogenic nature have provided evidence that the memory impairment is more likely to involve certain kinds of memory (in particular the memory for personal events). Therefore any employment of the term amnesia should nowadays be accompanied by a clear description of the type and severity of the memory problems involved, which presupposes an understanding of the current main classifications of the memory systems and memory processes. 3. Autobiographical–episodic memory system and its place within memory systems Memory is not unitary, but is divided along a chronological and content axis, respectively. Along the time axis, memory is classified in short-term and long-term memory. The term short-term memory has been employed to describe the online holding of information such as telephone numbers. It has a limited capacity of a few bits (4–7) (G.A. Cowan, 2001; Miller, 1956) and encompasses a time range of seconds to minutes. Any information that is not lost and exceeds the limited capacity
780
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
of short-term and working memory is assigned to long-term memory stores. The time-related dichotomy of memory was further refined by the introduction of the term ‘working memory’ by Alan Baddeley (Baddeley, 2000; Baddeley & Hitch, 1974). As captured by its name, working memory refers to working with memory – this involves not only time-limited online holding of new information, but also retrieving portions of old, already stored information. Further time-related categorizations of memories involve distinctions between old and new and anterograde and retrograde memories, respectively (Fig. 1). The compromised ability to access information that happened before the memoryimpairing incident corresponds to retrograde memory impairment, while anterograde memory impairment refers to the compromised capacity to long-term acquire new information after the incident. Anterograde amnesia can occur with minimal or no retrograde amnesia. Though traditional accounts of amnesia had led to the assumption that retrograde amnesia should be always accompanied by anterograde amnesia, cases of isolated retrograde amnesia have been reported after either neurological or psychological incidents. In some cases of isolated retrograde amnesia following a neurological insult (e.g. severe traumatic brain injury), significant anterograde amnesia was present initially, but then resolved or became subtle (Levine, Svoboda, Turner, Mandic, & Mackey, 2009; Levine et al., 1998). The etiological mechanisms involved in isolated or disproportionate retrograde amnesia for personal events are still a source of debate, with some authors arguing that, even in the cases with clear evidence of brain pathology psychological factors might make substantial contribution to the presence of residual, disproportionate retrograde AEM loss (Kopelman, 2000, 2008). The categorization of memory along the content dimension has evolved based on data from patients with different types of memory impairments, corresponding to different types of brain lesions, neuroimaging studies of patients with memory impairments or normal subjects, animal memory research and human developmental studies. In the 1970s and 1980s a revival of early conceptualizations of memory subdivisions (Schneider, 1912, 1928; Semon, 1904) was proposed in the field of human memory research by Tulving (1972, 1983) and in the arena of animal research by Mortimer Mishkin (Mishkin & Petri, 1984). Mishkin distinguished between ‘habit’ and a ‘memory’ system; the habit system referred to procedures and routines, whereas the memory system was concerned with the acquisition of facts and relations between objects. Tulving initially differentiated between semantic and episodic memory, implying that semantic memory refers to general knowledge, while episodic memory refers to single episodes with a specific embedding in time and locus. Later, Tulving and other researchers expanded the categorization of memory systems, in particular by adding those systems involving automatic, implicit and subconscious levels of processing – such as procedural memory and the priming system. Procedural memory refers to highly automated sensory-motor skills, which include complex motor acts like driving a car, riding a bike, or skiing. Priming describes the higher probability of (‘‘automatically”) choosing a stimulus which was perceived earlier in the same or a similar manner. The ‘perceptual memory system’ was identified as a legitimate distinct system relatively recently. In contrast to the priming and procedural memory systems, this system acts ‘consciously’, but on a presemantic level and relies on familiarity judgments. An example is the conscious identification of an apple without hesitance, no matter what color it has or whether it is already half eaten or not. Patients with semantic dementia, who lose the capabilities for language and semantic memory, may still be able to distinguish for example an apple from a peach or pear without the need to access semantic information, by accessing perceptual representations of information via the perceptual memory system. Fig. 2 outlines these memory systems, accordingly to their assumed (Tulving, 2005) ontogenetic and phylogenetic hierarchy (starting with the simplest memory systems) and points to differences in their neuroanatomical substrates. The conceptualizations of episodic memory have undergone several revisions over the years (Tulving, 1972, 1983, 1985, 1995, 2000, 2002, 2005). While a few decades ago the term episodic could be applied to describe memory for laboratory stimuli with a specific embedding in time and place (Tulving, 1972), currently the episodic memory system is viewed as equivalent to the AEM system and AEM is defined as the conjunction of subjective time, autonoetic consciousness and the experiencing self (Tulving, 2005). One characteristic of AEM is mental time traveling through subjective time from
Fig. 1. Relations between anterograde and retrograde amnesia. The flash symbol represents the time point of a brain infarct or of a major psychic traumatic event, leading to either anterograde or retrograde amnesia or to both. Note that for retrograde amnesia the frequently observed gradient – termed Ribot’s law (cf. Markowitsch, 2008), is indicated by stating that usually very old memories are preserved in retrograde amnesia, while those close to the point of the event are impaired.
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
781
Fig. 2. The five long-term memory systems and their assumed brain bases (for further description see the text).
present to both past and future. Piolino, Desgranges, and Eustache (2009) suggest that mental time traveling is the last feature of AEM that becomes fully functional, but the first feature that is affected in most amnesic conditions. In comparison to semantic memory that is accompanied by noetic consciousness only, AEM requires a higher level of self-consciousness – autonoetic consciousness (Markowitsch, 2003). The terms autonoetic [‘‘self-conscious”, or ‘‘self-aware”], noetic [‘‘aware”] and anoetic [‘‘not-aware”] were introduced and elaborated on by Tulving (1995). Autonoetic consciousness was defined as ‘‘capacity that allows adult humans to mentally represent and to become aware of their protracted existence across subjective time” (Wheeler, Stuss, & Tulving, 1997, p. 335). Although the designations autobiographical and episodic are sometimes used interchangeably, not all autobiographical memories are, however episodic. A distinction should therefore be made between autobiographical–episodic and autobiographical–semantic memory. The latter refers to personal knowledge, such as one’s name or date of birth or trait self-knowledge, and might be preserved, relearned or updated despite blocked access to AEM and even in the presence of certain impairments of semantic knowledge of impersonal facts (Klein & Gangi, 2010; Markowitsch, Calabrese, Neufeld, Gehlen, & Durwen, 1999). This latter observation led to hypothesizing dissociations within the semantic memory system (Klein & Gangi, 2010). A significant amount of general evaluative self-knowledge is available from autobiographical–semantic memory, especially in adulthood (Pfeifer, Dapretto, & Lieberman, 2010), while during childhood and adolescence self-evaluations seem to have a stronger connection to AEM (Newen et al., 2009; Oddo et al., 2010; Pfeifer et al., 2010). This may explain why patients who suffer AEM impairments, but continue to have an intact autobiographical–semantic memory might be able to maintain some aspects of personal identity (Rathbone, Moulin, & Conway, 2009). Although both AEM and autobiographical–semantic memory are involved in sustaining personal identity, the maintenance of an enduring personal identity in a frequently changing and challenging environment most likely requires periodical and persuasive update and confirmation, which draw on emotionally-laden episodic memories of personal experience. Among memory systems, the AEM system develops the latest ontogenetically (Nelson & Fivush, 2004; Piolino et al., 2007; Tulving, 2005) and is arguably uniquely human (Clayton & Dickinson, 1998; Tulving, 2005). In comparison to other memory systems, AEM is more vulnerable to neuronal alterations (Markowitsch, 2008; Tulving, 2005). AEM system is also susceptible to distortions, misinformation and dynamic reshaping (Loftus, 2005). Already in 1896 Freud noted that ‘‘the material present in the form of memory traces [is] subjected from time to time to a rearrangement” (Masson, 1985, p. 207). Furthermore, Freud (1901) remarked that there is in general no guarantee for the correctness of our memory; nevertheless we much more frequently than is justified assume that we can trust its information. The recent research on false memory syndromes supports Freud’s early insight (Loftus, 2005). In a combined behavioral and neuroimaging study it was found that university students made roughly 45% errors when judging whether a shown scene had or had not been included in two short movies (Kühnel, Woermann, Mertens, & Markowitsch, 2008). These results were reflected in the same study by the patterns of functional brain activations (Fig. 3): It was found that correct identifications of previously perceived scenes resulted in medial prefrontal activations, while incorrect identifications led to bilateral activations in the visual association cortex and precuneus. It can be speculated that the medial prefrontal activations reflect processes of monitoring and confidence, while the posterior activations reflect processes of trying to match internally generated images or imaginations with those seen in the outer world.
782
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
Fig. 3. Horizontal sections through the human brain showing voxel-based bilateral activations (white regions) in the regions of the visual association cortex and the precuneus (left section) for falsely recognized pictures and in the medial prefrontal and anterior cingulate cortex (right section) for correctly identified pictures in the study of Kühnel et al. (2008).
The research on false memories has implications for psychogenic amnesia, for example with regards to the legal aspect of hypnosis recall. As we will elaborate below, some cases of psychogenic amnesia might occur in relationship to legal difficulties or a forensic background. Though hypnosis may facilitate access to dissociated memories in certain psychogenic amnesic conditions, there is some evidence suggesting a higher likelihood of confident errors accompanying hypnosis recall (Maldonado & Spiegel, 2008). 4. Autobiographical–episodic memory and the brain The processing of memory has been a source of debate since the beginning of brain research. A waxing and waning interest in theories that proposed a strict mosaic-like localization (Markowitsch, 1994) within specific brain structures alternated with a similarly fluctuating interest in theories which promoted a widespread, Gestalt-like representation within the brain (Markowitsch, 1992). Nowadays, a compromise has generated a theoretical approach to memory processing that integrates both views and reflects the neuroscientific stance that both specialization and integration characterize the human brain (Paus, 2010; Pessoa, 2008). It is assumed that information enters the brain via the sensory organs and is then further processed depending on the kind of information and the process selected or triggered. Somewhat simplified, this means that subconsciously processed information preponderantly engages unimodal neocortical structures (priming), or – for procedural learning – the basal ganglia, premotor and other motor-related areas. Consciously processed information recruits more widespread networks, which are still largely neocortical for perceptual learning, but include limbic regions for the other two memory systems – the semantic and AEM systems. These two memory systems require the activation of limbic structures where its biological and social relevance is extracted and where the information is compared with already existing, related memories and later bound to and integrated with these. This process is also named synchronization. Further consolidation occurs during sleep (Stickgold & Walker, 2005). The link between sleep and memory has possible relevance for psychogenic amnesia. Loewenstein (1991) for example remarked that dissociated memories in psychogenic amnesia can often reveal their presence indirectly in nightmares. Kritchevsky et al. (2004) also described the case of man with severe retrograde functional amnesia, who during recovery of AEM began to experience nightmares with content related to past personal traumatic events. Though a connection between REM sleep and consolidation of emotional memory has been put forth by several authors, it has not unequivocally been substantiated empirically (for a review see Peigneux, Schmitz, & Urbain, 2010). Storage of facts and events is largely a matter of the cerebral cortex, though it has to be emphasized that storage is never final, as new information or the retrieval of already existing one leads to re-consolidation and new storage in the context of the last re-consolidation (Fulton, 2006; Haist, 2001; C.A. Miller & Sweatt, 2006; Wood, 2003). Retrieving facts and events requires an engagement of at least three closely interacting networks, namely activating brainstem structures comprising portions of the reticular activating system, a neocortical network containing the main information of the respective fact or event, plus a limbic network providing the emotional tagging of events or episodes (cf. Fig. 4.29 in Markowitsch & Welzer, 2010). Importantly, while encoding is based on a hierarchical arrangement of memory systems from procedural to AEM, retrieval allows independence in that way that no matter how the information was encoded, it can be retrieved in any memory system. Tulving (1995) suggested this flexibility and termed it the SPI-model, SPI referring to SERIAL encoding, PARALLEL processing and INDEPENDENT retrieval. For example, a patient with retrograde amnesia no longer knew or remembered that he had possessed a precious collection of antique clocks, but was able to manipulate the fine and complicated mechanical components within them without hesitance. Consequently, while he did not consciously remember his skills (serially encoded), he could retrieve them in an automatic way (parallel retrieval of procedural memories) (Markowitsch, Calabrese, Haupts, et al., 1993). While the main approach of the last century memory research consisted of a careful collection of data from neurological patients in order to establish relations between brain regions and memory processing, this has recently been complemented
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
783
and in part even replaced by functional imaging investigations in control subjects. Nevertheless, there are numerous examples of painstaking analyses in single patients demonstrating that regions of the limbic system are of crucial importance for encoding autobiographical memories (Fig. 4). The limbic system is seen as an interface between neocortical and brain stem structures (Fig. 4A). It contains important circuits and regions implicated in emotion and cognition (Fig. 4B). Within it, the hippocampal formation has received by far the most attention, due to the many influential papers on the amnesic patient H.M. of the Canadian neuropsychologist Brenda Milner. H.M. was a young man who due to pharmacologically resistant epileptic attacks underwent in 1953 bilateral surgery of his medial temporal lobes (Fig. 5). After surgery he became severely anterogradely and in part retrogradely amnesic (cf. Fig. 1) and remained so over many decades until his death in 2008. Many scientists concluded that the region within the medial temporal lobes, responsible for the amnesic condition with respect to AEM, was the hippocampus proper (e.g. Vargha-Khadem et al., 1997). Two other regional complexes within the limbic system have been regarded to be important ‘‘bottleneck” structures (Brand & Markowitsch, 2003) for AEM encoding. One is the medial and anterior diencephalon, the second is the amygdaloid body. Bilateral damage to the medial – and to a somewhat lesser degree also to anterior – diencephalic structures (anterior and mediodorsal thalamic nuclei, paratenial and midline nuclei) regularly leads to severe anterograde amnesia (Markowitsch, 1988; Vann, 2009). Given that this region contains a number of fibers (mammillothalamic tract, internal medullary lamina),
Fig. 4. (A) Schematic sagittal section through a mammalian brain, illustrating the limbic system in two sagittal sections. The top schematic section reflects the frequently expressed idea of a ‘triune brain’ in which the limbic system works as an interface between the neocortex (controlling intellectual functions) and the brainstem (controlling basic physiological functions). (B) Schematic sagittal section through the human cerebrum, showing the main structures of the expanded limbic system.
784
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
Fig. 5. Sketch of the extent of surgical removal in patient H.M. Top: View of H.M.’s cerebrum from below. The extent of surgical removal of the temporal lobes is indicated for one hemisphere only. The marked levels a–d refer to the bottom coronal sections which illustrate the removed medial temporal regions. Again, for comparison, the opposite hemisphere is presented intact, while in fact it had the same extent of removal.
some diencephalic amnesic conditions may reflect a ‘‘disconnection syndrome” due to disruption of interactions between distant brain structures. The case of a former medical professor with above-average intelligence, illustrates this configuration of damage (Markowitsch, von Cramon, & Schuri, 1993). After a diencephalic infarct, this patient became totally anterogradely amnesic while his retrograde memory was partially preserved. His anterograde amnesia, which was also accompanied by a significant impairment of the ability to reflect on his condition, was so severe that his memory for new events lasted for seconds only. The other region, the amygdaloid body, is rarely damaged exclusively and bilaterally. An exception is the Urbach–Wiethe disease, a genetic condition that can lead to a selective calcification of both amygdalae (Markowitsch et al., 1994). As a consequence of bilateral damage of amygdala-structures, which are involved in the appraisal of incoming stimuli according to their biological and social significance, patients with this kind of damage may suffer problems with the processing of emotionally-laden AEMs (Cahill, Babinsky, Markowitsch, & McGaugh, 1995; Koenigs et al., 2008; Siebert, Markowitsch, & Bartel, 2003). As outlined above, the structures of the limbic system are engaged in a complementary, but closely interwoven manner in the acquisition of AEM information, while neocortical and in part limbic areas represent the major storage places. The
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
785
combined activation of right-hemispheric fronto-temporal regions serves as trigger stations for retrieving stored AEM events (Brand & Markowitsch, 2008; Fink et al., 1996; Kroll, Markowitsch, Knight, & von Cramon, 1997; LaBar & Cabeza, 2006). The corresponding regional complex in the left hemisphere seems to trigger semantic old memories (Markowitsch, Calabrese, et al., 1999). The fiber system, interconnecting the fronto-temporal regions is the uncinate fascicle (UF). UF has a temporal, frontal and insular part (Ebeling & von Cramon, 1992) and was ascribed functions in memory and emotional processing (Markowitsch, 1995). The ventromedial portion of UF primarily connects the amygdala and uncus with the gyrus rectus and the subcallosal area (Ebeling & von Cramon, 1992). According to a histopathological study, the right UF contains 33% more fibers and is 27% larger than the left one (Highley, Walker, Esiri, Crow, & Harrison, 2002). The ventral portion of the right UF has been implicated in the retrieval of episodic–autobiographical memories, in particular in ecphorizing affect-laden personal events (Fink et al., 1996; Levine et al., 1998; Markowitsch, 1995). According to some authors, UF may be involved in the formation of memory as well (Sepulcre et al., 2008). Levine et al. (1998, 2009) reported a case of isolated dense retrograde autobiographical–episodic amnesia covering the entire life, which occurred after a severe traumatic brain injury and was associated with a focal lesion of the frontal portion of the right UF. Despite normal performance on standard anterograde memory tests, the patient reported a feeling of disconnection from the post-accident autobiographical events. Subsequent refined testing of his anterograde AEM revealed that he assigned significantly less ‘‘remember‘‘ – ratings to his post-injury autobiographical events in comparison to normal subjects. He also generated less event-specific details from anterograde AEM than normal subjects, but this was not statistically significant. The lack of first person autonoetic connection with past personal events was also described in some psychopathological conditions accompanied by AEM impairments (Lemogne et al., 2006). One of the advanced models for the amnesia in dissociative identity disorder (DID) hypothesizes that the autobiographical traumatic information can in fact be accessed, but is discarded, because it is not perceived as belonging to the person’s autobiographical experience (Dorahy & Huntjens, 2007). As opposed to other brain fiber connections (such as fornix or parts of corpus callosum), UF matures later and more slowly and may continue its development beyond 30 years (Lebel, Walker, Leemans, Phillips, & Beaulieu, 2008). This may allow a higher structural plasticity in relationship to a variety of environmental influences. Alterations of the UF and connections between UF structural integrity and memory performance have been described in several psychiatric and neurological conditions (Diehl et al., 2008; Sepulcre et al., 2008; Staniloiu & Markowitsch, 2010; Yasmin et al., 2008). Microstructural changes of the UF were also identified in children, who were reared in a deprived, neglectful environment (Govindan, Behen, Helder, Makki, & Chugani, 2010). Interestingly, in children with a history of abuse or neglect, a reduced AEM specificity (overgeneral memory effect) was recently described by Valentino, Toth, and Cicchetti (2009). 5. Clinical aspects of psychogenic memory loss As the description of Ewald Hering implies AEM binds and integrates personal events and emotion with an autonoetic self. One view of retrograde ‘‘organic” AEM recollection impairments is that they may be accounted for by both loss of information and deficits in binding and reassembling details of the past (Rosenbaum, Gilboa, Levine, Winocur, & Moscovitch, 2009). Being the most advanced acquisition ontogenetically (Tulving, 2005), the AEM system displays a higher vulnerability to insults caused by various factors (including physical injuries and stress) in comparison to other memory systems (Markowitsch, 2008; Tulving, 2005). Stressful events can lead to disturbances of the integrated organization of memory, perception, consciousness and identity, causing so called dissociative disorders. Among Dissociative Disorders in DSM-IV-TR (2000), dissociative amnesia has as the central symptom the inability to recall important personal information usually of a traumatic nature (Brandt & van Gorp, 2006; Kopelman, 2000). The disturbance is precipitated by stressful experiences or psychological trauma and is not better explained by other psychiatric or medical conditions. The symptoms of dissociative amnesia cause significant impairment of functioning or distress. The degree of experienced distress depends on many variables, including the cultural views of dissociative experiences, selfhood and past (Seligman & Kirmayer, 2008). Dissociative amnesia is included in the psychological literature under the group of psychogenic amnesic disorders, which have as the main feature the preponderant contribution of psychological factors to their emergence. Psychogenic amnesic disturbances characterize not only dissociative amnesia, but could be part of other dissociative disorders, such as dissociative identity disorder (DID), dissociative fugue, Ganser syndrome, and dissociative trance disorder (possession trance), as well as certain anxiety disorders, such as acute stress disorder and post-traumatic stress disorder (PTSD) or personality disorders–borderline personality disorder (McKay & Kopelman, 2009; Staniloiu et al., 2009; Zanarini, Frankenburg, Jager-Hyman, Reich, & Fitzmaurice, 2008). Ganser syndrome is currently listed under the category of Dissociative Disorders Not Otherwise Specified in DSM-IV-TR and is defined by giving approximate answers to questions (vorbeireden). Ganser’s (1898, 1904) original clinical description of the syndrome was, however, much broader than the current DSM-IV-TR one. It included, in addition to the tendency to give approximate answers, a hysterical semi-trance condition, amnesia and hallucinations, being more consistent with subsequent views of this disorder as a brief reactive psychosis to stress. Dissociative memory symptoms in the form of hypermnesia (‘‘flashbacks”) or amnesia may also occur in PTSD. PTSD conditions which are accompanied by ‘‘positive” dissociative symptoms such as flashbacks and intrusions seem to engage different neural networks than PTSD conditions that are accompanied by ‘‘negative” dissociative symptoms such as amnesia (Markowitsch, 2000). Lanius et al. (2005) found that in general a network of prefronto–temporo-parietal areas was engaged
786
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
in all patients, but that the group with ‘‘positive” symptoms had – compared to normal subjects – in addition a greater covariation with the right insula and right visual association cortex (compared to the reference in the left ventrolateral thalamus). Between the two groups, that with negative symptoms (amnesia) showed – compared to the ‘‘positive” (flashbacks) group – a more significant covariation in the left inferior frontal gyrus, while vice versa the ‘‘positive” symptoms group had more significant covariations with posterior cingulate/precuneus regions, the right middle temporal and the left inferior frontal gyri. In a recent review of various imaging studies of PTSD, arguments are made for a mechanism of undermodulation of affect via failure of prefrontal inhibition of limbic regions (such as amygdala) underlying the re-experiencing/hyperarousal PTSD subtype and one of overmodulation of the emotional limbic responses in the (‘‘negative”) dissociative PTSD subtype (Lanius et al., 2010). Psychogenic amnesias could further be differentiated according to the timeframe (anterograde versus retrograde), degree (global versus selective) of impairment of autobiographical–episodic memory and the co-existence of deficits in autobiographical–semantic memory and general semantic knowledge (cf. Serra, Fadda, Buccione, Caltagirone, & Carlesimo, 2007). The most frequent manifestations of psychogenic amnesias are forms of retrograde amnesias. A particular type of psychogenic retrograde amnesia has been termed the ‘mnestic block syndrome’ (Markowitsch, Kessler, Russ, et al., 1999). This form is characterized by an AEM block, which may comprise the whole past life. Affected patients otherwise have largely preserved semantic memories; they typically can read, write, calculate and know how to behave in social situations. Furthermore, they can store new long-term AEMs, though the acquisition of these new events may be less emotionally tagged than in normal subjects (Brand & Markowitsch, 2009; Reinhold & Markowitsch, 2009). When retrograde psychogenic amnesia is accompanied by sudden leaving of the customary environment – home and city – and compromised knowledge about personal identity, the condition is referred to as psychogenic (dissociative) fugue (Markowitsch, Fink, Thöne, Kessler, & Heiss, 1997). One example of dissociative fugue is the case of a man in his 30’s, who one morning took his bike to go to the bakery, but then suffered a sudden episode of amnesia. He rode his bike along the river for a whole week, being unable to recall his past and not knowing who he was (Markowitsch, Fink, Thöne, et al., 1997). Fugues have been reported for over a century (Markowitsch, 1992), though they were frequently erroneously associated with epilepsy (e.g., Burgl, 1900). A century ago, the fugue condition was named Wanderlust in Germany (cf. e.g. Burgl, 1900). Most fugues were not found to involve the formation of a new identity and were reported to be brief (Maldonado & Spiegel, 2008). However, prolonged courses of fugues were also described (Hennig-Fast et al., 2008). With regards to the degree of impairment of AEM loss, sometimes the ‘blocked’ autobiographical–episodic material is content-specific (selective) and/or retrograde amnesia is limited to specific life epochs (Markowitsch, Thiel, Kessler, et al., 1997). Although anterograde memory deficits occasionally accompany retrograde psychogenic amnesia, cases of psychogenic anterograde amnesia (inability to store new AEM episodes long-term) with preserved retrograde autobiographical–episodic memory have only been reported rarely (Kessler et al., 1997; Kumar, Rao, Sunny, & Gangadhar, 2007; Markowitsch, Kessler, Kalbe, & Herholz, 1999; Smith et al., in press; Staniloiu, Borsutzky, & Markowitsch, 2010). In a previous article, this condition (anterograde psychogenic amnesia) was described in a young woman who suffered a whiplash injury and who manifested a severe AEM block, which persisted for more than 15 years. After the whiplash injury she displayed an ongoing inability to acquire any new knowledge or AEM long-term, while her memories up to the time of the event remained precise and vivid. Similar to cases reported by Kapur et al. (1997) and O’Connor, Sieggreen, Ahern, Schomer, and Mesulam (1997) this young woman manifested accelerated forgetting, while storing information successfully for one to four hours. Despite that some patients with psychogenic amnesia claim loss of various procedural skills (Serra et al., 2007; van der Hart & Nijenhuis, 2001), objective evidence of procedural memory impairments in psychogenic or functional amnesia remains limited or debatable (Glisky et al., 2004; Huntjens, Postma, Woertman, van der Hart, & Peters, 2005; Smith et al., in press). In comparison to the terms dissociative or psychogenic amnesia, functional amnesia includes also forms of amnesia without significant brain damage as detected by conventional structural brain imaging and no clearly identifiable psychological etiological mechanisms underlying the memory disorder (De Renzi, Lucchelli, Muggia, & Spinnler, 1997; Fujiwara et al., 2008). The lack of clearly identifiable psychological triggers in functional amnesia may be explained by several factors: (a) The memory disturbance for the stressful event due to amnesia. (b) An impaired capacity for emotional awareness and processing in the face of ongoing or repeated stresses that seems to premorbidly characterize some of the patients with psychogenic amnesia and in fact might predispose them to develop this condition (Staniloiu et al., 2009). (c) The possible involvement of mechanisms of kindling sensitization in the face of recurrent stresses, which may trigger an episode of illness after a seemingly minor stress (Post, Weiss, Smith, Rosen, & Frye, 1995). (d) The so called incubation effect of life adversity (see below). (e) The personal view of the trauma within the explanatory model of illness (Kleinman, 1980) that may include direct exposure to trauma as well as vicarious traumatization (Maldonado et al., 2002). (f) The so- called process of ‘‘re-contextualization of health memory” (Modell, 2006) and finally (g) The presence of morphological changes, which could not be captured by conventional structural brain investigations (Bigler, 2004) but may be visualized with other techniques (such as Diffusion Tensor Imaging [DTI]).
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
787
6. The Janus head of amnesia: psychogenic and organic As it is the case with a variety of psychiatric disorders, advances in neurosciences have challenged the validity of the old distinctions in organic and psychogenic (Markowitsch, 1996, 1999a; Ouellet, Rouleau, Labrecque, Bernier, & Scherzer, 2008). Functional imaging studies have delivered evidence for a long searched biological link between the organic and psychogenic amnesia, by demonstrating brain functional and metabolic alterations in psychogenic or functional amnesia, which involve anatomical regions that are agreed upon to exert a crucial role in memory processing (Markowitsch, 1999b; Staniloiu & Markowitsch, in press). In a single case study of a patient with psychogenic amnesia after a traumatic event, involving severe retrograde and anterograde memory deficits, we identified via positron-emission-tomography (PET) an overall brain glucose hypometabolism with a more severe reduction of metabolism in the patient’s anterior and medial temporal lobes and the diencephalon (Markowitsch, Kessler, Van der Ven, Weber-Luxenburger, & Heiss, 1998). The patient’s memory recovery was observed approximately 8–12 months after the initiation of psychotherapeutic and psychopharmacological treatment and was paralleled by a normalization of the brain’s glucose metabolism (Markowitsch et al., 2000). Reductions in the brain’s glucose consumption, which preponderantly involved brain regions known to be part of the AEM network, were also found in subsequent investigations of other patients with psychogenic amnesia and are reviewed in Reinhold, Kühnel, Brand, and Markowitsch (2006), Brand and Markowitsch (2009) and Staniloiu and Markowitsch (in press). In a recent study of Brand, Eggers et al. (2009) that analyzed the results of functional brain imaging performed in resting state in 14 patients with psychogenic amnesia, reductions in metabolism of the right temporofrontal regions were identified, with a common significant hypometabolic zone in the right inferolateral prefrontal cortex (compared to the metabolism of control subjects). Similarly, Tramoni et al. (2009) found evidence of changes in the right prefrontal area in a patient with functional amnesia. By combining functional imaging with structural MRI methods sensitive for assessing white matter integrity, the latter identified subtle structural changes in several long fiber tracts passing through the inferolateral prefrontal cortex or its neighborhood. As the case-report of Tramoni et al. (2009) suggests, the gap between psychogenic and organic has furthermore been narrowed since the advent of newer structural techniques, such as DTI or magnetization transfer ratio measurements, which improved the ability to detect the integrity of brain white matter. Though cases of ‘‘organic” amnesia associated with fiber tract damage have been reported since long, the expanding use of the above mentioned techniques has led to an increase in findings of white matter damage in association with memory impairment or amnesia as well as a renewed interest in the mechanism of disconnection (Markowitsch, 1984). Techniques like DTI (Beaulieu, 2009) or magnetization transfer measurement (Tramoni et al., 2009) may in our opinion prove in the future to be valuable tools for identifying the neural correlates of certain forms of functional amnesia, such as the ones occurring after a mild traumatic brain injury (Staniloiu et al., 2009; Staniloiu, Markowitsch, & Borsutzky, 2010). This might be of importance, given that traumatic brain injuries are a frequent occurrence, but they are, on the one hand, underreported or documented (Ruff, Iverson, Barth, Bush, & Broshek, 2009). On the other hand, a substantial number of patients with mild traumatic brain injury (TBI) experience dissociative symptoms, which typically predict a poor outcome (Mooney & Speed, 2001). Finally, although in most of the cases of mild TBI conventional neuroimaging reveals no objective brain damage, recent studies evidenced via DTI white matter changes in specific brain regions of patients with mild TBI, which correlated with their degree of cognitive dysfunction (Lipton et al., 2008, 2009). The validity of a strict demarcation between organic and psychogenic amnesia has not only been questioned by the results of the functional brain imaging, but also by the observation that in some cases of organic amnesia psychological factors might play a significant role in exacerbating or maintaining the symptomatology (Kopelman, 2000, 2008). Furthermore symptom exaggeration has been reported in a variety of psychiatric and non-psychiatric conditions, including traumatic brain injury, depression, and dissociative disorders (Bass & Halligan, 2007). As suggested already in 1943 by Lennox, feigned amnesia (for primary or secondary gains) may accompany both organic and psychogenic amnesia. For example, ‘‘an epileptic, who also has hysteria and is a malingerer, may have periods of amnesia which exhibit features of all three types” (‘‘pathological”-organic, ‘‘psychological” and ‘‘feigned”; Lennox, 1943, p. 741). Differentiating psychogenic (dissociative) amnesia from disorders, which involve intentional production or feigning of symptoms (such as malingering and factious disorders), constitutes a diagnostic requirement that may pose a variety of challenges, especially in the forensic settings. Part of the challenge arises from the fact that some cases of psychogenic amnesia occur on a forensic background (Kritchevsky et al., 2004; Markowitsch, 1992; Markowitsch, Calabrese, Fink, et al., 1997; Staniloiu & Markowitsch, in press). Although dissociative (conversion) disorders (which were formerly captured under the name of hysteria) have traditionally been associated preponderantly with female gender, recent findings suggest that the frequency of dissociative disorders in men may have been underestimated, arguably partly because men with dissociative disorders may enter more frequently the legal system facilities than psychiatric settings (Spitzer & Freyberger, 2008). On the other hand, subjects who after a traumatic incident malinger cognitive deficits for financial gains, tend to typically claim memory impairments (Serra et al., 2007). In comparison to malingering that involves the intentional feigning of symptoms for legal, financial or economic gains, the intentional production of symptoms in factitious disorder is considered to be solely motivated by the wish to assume the sick role. While intentionality is not part of the diagnostic description of psychogenic (dissociative) disorders, an accurate clinical assessment of someone‘s motivations to assume the sick role remains a difficult task (Bass & Halligan, 2007).
788
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
This observation has led several authors to propose a working model where dissociative (conversion) disorders, somatoform disorders and factitious disorders are conceptualized as being part of a continuum rather than sharply delineated diagnostic entities (Barbarotto, Laiacona, & Cocchini, 1996; Di Fiorino, 2003; Krahn, Bostwick, & Stonnington, 2008; Mayou, Kirmayer, Simon, Kroenke, & Sharpe, 2005). 7. Stress and psychogenic amnesia Psychogenic amnesias have been described in many cultures and recognized to occur in response to traumatic or psychological stress (Seligman & Kirmayer, 2008; Staniloiu & Markowitsch, in press). They could co-exist with other psychiatric disorders, such as major depressive disorder, bulimia nervosa, alcohol abuse or borderline, histrionic or narcissistic personality disorder (Maldonado & Spiegel, 2008). Psychogenic amnesias tend to affect younger people (Reinhold & Markowitsch, 2007; Yehuda, Schmeidler, Siever, Binder-Brynes, & Elkin, 1997), being more commonly reported in the third and fourth decade of life (Coons & Milstein, 1992; Kanzer, 1939). The age distribution on one hand echoes the findings of a negative correlation between age and dissociation scale scores (Putnam, 1997). On the other hand, it may reflect the differences in windows of vulnerability to stress of the main brain structures involved in AEM processes (Lupien, McEwen, Gunnar, & Heim, 2009) as well as age-related changes in brain connectivity patterns subserving AEM processes (Addis, Leclerc, Muscatell, & Kensinger, 2010; St Jacques, BessetteSymons, & Cabeza, 2009). Although in several cases a direct relationship between the severity of exposure to trauma and incidence of amnesia was reported, there are case reports of psychogenic amnesia after a seemingly minor stressor (Markowitsch et al., 1998; Reinhold & Markowitsch, 2007, 2009). In most of the latter cases the collateral information revealed a history of previous repeated traumatic experiences, suggesting a pathogenetic model of kindling sensitization (Post et al., 1995) or alternatively an incubation effect (Lupien et al., 2009). The concept of kindling was introduced in psychiatry from epileptology, where it described the process by which repeated sub-threshold stimulation of a brain area eventually resulted in a seizure. In psychopathology kindling could explain among other the observed inverse relationship between the number of experienced depressive episodes and precipitating life events. The incubation effect means that the effects of adversity, which occur during a window of vulnerability (such as a critical period of synaptic organization), do not become evident at the time of adversity, but later (such as when the synaptic organization is completed) (Lupien et al., 2009). Prior to elaborating on the connection between stress and psychogenic (dissociative) amnesia, we would like to emphasize that both stress and dissociative responses could be conceptualized as part of a spectrum that ranges from physiological, adaptive or non-pathological to maladaptive and pathological. Some non-pathological dissociative experiences, such as day dreaming, absorption, reverie may involve positive emotions and/or may enhance performance by focused allocation of cognitive resources. Maldonado and Spiegel (2008) give the example of athletes who can perform extremely well while focussing on a particular detail of an event, suspending any critical thinking and allowing their bodies to automatically do what they are supposed to do. Stress hormones, such as glucocorticoids promote under normal conditions brain maturation and remodelling, cell survival and learning and memory. Elevated or decreased levels may, however, have negative consequences for brain function or morphology (Du et al., 2009). Our pathogenetic model of the mnestic block syndrome hypothesizes that increased glucocorticoid levels play a significant role in the memory retrieval block (Markowitsch, 1996, 2000) and has received empirical support from several studies (for a review, see De Kloet & Rinne, 2007). Although glucocorticoids are critically involved in stress responses, a variety of other hormones (including arginin-vasopressin [AVP], oxytocin) may, however, have modulating effects (Joels & Baram, 2009). The degree to which chronic repeated stress or massive acute stress may affect an individual’s homeostasis and lead to psychiatric and/or non-psychiatric disorders depends on a range of factors, such as genetic dispositions, type of stress, duration of stress, developmental phase, age, gender, context, prior experiences, personality characteristics. Genetic factors might influence not only hormonal stress responses (glucocorticoids, AVP), but also the development of brain structures, brain plasticity, brain function (de Quervain et al., 2007; Rasch et al., 2009), and the temperament (including proneness to dissociative experiences; Becker-Blease et al., 2004). Carriers of a functionally relevant deletion variant of ADRA2B, the gene encoding the a2b-adrenoreceptor, were for example found to exhibit enhanced memory for emotional material (de Quervain et al., 2007), likely via modulation of amygdala activity (Rasch et al., 2009). Although Hurlemann et al. (2005) proposed that both hypermnesia and peri-emotional amnesia (decreased memory for neutral events during simultaneous enhanced encoding of an aversive event) are amygdala-dependent and vary as a function of noradrenergic-glucocorticoid input to the amygdala, the genetic underpinnings of psychogenic memory loss remain still unclear. Certain genetic polymorphisms may bias the normal maturation trajectories of key brain structures that are involved in processes of memory and emotion, possibly rendering them more sensitive to the effects of stress. A recent study (Pacheco et al., 2009) that examined the impact of the polymorphism of the serotonin transporter gene promoter region (5-HTTLPR) on white matter tracts connections in young women identified a significant association between the number of low expressing 5-HTTLPR alleles and the microstructural changes of the uncinate fascicle. The creation of brain structure is however not the product of the genes alone, but the result of the frequently synergistic interplay between genes and experience (environment) (Rutter, Moffitt, & Caspi, 2006). Both animal and human studies demonstrate that early life experiences could alter the gene expression via epigenetic modifications and lead to long lasting
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
789
changes in stress hormonal responses, synaptic plasticity and behavior. McGowan et al. (2009) recently showed epigenetic modifications such as increased methylation of the promoters of several genes of interest in the brain of victims of suicide with a history of childhood abuse in comparison to the suicide completers without a history of childhood abuse. These findings mirrored the ones from animal studies, which revealed that early development experiences could have programming effects on the hypothalamo–pituitary–adrenal (HPA) axis (Champagne et al., 2008; Weaver et al., 2005). Accumulating evidence suggests that epigenetic changes might also be involved in processes of learning and memory such as the extinction of fear-related memories, or re-consolidation (Roth & Sweatt, 2009). Several key brain structures for autobiographical memory and emotional processing have been identified as being sensitive to the consequences of exposure to negative or stressful experiences (such as amygdala and hippocampal formation, prefrontal cortex and specific white matter tracts). Reductions in the hippocampal volumes and AEM impairments have been reported in patients with stress-related psychiatric conditions, such as post-traumatic stress disorders (Bremner et al., 1997), recurrent major depressive disorders (especially the ones with a history of trauma) (Campbell, Marriott, Nahmias, & MacQueen, 2004; Vythilingam et al., 2002) and dissociative identity disorder (Vermetten, Schmahl, Lindner, Loewenstein, & Bremner, 2006). The effects of stress on the above mentioned brain structures depend on both their vulnerability and the magnitude of neurotoxic (such as glucocorticoid) cascades (O’Brien, 1997). The stage of development or declining of the respective structures influences their susceptibility to the stress effects (Lupien et al., 2009). This may partly explain why the same type of traumatic experiences may be associated with different brain morphological or functional changes and psychopathology (De Kloet & Rinne, 2007). Along this line, one study found that repeated episodes of sexual abuse were accompanied by reduced hippocampal volume when the traumatic experiences occurred early in childhood, but reduced prefrontal cortex gray matter volume when the traumatic experiences occurred in adolescence (Andersen & Teicher, 2008). The latter findings are explained by the fact that the hippocampus is fully organized during adolescence, while other structures, such as amygdala and prefrontal cortex are still developing. As opposed to the hippocampi, amygdala volumes may increase in response to stress as a result of persistent dendritic growth (Mehta et al., 2009; Roozendaal, McEwen, & Chattarji, 2009). The existence of sexually dimorphic brain structures and functions may get translated in gender differentiated responses to stress (Cahill, 2006; Lanius, Hopper, & Menon, 2003; Piefke, Weiss, Markowitsch, & Fink, 2005; Schulte-Rüther, Markowitsch, Shah, Fink, & Piefke, 2008; Tranel & Bechara, 2009). Changes in white matter tracts have also been reported in response to early stress. In children with a history of early deprivation, microstructural morphological changes of the UF, which were more prominent on the right side, were reported recently (Govindan et al., 2010). These changes correlated with the length of the time spent in the deprived environment. The findings of microstructural alterations in the UF, as detected by DTI, in relation to traumatic stress led us to hypothesize that UF plays a role in some cases of psychogenic amnesia (where a hypometabolism of the right inferior lateral prefrontal cortex was evidenced; Brand et al., 2009). The impact of early experiences on AEM development has also received support from the observation that mothers, who build up secure attachments with their children, engage in more elaborated reminiscing, which facilitates an earlier onset of AEMs and promotes a more elaborated encoding of AEMs in their offspring (Nelson & Fivush, 2004; Siegel, 2004). A reduced AEM specificity was described in children with a history of abuse or neglect (Valentino et al., 2009) and a combination of trauma and disorganized parent–child attachment in early life had empirically been linked to dissociative tendencies in late adolescence (Carlson, 1998; Siegel, 2004). 8. Psychogenic amnesia, emotions/feelings and personal meaning A central feature of AEM is its relationship with emotions. Emotions have an intimate connection to (one-)self and are used for appraising one‘s own position with respect to environment. Emotional evaluation is of importance for the acts of encoding, storage as well as for the recall of AEM. A ‘‘non-specific blurring or flattening” (Markowitsch, 1999b) in AEM (or so called overgeneralization phenomenon; Williams & Scott, 1988) has, for example, been described in patients with major depressive disorder. In contrast to semantic memories, which only display a low correlation with emotion (Welzer & Markowitsch, 2005), most AEM reminiscences are significantly affectively laden. At the brain level, this is reflected by findings that the recall of AEM memories typically engages networks involved in integrating emotion and cognition such as right temporal–frontal areas and right amygdala. The observation that the feelings have motivational functions and assign ‘‘value to what is meaningful” (Modell, 2006, p. 151), may partly explain the dependency of the experience of a coherent personal identity on the explicit episodic memories of elements in one’s own life history. Along this line, our own observations revealed that patients with psychogenic amnesia often displayed a blunted affective disposition along with a resignation to the present situation. Furthermore, several patients with psychogenic amnesia encountered difficulties with judging the feelings (Kritchevsky et al., 2004) and intentions of others (Reinhold & Markowitsch, 2007) as well as interpersonal relationships. This is not surprising as our emotions are molded or modulated by social environment and as ‘‘social animals” (Darwin, 1871/2004) we are particularly dependent on appropriately regulating, expressing and communicating our feelings and emotions in a social context (Leary, 2007). Furthermore certain so called social brain structures, such as the amygdala are also involved in AEM processes. Cahill et al. (1995) found for example that patients with bilateral amygdala degeneration are impaired in getting the gist of a told story; they fail
790
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
to repeat (and consequently to remember) the emotionally important parts of it and instead tend to concentrate on the emotionally less important parts. These patients furthermore appear socially impaired in everyday life situations (Siebert et al., 2003). Similarly, patients with amygdala damage fail to appropriate judge emotions of others, which may be partly related to the role played by amygdala in directing the gaze towards the eye region of the other person (Gamer & Büchel, 2009; Kawashima et al., 1999; Vuilleumier & Pourtois, 2007). These findings suggest that not only the feeling of a coherent personal identity (being an autonoetic self), but also the ‘‘feeling” and ‘‘imagining” (Modell, 2006) of others (which facilitates the being of oneself with others) may have linkages to AEM. 9. Psychogenic amnesia and self versus other AEM, self-consciousness and theory of mind (ToM) seem to be correlated early during the ontogenetic development (Nelson & Fivush, 2004; Perner, 2000). The ToM capacity, which is defined as the ability to attribute and make inferences about the mental states of other people, such as desires, beliefs, intentions and feelings, and to differentiate between own and others’ mental states, undergoes significant and gradual changes during childhood. Between 16 and 24 months of age children pass the mirror recognition test, which marks the transition from a core self (Damasio, 1999; Nelson & Fivush, 2004) to a cognitive self (Howe & Courage, 1993) and is opined to be a necessary, but not sufficient condition for the development of episodic (Howe & Courage, 1993) or ‘‘evocative” memory (Adler, 1981). An enduring sense of self across time as detected through the delayed self recognition paradigm of Povinelli, Landau, and Perilloux (1996) is unusual in 2–3 year-old children, but is typically described in most 4 and 5 years old ones. Although individual, cultural and gender variations exist, the earliest childhood AEM dates to around age 3.5 years (Harpaz-Rotem & Hirst, 2005; Q. Wang, 2001). In the ToM domain, children seem to understand certain aspects of the mind from a strikingly early age, but it is not until around age 4 when they succeed in a standard false belief task (Baron-Cohen, Leslie, & Frith, 1985; Wimmer & Perner, 1983). These data led to the argument that meta-representational abilities, where there might be an (experiential) awareness – a state of consciousness similar to autonoetic consciousness or meta-representational self-consciousness (Newen and Vogeley, 2003) – of the relation between knowledge sources and present knowledge states (Perner, 2000) play a critical role in the development of ToM abilities. From an evolutionary point of view, it makes sense to believe that AEM, ToM and autonoetic self-consciousness also coevolved phylogenetically, driven by the forces of the social selection process (Corballis, 2009; Spreng, Mar, & Kim, 2008). In addition to directive, self representation and survival functions AEM has been attributed a significant social role (Bluck, Alea, Habermas, & Rubin, 2005). AEM contains a wealth of information about people and social interactions and the exchange of AEM enhances social and romantic bonding, understanding of others’ inner world and perspective. Additionally, it enables the intergenerational sharing of the past experiences, including the moral perspective (Nelson & Fivush, 2004). As Tulving (2005) remarked, possible ties between moral judgment and AEM were suggested by Darwin’s description of ‘‘moral being”, which underlined several features that match essential characteristics of AEM such as the capacity for autonoetic consciousness. While the relationship between AEM and the capacity for moral development remains still largely unexplored (Robertson et al., 2007), AEM impairments (especially for emotional events) were reported in offenders with psychopathy by several authors (Bourget & Whitehurst, 2007; M.C. Craig et al., 2009). Furthermore a recent study found evidence of microstructural integrity in the right uncinate fasciculus of forensic patients with psychopathy (M.C. Craig et al., 2009). Though language is not a necessary condition for AEM (Tulving, 2005), it supports and enriches the development of AEM capacities. It has been argued that language and mental time traveling (Suddendorf, Addis, & Corballis, 2009; Tulving & Kim, 2007) have co-evolved (Corballis, 2010). In contrast to the Chomskian’s view, some authors hypothesized that the languagebased communication system might have been preceded by the mimetic gesture (Corballis, 2009; Tomasello, 2008). Therefore language, similar to certain mechanisms involved in ToM (Boria et al., 2009; Cattaneo & Rizzolatti, 2009; Rizzolatti, Fabbri-Destro, & Cattaneo, 2009) might have had its roots in mirror neurons-containing brain regions such as territories including areas in the inferior frontal gyri. ToM abilities stem from innate neuro-cognitive capacities that are subserved by specific neural networks, which include right and left temporo-parietal junction, medial prefrontal cortex, precuneus, anterior temporal sulci and amygdala (Carrington & Bailey, 2009; Shaw et al., 2004). Congenitally blind people seem to activate similar neural networks during ToM tasks as people with normal vision (Bedny, Pascual-Leone, & Saxe, 2009). Partly dissociable neural systems underlie affective ToM functions (that are related to empathic processes) versus cognitive ToM functions (Shamay-Tsoory & Aharon-Peretz, 2007), with a preponderant role of ventromedial-prefrontal cortex areas for the affective ToM and cognitive aspects of empathy (Shamay-Tsoory, Aharon-Peretz, & Perry, 2009). Similar to AEM the ‘‘emergence” of the ToM capacity of an individual depends on other cognitive functions as well as environmental factors, including early life experiences and socio-cultural milieu. The nature of early attachments with primary care givers and mothers’ reminiscing style influence not only the development of AEM, but also of ToM capacities. The impaired performance on ToM tasks of children who were raised in deprived institutional settings reflects the importance of early life experiences with caregivers for the development of the ToM (Bedny et al., 2009; Colvert et al., 2008). In the field of brain functional imaging significant overlaps in the brain networks that become activated during the assessment of AEM, ToM, prospection (constructive episodic simulation), self projection, scene construction, navigation and the
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
791
default mode (Addis, Pan, Vu, Laiser, & Schacter, 2009; Buckner & Carroll, 2007; Hassabis & Maguire, 2007; Rabin, Gilboa, Stuss, Mar, & Rosenbaum, 2010) have been identified. In a recent meta-analysis the strongest overlap was reported for the neural networks subserving AEM and ToM tasks (Spreng et al., 2008). In spite of these results and sophisticated elaborations in the theoretical domains, the relationship between AEM and ToM has however been insufficiently experimentally investigated (Rabin et al., 2010; Spreng & Grady, 2010). Co-occurring impairments of AEM, ToM and autonoetic self awareness have been described in certain psychiatric disorders, such as schizophrenia (Corcoran & Frith, 2003), but it is unclear to which degree they reflect a developmental arrest of closely in time emerging neuro-cognitive functions or an early functional interdependence. Co-existing impairments of AEM and ToM have also been reported in the autistic spectrum disorders (Lind & Bowler, 2009; Shalom, 2003), as well as in mood disorders, such as major depressive disorder (MDD) and bipolar disorder (Inoue, Tonooka, Yamada, & Kanba, 2004; ShamayTsoory, Harari, Szepsenwol, & Levkovitz, 2009). Interestingly, in all these disorders microstructual changes in white matter tracts – particularly in the uncinate fascicle – were reported or suspected to exist (Kubicki et al., 2007; Pugliese et al., 2009; Taylor, Macfall, Gerig, & Krishnan, 2007; F. Wang et al., 2009). Lesions of the UF as well as of the capsula interna have been linked to memory impairments (Diehl et al., 2008; Fouquet et al., 2009; Markowitsch, von Cramon, Hofmann, Sick, & Kinzler, 1990; Sepulcre et al., 2008); furthermore, there are a few reports of ToM impairments occurring after white matter tract damage (Bach et al., 1998; Happé, Malhi, & Checkley, 2001). Several different theoretical models have attempted to capture the developmental interdependence between theory of mind and AEM (Saxe, Moran, Scholz, & Gabrieli, 2006). One model suggests that AEM depends on ToM and received some support from findings that performance on ToM tasks predicted performance on AEM tasks (Perner, 2001) and good ToM abilities are protective against false memories or contamination of AEM through suggestion (Bright-Paul, Jarrold, & Wright, 2008; Welch-Ross, Diecidue, & Miller, 1997). Other models propose that ToM abilities are at least partly dependent on AEM memory capacities or alternatively that the apparent connection between ToM and AEM tasks is mediated by autonoetic consciousness. Though many researchers would agree – or at least would not reject – that some interdependence between ToM and AEM characterizes early human development, findings from adult patients with brain lesions suggest that adult ToM and AEM abilities are at least partly dissociable (Bird, Castelli, Malik, Frith, & Husain, 2004). Rosenbaum, Stuss, Levine, and Tulving (2007) concluded from their observation in two patients with organic brain damage that a severe impairment of AEM and autonoetic consciousness does not affect the expression of premorbidly acquired ToM capacities; our findings from patients with psychogenic amnesia, however, suggest a certain degree of interdependence between ToM and AEM, which extends beyond childhood development (e.g., Brand et al., 2009; Fujiwara et al., 2008; Reinhold & Markowitsch, 2007, 2008, 2009). We advance several possible explanations for our results: (1) One possibility is that the acquisition of ToM and AEM capacities might initially engage similar neural networks. Once the ToM capacities are fully acquired and matured, they might be subserved by partially different neural networks, but the initial networks might remain available and might still be activated to accomplish or facilitate the performance of more difficult or complex novel tasks (Bird et al., 2004). (2) AEMs represent valuable reservoirs of information that could be flexibly used during social problem solving. In healthy subjects, however, it is unlikely that these reservoirs are accessed for habitual tasks, as for these it is relied on semantic and procedural knowledge (Rabin et al., 2010), which are more cognitively economical. Indeed in Hyman and Faries‘s (1992) research only few self-reports were recorded where AEM was used directively to solve problems. Instead a much more common reason for recalling or retelling memories was the sharing of experience and giving advice. (3) Depending on context, task requirements, gender, developmental stage, experience and cultural affiliations (e.g. Han & Northoff, 2008; Kobayashi, Glover, & Temple, 2007; Ray et al., 2010) individuals might resort to different strategies during ToM tasks. While some subjects might preponderantly rely on semantic memory and general knowledge for solving ToM tasks, others might engage more AEM resources. Several neuroimaging studies that investigated the neural correlates of ToM tasks (such as other minds‘ representation) (Legrand & Ruby, 2009; Schulte-Rüther, Markowitsch, Fink, & Piefke, 2007) produced activations of brain regions that are important for the recall of AEM events and elaboration of future events (Markowitsch, 2005). One possible interpretation of these results is an increased need of autobiographical–episodic recall for another versus self ToM tasks. Evidence for the engagement of different strategies and neural networks during ToM tasks was also provided by studies that investigated the reading of minds of dissimilar others versus similar others and neutral mind reading versus mind reading in emotional context (J.P. Legrand & Ruby, 2009; Mitchell, Macrae, & Banaji, 2006). (4) Patients with impairments of AEM, who perform well on laboratory ToM tasks, might still experience difficulties with ToM tasks in naturalistic settings, when confronted with novel, challenging social environments that may benefit from the use of AEM database or a higher degree of imagination. (5) Certain contexts might be conducive to more engagement of AEM during ToM tasks. In a compelling theoretical review of empathy as applied to psychoanalytical context, Buie (1981) described several types of empathy that could unfold in the therapeutic process: conceptual, resonant, self-experiential, imaginative imitation empathy. The latter occurs when a therapist must resort to his imagination and fantasy to find what it would be like to be in the other person’s shoes (i.e., in approximating experiential understanding) (Nagel, 1974). We would speculate that the professional use
792
(6)
(7)
(8) (9)
(10)
(11)
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
of empathy in the psychoanalytical context might offer an example of a setting that would promote the use of AEM resources to modulate empathetic responses, such as during self-experiential empathy or so called imaginative imitation empathy. Gender differences may influence the degree to which the AEMs are engaged during theory of mind tasks and empathic responses, as suggested by a study that showed that women who had a similar experience reported more empathy than women who had not, while men who had a similar experience reported no more empathy than men who had not (Batson et al., 1996). The above described gender differences might partly reflect variations in socialization. Parental expectations of girls versus boys and gender differences in language ability, temperament and interest in social interaction may elicit different maternal reminiscing styles (Nelson & Fivush, 2004). We speculate that this might not only be conducive to gender differences in AEM encoding (such as facilitating a more detailed and emotionally-laden AEM encoding in girls versus boys), but may also promote a gender-differentiated inclination to use of AEM information to modulate ToM functioning and empathic responses. Certain pathological conditions might be accompanied by an increased reliance on AEM or self-reflection for ToM tasks (which may not necessarily lead to an enhanced performance on ToM tasks, but sometimes on the contrary). For example, it has been observed that patients with major depressive disorder display a higher tendency to attribute negative emotions to neutral faces (Phillips, Drevets, Rauch, & Lane, 2003). In addition especially in patients suffering from chronic forms of depression deficits of ToM and capacity for true empathic engagement were described. We conjecture that these deficits may reflect several interacting factors, such as a pervasively egocentric perspective (McCullough, 2000) that might bias these patients towards using their own minds as a template for other people’s experience via simulation, an increased self focus (Grimm et al., 2009), a failure of inhibition of self perspective and/or a poverty of models of the inner world of others (Newen et al., 2009). Age, perhaps as a measure of experience and /or brain networks’ functional connectivity may also influence the degree to which information from AEM is accessed for the performance of ToM tasks (Pfeifer et al., 2010). Another possible explanation is that subtle deficits of ToM capacities might have been present in patients with psychogenic amnesia premorbidly, but that they were significantly exacerbated by the onset of amnesia. Depressive disorders, subclinical depressive symptoms and emotional processing difficulties have been reported to either precede or accompany dissociative disorders (Markowitsch, Kessler, Russ, et al., 1999; Markowitsch et al., 1998). Alexithymic traits – consisting of difficulty with identifying feelings and distinguishing between feelings and the bodily sensations of emotional arousal and verbalizing feelings to other people, reduced imagination and fantasy and externally oriented cognitive style – have been reported to be more common in men and have been linked to increased proneness to dissociation (Modestin, Lotscher, & Erni, 2002). They could accompany several conditions such as conversion disorder, somatoform disorders, fibromyalgia, depression, anxiety and traumatic brain injuries (including mild ones) (Koponen et al., 2005; Mattila et al., 2009). Alexithymic traits have been linked to early traumatic experiences and a dysfunction of the right hemisphere (Moriguchi et al., 2006; Schore, 2002) or inter-hemispheric transfer (Romei et al., 2008). High alexithymic scores have been found to correlate with impaired ToM, deficient self awareness (Moriguchi et al., 2009) and decreased gray matter volumes of brain regions that are known to be important for emotional processing, self awareness, ToM and imagination (e.g., precuneus) (Borsci et al., 2009). Another explanatory avenue is that some, but not all of our patients with psychogenic amnesia and co-existing impairments of AEM and ToM displayed a degree of executive (such as cognitive flexibility) dysfunctions and/or semantic knowledge impairments (Bull, Phillips, & Conway, 2008). Finally, another hypothesis is that the lack of significant impairment on ToM tasks that was observed in some of the patients with ‘‘organic” episodic memory impairment reflected a process of brain reorganization, which allowed other neighboring brain structures to undertake the ToM tasks (Reinhold & Markowitsch, 2008). Degenerative disorders usually develop over a long period of time (Mondadori et al., 2006). An interesting observation is that the two cases presented by Rosenbaum et al. (2007), who experienced significant memory impairments, had their ToM capabilities tested several years after the onset of their AEM problems. Consequently, one could argue that in the meantime significant neural reorganization could have occurred. In fact, brain reorganization has been described in several conditions, including a recent case who despite removal of the whole Broca area for a slowly developing brain tumor, suffered only very subtle speech impairment (Plaza, Gatignol, Leroy, & Duffau, 2009). Similarly, Henke et al. (2003) described a patient who suffered no major memory problems in spite of undergoing one-hemispheric surgical removal of his medial temporal lobe, including the hippocampal region, after epilepsy-related damage of the contralateral medial temporal lobe.
We would like to add that employing of AEM information for the ToM tasks might not necessarily get translated into an enhanced accuracy of performance. This might be in a way similar to accessing AEM information to simulate future events. Gilbert and Wilson (2007) described several errors which may accompany future simulation, such as relying on the most recent or salient AEM information. We could speculate that the reliance on most recent AEM to simulate the future may in young adults be related to the stronger connection of these memories to the self, as evidenced by an fMRI study in young women, which demonstrated in this group a specific role of medial prefrontal cortex in retrieving recent AEMs (Oddo et al., 2010).
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
793
10. Psychogenic amnesia and future memories of the self ‘‘Time present and time past Are both perhaps present in time future” (Eliot, 1969, p. 171) An extended self emerges ontogenetically in conjunction with AEM, which throughout lifetime continues to support the self‘s vigorousness and coherence. Through the medium of autonoetic awareness which accompanies AEM, the extended self is capable of subjective mental time traveling in the past, present and future (Markowitsch, 2003; Tulving, 2005). Though traditionally the AEM was viewed as acting primarily as a storehouse of the past (Augustine, 1907/2001), recent data suggest that AEM may serve a significant proscopic function. Neuroimaging studies have revealed that similar networks, which serve AEM are engaged in self projection and construction of future events (Schacter & Addis, 2009). These findings might finally offer the long-time searched explanation to the evolutionists who struggled to find a survival advantage for the existence in humans of an AEM system that only offered an imperfect repository of the past. Indeed, it has since long time been observed that AEM are fragile and susceptible to forgetting and distortion (Kühnel et al., 2008; Loftus, 2005). Freud noted in a letter to Fliess that ‘‘the material present in the form of memory traces (is) being subjected from time to time to a rearrangement in accordance to fresh circumstances-to a retranscription” (Masson, 1985, p. 207). Furthermore several authors remarked that exceptional AEM abilities were not typically perceived as a blessing by their owners (Luria, 1987; Parker, Cahill, & McGaugh, 2006). On the contrary, similarly to the prospect that ‘‘every second of our lives recurs an infinite number of times” (Kundera, 1991, p. 5), the prospect of ‘‘running” (remembering) our entire life on a daily basis (Parker et al., 2006) has seemed terrifying and burdening. Moreover the benefits of an optimum balance between forgetting and remembering have been suggested since Ribot’s time (Hacking, 1995; Markowitsch & Brand, 2010). This provides further evidence that a main function of AEM in a permanently changing environment is to prepare human beings for the future, by reshaping and reconstructing the past to support current aspects of the self and match future goals that are coherent with one individual’s goals, self image and system of beliefs (Conway, 2009). The importance of AEM for mental time traveling (experiencing time versus knowing time) in the future is corroborated by findings from patients with organic or psychogenic amnesia as well other psychiatric disorders that are accompanied by AEM deficits (such as major depressive disorder). Patients with psychogenic or organic amnesia are often unable to plan for their personal future, being imprisoned or trapped in an extended or forever ‘noetic’ present (Suddendorf et al., 2009). Patients with psychogenic amnesia frequently show resignation to their present situation and an apparent lack of concern (la belle indifference; Janet, 1907) about their symptoms (Reinhold & Markowitsch, 2007; Serra et al., 2007). A main characteristic of patients with major depressive disorder is the inability to imagine personal positive future events (Sharot, Riccardi, Raio, & Phelps, 2007), which may lead them to attempt suicide. This inability may reflect a disruption of the balance between the neural networks that subserve the encoding and retrieval of positive versus negative AEMs (Lemogne et al., 2006; Markowitsch, Vandekerckhove, Lanfermann, & Russ, 2003), which may partly be mediated by an impaired sensitivity to positive experiences (Addis et al., 2010; Pizzagalli et al., 2009). Interestingly, several structures involved in AEM processing (such as basal forebrain, amygdala, ventromedial-prefrontal cortex) have also been reported to be engaged in reward-related processing, decision making and future-minded choice behavior (e.g., Daniel & Pollmann, 2010; Peters & Büchel, 2010; Ulrich-Lai & Herman, 2009). These may partly explain descriptions of changes in eating preferences, smoking or drinking habits or other previously rewarding activities (such as car driving) after the onset of psychogenic amnesia, which occurred without evidence for impairments in procedural knowledge (Fujiwara et al., 2008; Kritchevsky et al., 2004; Markowitsch, Fink, Thöne, et al., 1997; Thomas-Antérion, Mazzola, Foyatier-Michel, & Laurent, 2008). 11. Searching for the neural signature of the constricted self in the psychogenic amnesia Despite the risk of being criticized for adding a new form of self to the already existing long list of selves (Legrand & Ruby, 2009), we employed the term constricted self to metaphorically capture the symptomatology of patients with psychogenic amnesia (including the constriction of mental time traveling in the future). Janet (1907) talked about hysteria as a disorder of ‘‘personal synthesis”. As some authors pointed out (Spiegel, 2006; van der Hart & Dorahy, 2006), he often used the term désagregation (in spite of the availability in both his own and French traditional psychological vocabulary of the French term dissociation) (Janet, 1920). This suggests that he might have already understood this condition as involving not only a simple separation, but a failure of integration of various aspects of cognition (memory), self, consciousness and emotion (Maldonado & Spiegel, 2008). Similarly in several patients with psychogenic amnesia not only deficits of AEM, but also impairments of the capacities for self-consciousness, subjective time traveling, executive functions and ‘‘feeling” and ‘‘being” with others were observed (Brand et al., 2009). Repeatedly a neural signature of psychogenic amnesia has been identified in a particular area in the right prefrontal cortex. In a relatively large sample of patients with psychogenic amnesia characterized by severe retrograde AEM deficits and no overt structural abnormalities as detected by conventional imaging methods, evidence of functional changes (hypometabolism) during resting state in the right temporofrontal region with a significant decrease in the inferolateral prefrontal cortex was found (Brand et al., 2009). These changes in our opinion reflect the stress-mediated impairment of AEM retrieval. The detected location is consistent with previous findings from both cases of organic and psychogenic amnesia (Kroll et al., 1997; Piolino et al., 2005). The vicinity of this identified area with regions involved in processing of emotion, self-evaluation, self-regulation,
794
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
Fig. 6. The place of the insula in the fronto-temporal junction area.
consciousness, time perception, executive functions and theory of mind (Brand et al., 2009) offers an explanation for the observed malfunction of personal synthesis in patients with psychogenic amnesia. A neighboring brain region that has recently been receiving increased attention (A.D. Craig, 2010) in stress-related disturbances of self and consciousness (Lanius et al., 2005, 2010), including the ones accompanying the psychogenic memory loss (Brand et al., 2009; Thomas-Anterion, Guedj, Decousus, & Laurent, 2010) is the insula. Via glucose PET evidence of insular hypometabolism in patients with psychogenic amnesia was provided (Brand et al., 2009; Markowitsch et al., 1998; Thomas-Anterion et al., 2010). Hidden in the Sylvian sulcus in the triangle of frontal, parietal, and temporal cortex, the insula (Fig. 6) is connected to a variety of structures, including inferior frontal gyrus, septum and amygdala (Allman et al., 2010; Markowitsch, Emmans, Irle, Streicher, & Preilowski, 1985; Nieuwenhuys, Voogd, & van Huijzen, 1978). It sends projections to hippocampus and receives efferents from the entorhinal cortex (Nieuwenhuys et al., 1978). The most anterior and ventral portion of the insula that is close to the frontal operculum contains the so-called von Economo neurons (VEN) (Allman, Watson, Tetreault, & Hakeem, 2005). In the right hemisphere the human postnatal insular-frontal cortex contains significantly more VENs than in the left (Allman et al., 2010). The VEN neurons, which are also present in the anterior cingular cortex (ACC), have been ascribed functions in consciousness, social cognition and regulation of appetite (Allman et al., 2010). Their degeneration was for example found to be associated with loss of emotional awareness and altered self-consciousness in patients with certain variants of fronto-temporal dementia (Seeley, 2008; Sturm, Rosen, Allison, Miller, & Levenson, 2006) as well as aberrant eating habits (Seeley, 2008). Insula has also been assigned functions in verbal memory tasks (Grasby et al., 1994; Morin, 2009; Morin & Michaud, 2007), inner speech (Vercammen, Knegtering, Bruggeman, & Aleman, 2010), time perception (Rao, Mayer, & Harrington, 2001) and drug (smoking) cravings (Naqvi, Rudrauf, Damasio, & Bechara, 2007.) Recent data from Arzy, Collette, Ionta, Fornari, and Blanke (2009) indicate that also self projection in time and ‘‘facilitation of future judgments with respect to one’s past ‘‘(p. 2016) engage the insula. Distortions of time perception have been described in patients with psychogenic (dissociative) amnesia (Steinberg, 2000). As mentioned above, smoking cessation and changes in eating preferences have also been reported after the onset of psychogenic memory loss (Fujiwara et al., 2008; Thomas-Antérion, Mazzola, Foyatier-Michel, & Laurent, 2008), though their neurobiological underpinnings are still unclear. In normal subjects increased activations of the insula have been evidenced in tasks involving self versus other conditions (Schilbach et al., 2006) and the right insula was particularly found to be activated during self face recognition (Devue et al., 2007). The insula has subsequently been proposed to be a structure that plays a critical role in supporting the so-called ‘‘sentient” (feeling) self (A.D. Craig, 2009). In light of these findings we would speculate that the efferents that the right inferior frontal gyrus receives from the right insula (J.R. Augustine, 1996), might provide gateways from the more primitive ‘‘feeling” self towards the extended or autobiographical self. 12. Conclusions The topic of self and self-consciousness has generated the interest of humanity for centuries. Adherents to Aristotelian or Heraclitean views of self (Moldoveanu & Stevenson, 2001) or opponents to the notion of self have all generated a significant
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
795
amount of sophisticated writings, reinforcing the idea that – in spite of differences of opinions regarding self-specificity or what it means to have a self – self is a topic of universal interest in multiple disciplines. Similar to the topic of selfhood, psychogenic amnesia has also ignited the interest of neurologists, psychiatrists and psychotherapists for more than a century and has generated a lot of debate. By providing a description of psychogenic amnesia in the present article we hope that we were able to reiterate the importance of AEM for supporting and maintaining a robust sense of autonoetic self, and to offer, in addition, some glimpses into the neurobiology that underlies the autobiographical– episodic remembrances of the past, as well as the self‘s memories of the future (Ingvar, 1985).
References Addis, D. R., Leclerc, C. M., Muscatell, K. A., & Kensinger, E. A. (2010). There are age-related changes in neural connectivity during the encoding of positive, but not negative information. Cortex, 46, 425–433. Addis, D. R., Pan, L., Vu, M.-A., Laiser, N., & Schacter, D. L. (2009). Constructive episodic simulation of the future and the past: Distinct subsystems of a core brain network mediate imaging and remembering. Neuropsychologia, 42, 2222–2238. Adler, G. (1981). The borderline–narcissistic personality disorder continuum. American Journal of Psychiatry, 138, 46–50. Allman, J. M., Tetreault, N. A., Hakeem, A. Y., Manaye, K. F., Semendeferi, K., Erwin, J. M., et al (2010). The von Economo neurons in frontoinsular and anterior cingulate cortex in great apes and humans. Brain Structure and Function, 214, 495–517. Allman, J. M., Watson, K. K., Tetreault, N. A., & Hakeem, A. Y. (2005). Intuition and autism: A possible role for Von Economo neurons. Trends in Cognitive Sciences, 9, 367–373. Andersen, S. L., & Teicher, M. H. (2008). Stress, sensitive periods and maturational events in adolescent depression. Trends in Neurosciences, 31, 183–191. Arzy, S., Collette, S., Ionta, S., Fornari, E., & Blanke, O. (2009). Subjective mental time: The functional architecture of projecting the self to past and future. European Journal of Neuroscience, 30, 2009–2017. Augustine, J. R. (1996). Circuitry and functional aspects of the insular lobe in primates including humans. Brain Research Reviews, 22, 229–244. Augustine (2001). The confessions. New York: A.A. Knopf. Bach, I., Davis, S., Colvin, C., Wijerante, C., Happé, F., & Howard, R. (1998). A neuropsychological investigation of theory of mind in an elderly lady with frontal leucotomy. Cognitive Neuropsychiatry, 3, 139–159. Baddeley, A. (2000). The episodic buffer: a new component of working memory? Trends in Cognitive Sciences, 4, 417–423. Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. A. Bower (Ed.), Recent advances in learning and motivation (pp. 47–89). New York: Academic Press. Barbarotto, R., Laiacona, M., & Cocchini, G. (1996). A case of simulated, psychogenic or focal pure retrograde amnesia: Did an entire life become unconscious? Neuropsychologia, 34, 575–585. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘‘theory of mind”? Cognition, 21, 37–46. Bass, C., & Halligan, P. W. (2007). Illness related deception: Social or psychiatric problem? Journal of the Royal Society of Medicine, 100, 81–84. Batson, C. D., Sympson, S. C., Hindman, J. L., Decruz, P., Todd, R. M., Weeks, J. L., et al (1996). ‘‘I’ve been there, too”: Effect on empathy of prior experience with a need. Personality and Social Psychology Bulletin, 22, 474–482. Beaulieu, C. (2009). The biological basis of diffusion anisotropy. In H. Johansen-Berg & T. E. J. Behring (Eds.), Diffusion MRI: From quantitative measurement to in vivo neuroanatomy (pp. 105–126). London, UK: Elsevier. Becker-Blease, K. A., Deater-Deckard, K., Eley, T., Freyd, J. J., Stevenson, J., & Plomin, R. (2004). A genetic analysis of individual differences in dissociative behaviors in childhood and adolescence. Journal of Child Psychology and Psychiatry, 45, 522–532. Bedny, M., Pascual-Leone, A., & Saxe, R. R. (2009). Growing up blind does not change the neural bases of Theory of Mind. Proceedings of the National Academy of Science of the USA, 106, 11312–11317. Bigler, E. D. (2004). Neuropsychological results and neuropathological findings at autopsy in a case of mild traumatic brain injury. Journal of the International Neuropsychological Society, 10, 794–806. Bird, C. M., Castelli, F., Malik, O., Frith, U., & Husain, M. (2004). The impact of extensive medial frontal lobe damage on ‘Theory of mind” and cognition. Brain, 127, 914–928. Bluck, S., Alea, N., Habermas, T., & Rubin, D. (2005). A tale of three functions: The self-reported uses of autobiographical memory. Social Cognition, 23, 91–117. Boria, S., Fabbri-Destro, M., Cattaneo, L., Sparaci, L., Sinigaglia, C., Santelli, E., et al (2009). Intention understanding in autism. PLoS One, 4, e5596. Borsci, G., Boccardi, M., Rossi, R., Rossi, G., Perez, J., Bonetti, M., et al (2009). Alexithymia in healthy women: A brain morphology study. Journal of Affective Disorder, 114, 208–215. Bourget, D., & Whitehurst, L. (2007). Amnesia and crime. Journal of the American Academy of Psychiatry and the Law, 35, 469–480. Brand, M., Eggers, C., Reinhold, N., Fujiwara, E., Kessler, J., Heiss, W.-D., et al (2009). Functional brain imaging in fourteen patients with dissociative amnesia reveals right inferolateral prefrontal hypometabolism. Psychiatry Research: Neuroimaging Section, 174, 132–139. Brand, M., & Markowitsch, H. J. (2003). The principle of bottleneck structures. In R. H. Kluwe, G. Lüer, & F. Rösler (Eds.), Principles of learning and memory (pp. 171–184). Basel: Birkhäuser. Brand, M., & Markowitsch, H. J. (2008). The role of the prefrontal cortex in episodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.). Handbook of behavioral neuroscience: Episodic memory research (Vol. 18, pp. 317–342). Amsterdam: Elsevier. Brand, M., & Markowitsch, H. J. (2009). Environmental influences on autobiographical memory: The mnestic block syndrome. In L. Bäckman & L. Nyberg (Eds.), Festschrift in honor of Lars-Göran Nilsson (pp. 229–264). Hove, UK: Psychology Press. Brandt, J., & van Gorp, W. G. (2006). Functional (‘‘psychogenic”) amnesia. Seminars in Neurology, 26, 331–340. Bremner, J. D., Randall, P., Vermetten, E., Staib, L., Bronen, R. A., Mazure, C., et al (1997). Magnetic resonance imaging-based measurement of hippocampal volume in posttraumatic stress disorder related to childhood physical and sexual abuse – A preliminary report. Biological Psychiatry, 41, 23–32. Bright-Paul, A., Jarrold, C., & Wright, D. B. (2008). Theory-of-mind development influences suggestibility and source monitoring. Developmental Psychology, 44, 1055–1068. Buckner, R. L., & Carroll, D. C. (2007). Self-projection and the brain. Trends in Cognitive Sciences, 11, 49–57. Buie, D. H. (1981). Empathy: Its nature and limitations. Journal of the American Psychoanalytic Associations, 29, 281–307. Bull, R., Phillips, L. H., & Conway, C. A. (2008). The role of control functions in mentalizing: Dual-task studies of theory of mind and executive function. Cognition, 107, 663–672. Burgl, G. (1900). Eine Reise in die Schweiz im epileptischen Dämmerzustande und die transitorischen Bewusstseinsstörungen der Epileptiker vor dem Strafrichter [A journey to Switzerland done in epileptic somnolence and the transitory disturbances of consciousness before the criminal judge]. Münchener medizinische Wochenschrift, 37, 1270–1273. Cahill, L. (2006). Why sex matters for neuroscience. Nature Reviews Neuroscience, 7, 477–484. Cahill, L., Babinsky, R., Markowitsch, H. J., & McGaugh, J. L. (1995). Involvement of the amygdaloid complex in emotional memory. Nature, 377, 295–296. Campbell, S., Marriott, M., Nahmias, C., & MacQueen, G. M. (2004). Lower hippocampal volume in patients suffering from depression: A meta-analysis. American Journal of Psychiatry, 161, 598–607. Carlson, E. A. (1998). A prospective longitudinal study of attachment disorganization/disorientation. Child Development, 69, 1107–1128.
796
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
Carrington, S. J., & Bailey, A. J. (2009). Are there theory of mind regions in the brain? A review of the neuroimaging literature. Human Brain Mapping, 30, 2313–2335. Cattaneo, L., & Rizzolatti, G. (2009). The mirror neuron system. Archives of Neurology, 66, 557–560. Champagne, D. L., Bagot, R. C., van Hasselt, F., Ramakers, G., Meaney, M. J., de Kloet, E. R., et al (2008). Maternal care and hippocampal plasticity: Evidence for experience-dependent structural plasticity, altered synaptic functioning, and differential responsiveness to glucocorticoids and stress. Journal of Neuroscience, 28, 6037–6045. Clayton, N. S., & Dickinson, A. (1998). Episodic-like memory during cache recovery in scrub-jays. Nature, 395, 272–274. Colvert, E., Rutter, M., Kreppner, J., Beckett, C., Castle, J., Groothues, C., et al (2008). Do theory of mind and executive function deficits underlie the adverse outcomes associated with profound early deprivation? Findings from the English and Romanian adoptees study. Journal of Abnormal Child Psychology, 36, 1057–1068. Conway, M. A. (2009). Episodic memories. Neuropsychologia, 47, 2305–2313. Coons, P. M., & Milstein, V. (1992). Psychogenic amnesia: A clinical investigation of 25 cases. Dissociation, 5, 73–79. Corballis, M. C. (2009). Mental time travel and the shaping of language. Experimental Brain Research, 192, 553–560. Corballis, M. C. (2010). Mirror neurons and the evolution of language. Brain and Language, 112, 25–35. Corcoran, R., & Frith, C. D. (2003). Autobiographical memory and theory of mind: Evidence of a relationship in schizophrenia. Psychological Medicine, 33, 897–905. Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24, 87–185. Craig, A. D. (2009). How do you feel – Now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10, 59–70. Craig, A. D. (2010). Once an island, now the focus of attention. Brain Structure and Function, 214, 395–396. Craig, M. C., Catani, M., Deeley, Q., Latham, R., Daly, E., Kanaan, R., et al (2009). Altered connections on the road to psychopathy. Molecular Psychiatry, 14, 946–953. Damasio, A. (1999). The feeling of what happens. Body and emotion in the making of consciousness. Orlando, FL: Harcourt, Inc. Daniel, R., & Pollmann, S. (2010). Comparing the neural basis of monetary reward and cognitive feedback during information-integration category learning. Journal of Neuroscience, 30, 47–55. Darwin, C. (2004). The descent of man, and selection in relation to sex. London, UK: Penguin Books. De Kloet, E. R., & Rinne, T. (2007). Neuroendocrine markers of early trauma. In E. Vermetten, M. J. Dorahy, & D. Spiegel (Eds.), Traumatic dissociation. Neurobiology and treatment (pp. 139–156). Arlington, VA: American Psychiatric Publ. De Quervain, D. J., Kolassa, I. T., Ertl, V., Onyut, P. L., Neuner, F., Elbert, T., et al (2007). A deletion variant of the alpha2b-adrenoceptor is related to emotional memory in Europeans and Africans. Nature Neuroscience, 10, 1137–1139. De Renzi, E., Lucchelli, F., Muggia, S., & Spinnler, H. (1997). Is memory loss without anatomical damage tantamount to a psychogenic deficit? The case of pure retrograde amnesia. Neuropsychologia, 35, 781–794. Devue, C., Collette, F., Balteau, E., Degueldre, C., Luxen, A., Maquet, P., et al (2007). Here I am: The cortical correlates of visual self-recognition. Brain Research, 1143, 169–182. Di Fiorino, M. (2003). Factitious disorders: The Münchausen and Ganserian patients. Bridging Eastern and Western Psychiatry, 1, 56–65. Diehl, B., Busch, R. M., Duncan, J. S., Piao, Z., Tkach, J., & Lüders, H. O. (2008). Abnormalities in diffusion tensor imaging of the uncinate fasciculus relate to reduced memory in temporal lobe epilepsy. Epilepsia, 49, 1409–1418. Dorahy, M. J., & Huntjens, R. J. C. (2007). Memory and attentional processes in dissociative identity disorder. In E. Vermetten, M. J. Dorahy, & D. Spiegel (Eds.), Traumatic dissociation. Neurobiology and treatment (pp. 55–75). Arlington, VA: American Psychiatric Publ. DSM-IV-TR (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. Du, J., Wang, Y., Hunter, R., Wei, Y., Blumenthal, R., Falke, C., et al (2009). Dynamic regulation of mitochondrial function by glucocorticoids. Proceedings of the National Academy of Sciences of the USA, 106, 3543–3548. Ebeling, U., & von Cramon, D. (1992). Topography of the uncinate fascicle and adjacent temporal fiber tracts. Acta Neurochirurgica, 115, 143–148. Eliot, T. S. (1969). The complete poems and plays. London, UK: Faber & Faber Limited. Fink, G. R., Markowitsch, H. J., Reinkemeier, M., Bruckbauer, T., Kessler, J., & Heiss, W.-D. (1996). Cerebral representation of one’s own past: Neural networks involved in autobiographical memory. Journal of Neuroscience, 16, 4275–4282. Fouquet, M., Desgranges, B., Landeau, B., Duchesnay, E., Mezenge, F., de la Sayette, V., et al (2009). Longitudinal brain metabolic changes from amnestic mild cognitive impairment to Alzheimer’s disease. Brain, 132, 2058–2067. Freud, S. (1901). Zum psychischen Mechanismus der Vergesslichkeit [On the psychic mechanism of forgetfulness]. Monatsschrift für Psychiatrie und Neurologie, 4, 436–443. Fujiwara, E., Brand, M., Kracht, L., Kessler, J., Diebel, A., Netz, J., et al (2008). Functional retrograde amnesia: A multiple case study. Cortex, 44, 29–45. Fulton, D. (2006). Reconsolidation: Does the past linger on? Journal of Neuroscience, 26, 10935–10936. Gamer, M., & Büchel, C. (2009). Amygdala activation predicts gaze toward fearful eyes. Journal of Neuroscience, 29, 9123–9126. Ganser, S. J. (1898). Ueber einen eigenartigen hysterischen Dämmerzustand [On a peculiar hysterical state of somnolence]. Archiv für Psychiatrie und Nervenkrankheiten, 30, 633–640. Ganser, S. J. (1904). Zur Lehre vom hysterischen Dämmerzustande [On the theory of the hysterical state of somnolence]. Archiv für Psychiatrie und Nervenkrankheiten, 38, 34–46. Gilbert, D. T., & Wilson, T. D. (2007). Prospection: Experiencing the future. Science, 317, 1351–1354. Glisky, E. L., Ryan, L., Reminger, S., Hardt, O., Hayes, S. M., & Hupbach, A. (2004). A case of psychogenic fugue: I understand, aber ich verstehe nichts. Neuropsychologia, 42, 1132–1147. Govindan, R. M., Behen, M. E., Helder, E., Makki, M. I., & Chugani, H. T. (2010). Altered water diffusivity in cortical association tracts in children with early deprivation identified with tract-based spatial statistics. Cerebral Cortex, 20, 561–569. Grasby, P. M., Frith, C. D., Friston, K. J., Simpson, J., Fletcher, P. C., Frackowiak, R. S., et al (1994). A graded task approach to the functional mapping of brain areas implicated in auditory-verbal memory. Brain, 117, 1271–1282. Grimm, S., Ernst, J., Boesiger, P., Schuepbach, D., Hell, D., Boeker, H., et al (2009). Increased self-focus in major depressive disorder is related to neural abnormalities in subcortical–cortical midline structures. Human Brain Mapping, 30, 2617–2627. Hacking, I. (1995). Rewriting the soul. Multiple personality and the sciences of memory. Princeton, NJ: Princeton University Press. Haist, F. (2001). Consolidation of human memory over decades revealed by functional magnetic resonance imaging. Neuroscience, 4, 1139–1145. Han, S., & Northoff, G. (2008). Culture-sensitive neural substrates of human cognition: A transcultural neuroimaging approach. Nature Reviews Neuroscience, 9, 646–654. Happé, F., Malhi, G. S., & Checkley, S. (2001). Acquired mind-blindness following frontal lobe surgery? A single case study of impaired ‘theory of mind’ in a patient treated with stereotactic anterior capsulotomy. Neuropsychologia, 39, 83–90. Harpaz-Rotem, I., & Hirst, W. (2005). The earliest memories in individuals raised in either traditional and reformed kibbuz or outside the kibbuz. Memory, 13, 51–62. Hassabis, D., & Maguire, E. A. (2007). Deconstructing episodic memory with construction. Trends in Cognitive Sciences, 11, 299–306. Henke, K., Treyer, V., Weber, B., Nitsch, R. M., Hock, C., Wieser, H. G., et al (2003). Functional neuroimaging predicts individual memory outcome after amygdalohippocampectomy. NeuroReport, 14, 1197–1202. Hennig-Fast, K., Meister, F., Frodl, T., Beraldi, A., Padberg, F., Engel, R. R., et al (2008). A case of persistent retrograde amnesia following a dissociative fugue: Neuropsychological and neurofunctional underpinnings of loss of autobiographical memory and self-awareness. Neuropsychologia, 46, 2993–3005.
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
797
Hering, E. (1870/1895). Ueber das Gedächtnis als eine allgemeine Funktion der organisierten Materie. Vortrag gehalten in der feierlichen Sitzung der Kaiserlichen Akademie der Wissenschaften in Wien am XXX. Mai MDCCCLXX. Leipzig: Akademische Verlagsgesellschaft. (Transl.: Hering, E. (1895). Memory as a general function of organized matter. Chicago: Open Court). Highley, J. R., Walker, M. A., Esiri, M. M., Crow, T. J., & Harrison, P. J. (2002). Asymmetry of the uncinate fasciculus: A post-mortem study of normal subjects and patients with schizophrenia. Cerebral Cortex, 12, 1218–1224. Howe, M. L., & Courage, M. L. (1993). On resolving the enigma of infantile amnesia. Psychological Bulletin, 113, 305–326. Huntjens, R. J., Postma, A., Woertman, L., van der Hart, O., & Peters, M. L. (2005). Procedural memory in dissociative identity disorder: When can interidentity amnesia be truly established? Consciousness and Cognition, 14, 377–389. Hurlemann, R., Hawellek, B., Matusch, A., Kolsch, H., Wollersen, H., Madea, B., et al (2005). Noradrenergic modulation of emotion-induced forgetting and remembering. Journal of Neuroscience, 25, 6343–6349. Hyman, I. E., & Faries, J. M. (1992). The functions of autobiographical memory. In M. A. Conway, D. C. Rubin, H. Spinnler, & W. A. Wagenaar (Eds.), Theoretical perspectives on autobiographical memory (pp. 207–221). Dordrecht, Netherlands: Kluwer Academic Publishers. ICD-10 (1992). Classification of mental and behavioral disorders: Clinical descriptions and diagnostic guidelines. Geneva: World Health Organization. Ingvar, D. H. (1985). ‘‘Memory of the future”: An essay on the temporal organization of conscious awareness. Human Neurobiology, 4, 127–136. Inoue, Y., Tonooka, Y., Yamada, K., & Kanba, S. (2004). Deficiency of theory of mind in patients with remitted mood disorder. Journal of Affective Disorder, 82, 403–409. James, W. (1950). The principles of psychology (Vol. I). New York, NY: Dover Publications. Janet, P. (1907). The major symptoms of hysteria. New York: Macmillan. Janet, P. (1920). The major symptoms of hysteria: Fifteen lectures given in the Medical School of Harvard University (2nd ed.). New York: Macmillan. Joels, M., & Baram, T. Z. (2009). The neuro-symphony of stress. Nature Reviews Neuroscience, 10, 459–466. Kanzer, M. (1939). Amnesia, a statistical study. American Journal of Psychiatry, 96, 711–716. Kapur, N., Millar, J., Colbourn, C., Abbott, P., Kennedy, P., & Docherty, T. (1997). Very long-term amnesia in association with temporal lobe epilepsy: Evidence for multiple-stage consolidation processes. Brain and Cognition, 35, 58–70. Kawashima, R., Sugiura, M., Kato, T., Nakamura, A., Hatano, K., Ito, K., et al (1999). The human amygdala plays an important role in gaze monitoring. Brain, 122, 779–783. Kessler, J., Markowitsch, H. J., Huber, M., Kalbe, E., Weber-Luxenburger, G., & Kock, P. (1997). Massive and persistent anterograde amnesia in the absence of detectable brain damage: Anterograde psychogenic amnesia or gross reduction in sustained effort? Journal of Clinical and Experimental Neuropsychology, 19, 604–614. Klein, S. B., & Gangi, C. E. (2010). The multiplicity of self: Neuropsychological evidence and its implications for the self as a construct in psychological research. Annals of the New York Academy of Sciences, 1191, 1–15. Kleinman, A. (1980). Patients and healers in the context of culture. Beverley, CA: University of California Press. Kobayashi, C., Glover, G. H., & Temple, E. (2007). Cultural and linguistic effects on neural bases of ‘Theory of Mind’ in American and Japanese children. Brain Research, 1164, 95–107. Koenigs, M., Huey, E. D., Raymont, V., Cheon, B., Solomon, J., Wassermann, E. M., et al (2008). Focal brain damage protects against post-traumatic stress disorder in combat veterans. Nature Neuroscience, 11, 232–237. Kopelman, M. D. (2000). Focal retrograde amnesia and the attribution of causality: An exceptionally critical review. Cognitive Neuropsychology, 17, 585–621. Kopelman, M. D. (2008). Retrograde memory loss. In G. Goldenberg & B. Miller (Eds.). Handbook of clinical neurology (Vol. 88, pp. 185–202). Amsterdam: Elsevier. Koponen, S., Taiminen, T., Honkalampi, K., Joukamaa, M., Viinamaki, H., Kurki, T., et al (2005). Alexithymia after traumatic brain injury: Its relation to magnetic resonance imaging findings and psychiatric disorders. Psychosomatic Medicine, 67, 807–812. Krahn, L. E., Bostwick, J. M., & Stonnington, C. M. (2008). Looking toward DSM-V: Should factitious disorder become a subtype of somatoform disorder? Psychosomatics, 49, 277–282. Kritchevsky, M., Chang, J., & Squire, L. R. (2004). Functional amnesia: Clinical description and neuropsychological profile of 10 cases. Learning and Memory, 11, 213–226. Kroll, N., Markowitsch, H. J., Knight, R., & von Cramon, D. Y. (1997). Retrieval of old memories – The temporo-frontal hypothesis. Brain, 120, 1377–1399. Kubicki, M., McCarley, R., Westin, C. F., Park, H. J., Maier, S., Kikinis, R., et al (2007). A review of diffusion tensor imaging studies in schizophrenia. Journal of Psychiatric Research, 41, 15–30. Kühnel, S., Woermann, F. G., Mertens, M., & Markowitsch, H. J. (2008). Involvement of the orbitofrontal cortex during correct and false recognitions of visual stimuli. Implications for eyewitness decisions on an fMRI study using a film paradigm. Brain Imaging and Behavior, 2, 163–176. Kumar, S., Rao, S. L., Sunny, B., & Gangadhar, B. N. (2007). Widespread cognitive impairment in psychogenic anterograde amnesia. Psychiatry and Clinical Neuroscience, 61, 583–586. Kundera, M. (1991). The unbearable lightness of being. New York, NY: Harper Perennial. LaBar, K. S., & Cabeza, R. (2006). Cognitive neuroscience of emotional memory. Nature Reviews Neuroscience, 7, 54–64. Lanius, R. A., Hopper, J. W., & Menon, R. S. (2003). Individual differences in a husband and wife who developed PTSD after a motor vehicle accident: A functional MRI case study. American Journal of Psychiatry, 160, 667–669. Lanius, R. A., Vermetten, E., Loewenstein, R. J., Brand, B., Schmahl, C., Bremner, J. D., et al (2010). Emotion modulation in PTSD: Clinical and neurobiological evidence for a dissociative subtype. American Journal of Psychiatry, 167, 640–647. Lanius, R. A., Williamson, P. C., Bluhm, R. L., Densmore, M., Boksman, K., Neufeld, R. W., et al (2005). Functional connectivity of dissociative responses in posttraumatic stress disorder: A functional magnetic resonance imaging investigation. Biological Psychiatry, 57, 873–884. Leary, M. R. (2007). Motivational and emotional aspects of the self. Annual Reviews of Psychology, 58, 317–344. Lebel, C., Walker, L., Leemans, A., Phillips, L., & Beaulieu, C. (2008). Microstructural maturation of the brain from childhood to adulthood. NeuroImage, 20, 1044–1055. Legrand, D., & Ruby, P. (2009). What is self-specific? Theoretical investigation and critical review of neuroimaging results. Psychological Review, 116, 252–282. Lemogne, C., Bergouignan, L., Piolino, P., Jouvent, R., Allilaire, J. F., & Fossati, P. (2009). Cognitive avoidance of intrusive memories and autobiographical memory: Specificity, autonoetic consciousness, and self-perspective. Memory, 17, 1–7. Lemogne, C., Piolino, P., Friszer, S., Claret, A., Girault, N., Jouvent, R., et al (2006). Episodic autobiographical memory in depression: Specificity, autonoetic consciousness, and self perspective. Consciousness and Cognition, 15, 258–268. Lennox, W. G. (1943). Amnesia, real and feigned. American Journal of Psychiatry, 99, 732–743. Levine, B., Black, S. E., Cabeza, R., Sinden, M., McIntosh, A. R., Toth, J. P., et al (1998). Episodic memory and the self in a case of isolated retrograde amnesia. Brain, 121, 1951–1973. Levine, B., Svoboda, E., Turner, G. R., Mandic, M., & Mackey, A. (2009). Behavioral and functional neuroanatomical correlates of anterograde autobiographical memory in isolated retrograde amnesic patient M.L. Neuropsychologia, 47, 2188–2196. Lind, S. E., & Bowler, S. M. (2009). Recognition memory, self-other source memory, and theory-of-mind in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 39, 1231–1239. Lipton, M. L., Gellella, E., Lo, C., Gold, T., Ardekani, B. A., Shifteh, K., et al (2008). Multifocal white matter ultrastructural abnormalities in mild traumatic brain injury with cognitive disability: A voxel-wise analysis of diffusion tensor imaging. Journal of Neurotrauma, 25, 1335–1342. Lipton, M. L., Gulko, E., Zimmerman, M. E., Friedman, B. W., Kim, M., Gelella, E., et al (2009). Diffusion-tensor imaging implicates prefrontal axonal injury in executive function impairment following very mild traumatic brain injury. Neurology, 252, 816–824.
798
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
Loewenstein, R. J. (1991). Psychogenic amnesia and psychogenic fugue: A comprehensive review. American Psychiatric Press Review of Psychiatry, 10, 189–222. Loftus, E. F. (2005). Planting misinformation in the human mind: A 30-year investigation of the malleability of memory. Learning and Memory, 12, 361–366. Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nature Reviews Neuroscience, 10, 434–445. Luria, A. R. (1987). The mind of a mnemonist. A little book about a vast memory. New York: Harvard University Press. Maldonado, J. R., Page, K., Koopman, C., Butler, L. D., Stein, H., & Spiegel, D. (2002). Acute stress reactions following the assassination of Mexican presidential candidate Colosio. Journal of Traumatic Stress, 15, 401–405. Maldonado, J. R., & Spiegel, D. (2008). Dissociative disorders. In E. Hales, S. C. Yudofsky, & G. O. Gabbard (Eds.), The American psychiatric publishing textbook of psychiatry (5th ed., pp. 665–710). Arlington, VA: American Psychiatric Publ. Markowitsch, H. J. (1984). Can amnesia be caused by damage of a single brain structure? Cortex, 20, 27–45. Markowitsch, H. J. (1988). Diencephalic amnesia: A reorientation towards tracts? Brain Research Reviews, 13, 351–370. Markowitsch, H. J. (1992). Intellectual functions and the brain. An historical perspective. Toronto: Hogrefe & Huber Publ. Markowitsch, H. J. (1994). The memory storehouse. Trends in Neuroscience, 17, 513. Markowitsch, H. J. (1995). Which brain regions are critically involved in the retrieval of old autobiographic memory? Brain Research Reviews, 21, 117–127. Markowitsch, H. J. (1996). Organic and psychogenic retrograde amnesia: Two sides of the same coin? Neurocase, 2, 357–371. Markowitsch, H. J. (1999a). Neuroimaging and mechanisms of brain function in psychiatric disorders. Current Opinion in Psychiatry, 12, 331–337. Markowitsch, H. J. (1999b). Functional neuroimaging correlates of functional amnesia. Memory, 7, 561–583. Markowitsch, H. J. (2000). Neuroanatomy of memory. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 465–484). New York: Oxford University Press. Markowitsch, H. J. (2003). Autonoëtic consciousness. In A. S. David & T. Kircher (Eds.), The self in neuroscience and psychiatry (pp. 180–196). Cambridge: Cambridge University Press. Markowitsch, H. J. (2005). Time, memory, and consciousness. A view from the brain. In R. Buccheri, A. C. Elitzur, & M. Saniga (Eds.), Endophysics, time, quantum, and the subjective (pp. 131–147). Singapore: World Scientific Publ. Markowitsch, H. J. (2008). Anterograde amnesia. In G. Goldenberg & B. L. Miller (Eds.). Handbook of clinical neurology (2nd Series, Vol. 88, pp. 155–183). New York: Elsevier. Markowitsch, H. J., & Brand, M. (2010). Forgetting – An historical perspective. In S. Della Sala (Ed.), Forgetting (pp. 23–34). Hove, UK: Psychology Press. Markowitsch, H. J., Calabrese, P., Fink, G. R., Durwen, H. F., Kessler, J., Härting, C., et al (1997). Impaired episodic memory retrieval in a case of probable psychogenic amnesia. Psychiatry Research: Neuroimaging Section, 74, 119–126. Markowitsch, H. J., Calabrese, P., Haupts, M., Durwen, H. F., Liess, J., & Gehlen, W. (1993). Searching for the anatomical basis of retrograde amnesia. Journal of Clinical and Experimental Neuropsychology, 15, 947–967. Markowitsch, H. J., Calabrese, P., Neufeld, H., Gehlen, W., & Durwen, H. F. (1999). Retrograde amnesia for world knowledge and preserved memory for autobiographic events. A case report. Cortex, 35, 243–252. Markowitsch, H. J., Calabrese, P., Würker, M., Durwen, H. F., Kessler, J., Babinsky, R., et al (1994). The amygdala’s contribution to memory – A PET-study on two patients with Urbach–Wiethe disease. NeuroReport, 5, 1349–1352. Markowitsch, H. J., Emmans, D., Irle, E., Streicher, M., & Preilowski, B. (1985). Cortical and subcortical afferent connections of the primate’s temporal pole: A study of rhesus monkeys, squirrel monkeys, and marmosets. Journal of Comparative Neurology, 242, 425–458. Markowitsch, H. J., Fink, G. R., Thöne, A. I. M., Kessler, J., & Heiss, W.-D. (1997). Persistent psychogenic amnesia with a PET-proven organic basis. Cognitive Neuropsychiatry, 2, 135–158. Markowitsch, H. J., Kessler, J., Kalbe, E., & Herholz, K. (1999). Functional amnesia and memory consolidation. A case of persistent anterograde amnesia with rapid forgetting following whiplash injury. Neurocase, 5, 189–200. Markowitsch, H. J., Kessler, J., Russ, M. O., Frölich, L., Schneider, B., & Maurer, K. (1999). Mnestic block syndrome. Cortex, 35, 219–230. Markowitsch, H. J., Kessler, J., Van der Ven, C., Weber-Luxenburger, G., & Heiss, W.-D. (1998). Psychic trauma causing grossly reduced brain metabolism and cognitive deterioration. Neuropsychologia, 36, 77–82. Markowitsch, H. J., Kessler, J., Weber-Luxenburger, G., Van der Ven, C., Albers, M., & Heiss, W. D. (2000). Neuroimaging and behavioral correlates of recovery from mnestic block syndrome and other cognitive deteriorations. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 13, 60–66. Markowitsch, H. J., Thiel, A., Kessler, J., von Stockhausen, H.-M., & Heiss, W.-D. (1997). Ecphorizing semi-conscious episodic information via the right temporopolar cortex – A PET study. Neurocase, 3, 445–449. Markowitsch, H. J., Vandekerckhove, M. M. P., Lanfermann, H., & Russ, M. O. (2003). Engagement of lateral and medial prefrontal areas in the ecphory of sad and happy autobiographical memories. Cortex, 39, 643–665. Markowitsch, H. J., von Cramon, D. Y., Hofmann, E., Sick, C.-D., & Kinzler, P. (1990). Verbal memory deterioration after unilateral infarct of the internal capsule in an adolescent. Cortex, 26, 597–609. Markowitsch, H. J., von Cramon, D. Y., & Schuri, U. (1993). Mnestic performance profile of a bilateral diencephalic infarct patient with preserved intelligence and severe amnesic disturbances. Journal of Clinical and Experimental Neuropsychology, 15, 627–652. Markowitsch, H. J., & Welzer, H. (2010). The development of autobiographical memory. Hove, UK: Psychology Press. Masson, J. M. (Ed.). (1985). The complete letters of Sigmund Freud to Wilhelm Fliess 1887–1904. Cambridge, MA: The Belknap Press of Harvard University Press. Mattila, A. K., Saarni, S. I., Salminen, J. K., Huhtala, H., Sintonen, H., & Joukamaa, M. (2009). Alexithymia and health-related quality of life in a general population. Psychosomatics, 50, 59–68. Mayou, R., Kirmayer, L. J., Simon, G., Kroenke, K., & Sharpe, M. (2005). Somatoform disorders: Time for a new approach in DSM-V. American Journal of Psychiatry, 162, 847–855. McCullough, J. P. Jr., (2000). Treatment for chronic depression. Cognitive behavioural analysis system of psychotherapy. New York, NY: Guilford Press. McGowan, P. O., Sasaki, A., D’Alessio, A. C., Dymov, S., Labonte, B., Szyf, M., et al (2009). Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nature Neuroscience, 12, 342–348. McKay, G. C. M., & Kopelman, M. D. (2009). Psychogenic amnesia: When memory complaints are medically unexplained. Advances in Psychiatric Treatment, 15, 152–158. Mehta, M. A., Golembo, N. I., Nosarti, C., Colvert, E., Mota, A., Williams, S. C., et al (2009). Amygdala, hippocampal and corpus callosum size following severe early institutional deprivation: The English and Romanian adoptees study pilot. Journal of Child Psychology and Psychiatry, 50, 943–951. Metzinger, T. (2008). Empirical perspectives from the self-model theory of subjectivity: A brief summary with examples. Progress in Brain Research, 168, 215–245. Miller, G. A. (1956). The magical number seven plus minus two. Some limits on our capacity for processing information. Psychological Review, 63, 244–257. Miller, C. A., & Sweatt, J. D. (2006). Amnesia or retrieval deficit? Implications of a molecular approach to the question of reconsolidation. Learning and Memory, 13, 498–505. Mishkin, M., & Petri, H. L. (1984). Memories and habits: Some implications for the analysis of learning and retention. In L. R. Squire & N. Butters (Eds.), Neuropsychology of memory (pp. 287–296). New York: Guilford Press. Mitchell, S. A., & Black, M. J. (1995). Freud and beyond. A history of modern psychoanalytic thought. New York: Basic Books. Mitchell, J. P., Macrae, C. M., & Banaji, M. R. (2006). Dissociable medial prefrontal contributions to judgments of similar and dissimilar others. Neuron, 50, 655–663. Modell, A. H. (2006). Imagination and the meaningful brain. Cambridge, MA: Bradford. Modestin, J., Lotscher, K., & Erni, T. (2002). Dissociative experiences and their correlates in young non-patients. Psychology and Psychotherapy, 75, 53–64.
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
799
Moldoveanu, M., & Stevenson, H. (2001). The self as a problem: The intra-personal coordination of conflicting desires. Journal of Socio-Economics, 30, 295–330. Mondadori, C. R., Buchmann, A., Mustovic, H., Schmidt, C. F., Boesiger, P., Nitsch, R. M., et al (2006). Enhanced brain activity may precede the diagnosis of Alzheimer’s disease by 30 years. Brain, 129, 2908–2922. Mooney, G., & Speed, J. (2001). The association between mild traumatic brain injury and psychiatric conditions. Brain injury, 15, 865–877. Moriguchi, Y., Ohnishi, T., Decety, J., Hirakata, M., Maeda, M., Matsuda, H., et al (2009). The human mirror neuron system in a population with deficient selfawareness: An fMRI study in alexithymia. Human Brain Mapping, 30, 2063–2076. Moriguchi, Y., Ohnishi, T., Lane, R. D., Maeda, M., Mori, T., Nemoto, K., et al (2006). Impaired self-awareness and theory of mind: An fMRI study of mentalizing in alexithymia. NeuroImage, 32, 1472–1482. Morin, A. (2009). Self-awareness deficits following loss of inner speech: Dr. Jill Bolte Taylor’s case study. Consciousness and Cognition, 18, 524–529. Morin, A., & Michaud, J. (2007). Self-awareness and the left inferior frontal gyrus: Inner speech use during self-related processing. Brain Research Bulletin, 74, 387–396. Nagel, T. (1974). What is it like to be a bat? Philosophical Review, 82, 435–456. Naqvi, N. H., Rudrauf, D., Damasio, H., & Bechara, A. (2007). Damage to the insula disrupts addiction to cigarette smoking. Science, 315, 531–534. Nelson, K. (2003). Narrative and self, myth and memory: Emergence of the cultural self. In R. Fivush & C. Haden (Eds.), The development of autobiographical memory and self-understanding (pp. 3–28). Hillsdale, NJ: Erlbaum Assoc. Nelson, K. (2005). Evolution and development of human memory systems. In B. J. Ellis & D. F. Bjorklund (Eds.), Origins of the social mind (pp. 319–345). New York: Guilford Press. Nelson, K., & Fivush, R. (2004). The emergence of autobiographical memory: A social cultural developmental theory. Psychological Review, 111, 486–511. Newen, A., & Schlicht, T. (2009). Understanding other minds: A criticism of Goldman’s simulation theory and an outline of the person model theory. Grazer Philosophische Studien, 79, 209–242. Newen, A., & Vogeley, K. (2003). Self-representation: Searching for a neural signature of self-consciousness. Consciousness and Cognition, 12, 529–543. Nieuwenhuys, R., Voogd, J., & van Huijzen, C. (1978). The human central nervous system. Berlin: Springer. O’Brien, J. T. (1997). The ‘glucocorticoid cascade’ hypothesis in man. British Journal of Psychiatry, 170, 199–201. O’Connor, M., Sieggreen, M. A., Ahern, G., Schomer, D., & Mesulam, M. (1997). Accelerated forgetting in association with temporal lobe epilepsy and paraneoplastic encephalitis. Brain and Cognition, 35, 71–84. Oddo, S., Lux, S., Weiss, P. H., Schwab, A., Welzer, H., Markowitsch, H. J., et al (2010). Specific role of medial prefrontal cortex in retrieving recent autobiographical memories: An fMRI study of young female subjects. Cortex, 46, 29–39. Ouellet, J., Rouleau, I., Labrecque, R., Bernier, G., & Scherzer, P. B. (2008). Two routes to losing one’s past life: A brain trauma, an emotional trauma. Behavioural Neurology, 20, 27–38. Pacheco, J., Beevers, C. G., Benavides, C., McGeary, J., Stice, E., & Schnyer, D. M. (2009). Frontal-limbic white matter pathway associations with the serotonin transporter gene promoter region (5-HTTLPR) polymorphism. Journal of Neuroscience, 29, 6229–6233. Parker, E. S., Cahill, L., & McGaugh, J. L. (2006). A case of unusual autobiographical remembering. Neurocase, 12, 35–49. Paus, T. (2010). Growth of white matter in the adolescent brain: Myelin or axon? Brain and Cognition, 72, 26–35. Pearce, J. M. (2007). Amnesia. European Neurology, 57, 126. Peigneux, P., Schmitz, R., & Urbain, C. (2010). Sleep and forgetting. In S. Della Sala (Ed.), Forgetting (pp. 165–184). Hove, East Sussex: Psychology Press. Perner, J. (2000). Memory and theory of mind. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 297–314). New York: Oxford University Press. Perner, J. (2001). Episodic memory: Essential distinctions and developmental implications. In C. Moore & K. Lemmon (Eds.), The self in time: Developmental perspectives (pp. 181–202). Hillsdale, NJ: Erlbaum. Pessoa, L. (2008). On the relationship between emotion and cognition. Nature Reviews Neuroscience, 9, 148–158. Peters, J., & Büchel, C. (2010). Episodic future thinking reduces rewards delay discounting through an enhancement of prefrontal–mediotemporal interactions. Neuron, 66, 138–148. Pfeifer, J. H., Dapretto, M., & Lieberman, M. D. (2010). The neural foundations of evaluative self-knowledge in middle childhood, early adolescence and adulthood. In P. D. Zelazo, M. Chandler, & E. Crone (Eds.), Developmental social cognitive neuroscience (pp. 141–163). New York, NY: Psychology Press. Phillips, M. L., Drevets, W. C., Rauch, S. L., & Lane, R. (2003). Neurobiology of emotion perception. II: Implications for major psychiatric disorders. Biological Psychiatry, 54, 515–528. Piefke, M., Weiss, P. H., Markowitsch, H. J., & Fink, G. R. (2005). Gender differences in the functional neuroanatomy of emotional episodic autobiographical memory. Human Brain Mapping, 24, 313–324. Piolino, P., Desgranges, B., & Eustache, F. (2009). Episodic autobiographical memories over the course of time: Cognitive, neuropsychological and neuroimaging findings. Neuropsychologia, 47, 2314–2329. Piolino, P., Hannequin, D., Desgranges, B., Girard, C., Beaunieux, H., Gittard, B., et al (2005). Right ventral frontal hypometabolism and abnormal sense of self in a case of disproportionate retrograde amnesia. Cognitive Neuropsychology, 22, 1005–1034. Piolino, P., Hisland, M., Ruffeveille, I., Matuszewski, V., Jambaque, I., & Eustache, F. (2007). Do school-age children remember or know the personal past? Consciousness and Cognition, 16, 84–101. Pizzagalli, D. A., Holmes, A. J., Dillon, D. G., Goetz, E. L., Birk, J. L., Bogdan, R., et al (2009). Reduced caudate and nucleus accumbens response to rewards in unmedicated individuals with major depressive disorder. American Journal of Psychiatry, 166, 702–710. Plaza, M., Gatignol, P., Leroy, M., & Duffau, H. (2009). Speaking without Broca’s area after tumor resection. Neurocase, 15, 284–310. Post, R. M., Weiss, S. R., Smith, M., Rosen, J., & Frye, M. (1995). Stress, conditioning, and the temporal aspects of affective disorders. Annals of the New York Academy of Sciences, 771, 677–696. Povinelli, D. J., Landau, K. R., & Perilloux, H. K. (1996). Self-recognition in young children using delayed versus live feedback: Evidence of a developmental asynchrony. Child Development, 67, 1540–1554. Pugliese, L., Catani, M., Ameis, S., Dell’Acqua, F., Thiebaut de Schotten, M., et al (2009). The anatomy of extended limbic pathways in Asperger syndrome: A preliminary diffusion tensor imaging tractography study. NeuroImage, 47, 427–434. Putnam, F. (1997). Dissociation in children and adolescents. A developmental perspective. New York: Guilford Press. Rabin, J. S., Gilboa, A., Stuss, D., Mar, R. A., & Rosenbaum, R. S. (2010). Common and unique neural correlates of autobiographical memory and theory of mind. Journal of Cognitive Neuroscience, 22, 1095–1111. Rao, S. M., Mayer, A. R., & Harrington, D. L. (2001). The evolution of brain activation during temporal processing. Nature Neuroscience, 4, 317–323. Rasch, B., Spalek, K., Buholzer, S., Luechinger, R., Boesiger, P., Papassotiropoulos, A., et al (2009). A genetic variation of the noradrenergic system is related to differential amygdala activation during encoding of emotional memories. Proceedings of the National Academy of Sciences of the USA, 105, 19191–19196. Rathbone, C. J., Moulin, C. J., & Conway, M. A. (2009). Autobiographical memory and amnesia: Using conceptual knowledge to ground the self. Neurocase, 15, 405–418. Ray, R. D., Shelton, A. L., Hollon, N. G., Matsumoto, D., Frankel, C. B., Gross, J. J., et al (2010). Interdependent self-construal and neural representations of self and mother. Social Cognitive and Affective Neuroscience, 5, 318–323. Reinhold, N., & Markowitsch, H. J. (2008). Theory of mind and the sense of self. Science Online [Letter to the editor]. Reinhold, N., Kühnel, S., Brand, M., & Markowitsch, H. J. (2006). Functional neuroimaging in memory and memory disturbances. Current Medical Imaging Reviews, 2, 35–57. Reinhold, N., & Markowitsch, H. J. (2007). Emotion and consciousness in adolescent psychogenic amnesia. Journal of Neuropsychology, 1, 53–64.
800
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
Reinhold, N., & Markowitsch, H. J. (2009). Retrograde episodic memory and emotion: A perspective from patients with dissociative amnesia. Neuropsychologia, 47, 2197–2206. Rizzolatti, G., Fabbri-Destro, M., & Cattaneo, L. (2009). Mirror neurons and their clinical relevance. Nature clinical practice. Neurology, 5, 24–34. Robertson, D., Snarey, J., Ousley, O., Harenski, K., DuBois Bowman, F., Gilkey, R., et al (2007). The neural processing of moral sensitivity to issues of justice and care. Neuropsychologia, 45, 755–766. Romei, V., De Gennaro, L., Fratello, F., Curcio, G., Ferrara, M., Pascual-Leone, A., et al (2008). Interhemispheric transfer deficit in alexithymia: A transcranial magnetic stimulation study. Psychotherapy and Psychosomatics, 77, 175–181. Roozendaal, B., McEwen, B. S., & Chattarji, S. (2009). Stress, memory and the amygdala. Nature Reviews Neuroscience, 10, 429–433. Rosenbaum, R. S., Gilboa, A., Levine, B., Winocur, G., & Moscovitch, M. (2009). Amnesia as an impairment of detail generation and binding: Evidence from personal, fictional, and semantic narratives in K.C. Neuropsychologia, 47, 2181–2187. Rosenbaum, R. S., Stuss, D. T., Levine, B., & Tulving, E. (2007). Theory of mind is independent of episodic memory. Science, 318, 1257. Roth, T. I., & Sweatt, J. D. (2009). Regulation of chromatin structure in memory formation. Current Opinion in Neurobiology, 19, 336–342. Ruff, R. M., Iverson, G. L., Barth, J. T., Bush, S. S., & Broshek, D. K. (2009). Recommendations for diagnosing a mild traumatic brain injury: A National Academy of Neuropsychology education paper. Archives of Clinical Neuropsychology, 24, 3–10. Rutter, M., Moffitt, T. E., & Caspi, A. (2006). Gene-environment interplay and psychopathology: Multiple varieties but real effects. Journal of Childhood Psychology and Psychiatry and Allied Disciplines, 47, 226–261. Saxe, R., Moran, J. M., Scholz, J., & Gabrieli, J. (2006). Overlapping and non-overlapping brain regions for theory of mind and self reflection in individual subjects. Social Cognitive and Affective Neuroscience, 1, 229–234. Schacter, D. L., & Addis, D. R. (2009). On the nature of medial temporal lobe contributions to the constructive simulation of future events. Philosophical Transactions of the Royal Society, London B, 364, 1245–1253. Schilbach, L., Wohlschlaeger, A. M., Kraemer, N. C., Newen, A., Shah, N. J., Fink, G. R., et al (2006). Being with virtual others: Neural correlates of social interaction. Neuropsychologia, 44, 718–730. Schneider, K. (1912). Über einige klinisch-psychologische Untersuchungsmethoden und ihre Ergebnisse. [On some clinical–pathological methods of investigation and their results]. Zeitschrift für die gesamte Neurologie und Psychiatrie, 8, 553–615. Schneider, K. (1928). Die Störungen des Gedächtnisses [Disturbances of memory]. In O. Bumke (Ed.). Handbuch der Geisteskrankheiten (Vol. 1, pp. 508–529). Berlin: Springer. Schore, A. N. (2002). Dysregulation of the right brain: A fundamental mechanism of traumatic attachment and the psychopathogenesis of posttraumatic stress disorder. Australian and New Zealand Journal of Psychiatry, 36, 9–30. Schulte-Rüther, M., Markowitsch, H. J., Fink, G. R., & Piefke, M. (2007). Mirror neuron and theory of mind mechanisms involved in face-to-face interactions: An fMRI approach to empathy. Journal of Cognitive Neuroscience, 19, 1354–1372. Schulte-Rüther, M., Markowitsch, H. J., Shah, N. J., Fink, G. R., & Piefke, M. (2008). Gender differences in the functional neuroanatomy of emotional perspective taking. NeuroImage, 14, 393–403. Seeley, W. W. (2008). Selective functional, regional, and neuronal vulnerability in frontotemporal dementia. Current Opinion in Neurology, 21, 701–707. Seligman, R., & Kirmayer, L. J. (2008). Dissociative experience and cultural neuroscience: Narrative, metaphor and mechanism. Culture, Medicine and Psychiatry, 32, 31–64. Semon, R. (1904). Die Mneme als erhaltendes Prinzip im Wechsel des organischen Geschehens. Leipzig: Wilhelm Engelmann. Sepulcre, J., Masdeu, J. C., Sastre-Garriga, J., Goni, J., Velez-de-Mendizabal, N., Duque, B., et al (2008). Mapping the brain pathways of declarative verbal memory: Evidence from white matter lesions in the living human brain. NeuroImage, 42, 1237–1243. Serra, L., Fadda, L., Buccione, I., Caltagirone, C., & Carlesimo, G. A. (2007). Psychogenic and organic amnesia: A multidimensional assessment of clinical, neuroradiological, neuropsychological and psychopathological features. Behavioural Neurology, 18, 53–64. Shalom, D. B. (2003). Memory in autism: Review and synthesis. Cortex, 39, 1129–1138. Shamay-Tsoory, S. G., & Aharon-Peretz, J. (2007). Dissociable prefrontal networks for cognitive and affective theory of mind: A lesion study. Neuropsychologia, 45, 3054–3067. Shamay-Tsoory, S. G., Aharon-Peretz, J., & Perry, D. (2009). Two systems for empathy: A double dissociation between emotional and cognitive empathy in inferior frontal gyrus versus ventromedial prefrontal lesions. Brain, 132, 617–627. Shamay-Tsoory, S., Harari, H., Szepsenwol, O., & Levkovitz, Y. (2009). Neuropsychological evidence of impaired cognitive empathy in euthymic bipolar disorder. Journal of Neuropsychiatry and Clinical Neurosciences, 21, 59–67. Sharot, T., Riccardi, A. M., Raio, C. M., & Phelps, E. A. (2007). Neural mechanisms mediating optimism bias. Nature, 450, 102–105. Shaw, P., Lawrence, E. J., Radbourne, C., Bramham, J., Polkey, C. E., & David, A. S. (2004). The impact of early and late damage to the human amygdala on ‘theory of mind’ reasoning. Brain, 127, 1535–1548. Siebert, M., Markowitsch, H. J., & Bartel, P. (2003). Amygdala, affect, and cognition: Evidence from ten patients with Urbach–Wiethe disease. Brain, 126, 2627–2637. Siegel, D. J. (2004). Memory: An overview, with emphasis on developmental, interpersonal, and neurobiological aspects. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 997–1011. Smith, C. N., Frascino, J. C., Kripke, D. L., McHugh, P. R., Treisman, G. J., & Squire, L. R. (2010). Losing memories overnight: A unique form of human amnesia. Neuropsychologia (Epub ahead of print). Spiegel, D. (2006). Recognizing traumatic dissociation. American Journal of Psychiatry, 163, 566–568. Spitzer, C., & Freyberger, H. J. (2008). Geschlechtsunterschiede bei dissoziativen Störungen. Bundesgesundheitsblatt Gesundheitsforschung, Gesundheitsschutz, 51, 46–52. Spreng, R. N., & Grady, C. L. (2010). Patterns of brain activity supporting autobiographical memory, prospection, and theory of mind, and their relationship to the default mode network. Journal of Cognitive Neuroscience, 22, 1112–1123. Spreng, R. N., Mar, R. A., & Kim, A. S. N. (2008). The common neural basis of autobiographical memory, prospection, navigation, theory of mind and the default mode: A quantitative meta-analysis. Journal of Cognitive Neuroscience, 21, 489–510. St Jacques, P. L., Bessette-Symons, B., & Cabeza, R. (2009). Functional neuroimaging studies of aging and emotion: Fronto-amygdalar differences during emotional perception and episodic memory. Journal of the International Neuropsychological Society, 15, 819–825. Staniloiu, A., & Markowitsch, H. J. (2010). Looking at comorbidity through the glasses of neuroscientific memory research: A brain network perspective. Behavioral and Brain Sciences, 33, 170–171. Staniloiu, A., & Markowitsch, H. J. (in press). Searching for the anatomy of dissociative amnesia. Journal of Psychology. Staniloiu, A., Bender, A., Smolewska, K., Ellis, J., Abramowitz, C., & Markowitsch, H. J. (2009). Ganser syndrome with work-related onset in a patient with a background of immigration. Cognitive Neuropsychiatry, 14, 180–198. Staniloiu, A., Borsutzky, S., & Markowitsch, H. J. (2010). Dissociative memory disorders and immigration. In W. Christensen, E. Schier, & J. Sutton (Eds.), ASCS09: Proceedings of the 9th conference of the Australasian Society for Cognitive Science (pp. 316–324). Sydney: Macquarie Centre for Cognitive Science. Steinberg, M. (2000). Dissociative amnesia. In B. J. Sadock & V. A. Sadock (Eds.). Comprehensive textbook of psychiatry (7th ed., Vol. 1, pp. 1544–1549). Philadelphia, PA: Lippincott, Williams & Wilkins. Stickgold, R., & Walker, M. P. (2005). Memory consolidation and reconsolidation: What is the role of sleep? Trends in Neurosciences, 28, 408–415. Sturm, V. E., Rosen, H. J., Allison, S., Miller, B. L., & Levenson, R. W. (2006). Self-conscious emotion deficits in frontotemporal lobar degeneration. Brain, 129, 2508–2516. Suddendorf, T., Addis, D. R., & Corballis, M. C. (2009). Mental time travel and the shaping of the human mind. Philosophical Transactions of the Royal Society London B, 364, 1317–1324.
A. Staniloiu et al. / Consciousness and Cognition 19 (2010) 778–801
801
Taylor, W. D., Macfall, J. R., Gerig, G., & Krishnan, R. R. (2007). Structural integrity of the uncinate fasciculus in geriatric depression: Relationship with age of onset. Neuropsychiatric Disease and Treatment, 3, 669–674. Thomas-Anterion, C., Guedj, E., Decousus, M., & Laurent, B. (2010). Can we see personal identity loss? A functional imaging study of typical ‘hysterical amnesia’. Journal of Neurology, Neurosurgery, and Psychiatry, 81, 468–469. Thomas-Antérion, C., Mazzola, L., Foyatier-Michel, N., & Laurent, B. (2008). À la recherche de la mémoire perdue: Nature des troubles et mode de récupération d’un cas d’amnésie rétrograde pure. Revue Neurologique, 164, 271–277. Tomasello, M. (2008). Origins of human communication. Cambridge, MA: MIT Press. Tramoni, E., Aubert-Khalfa, S., Guye, M., Ranjeva, J. P., Felician, O., & Ceccaldi, M. (2009). Hypo-retrieval and hyper-suppression mechanisms in functional amnesia. Neuropsychologia, 47, 611–624. Tranel, D., & Bechara, A. (2009). Sex-related functional asymmetry of the amygdala: Preliminary evidence using a case-matched lesion approach. Neurocase, 15, 1–18. 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: Clarendon Press. Tulving, E. (1985). Memory and consciousness. Canadian Psychology, 26, 1–12. Tulving, E. (1995). Organization of memory: Quo vadis? In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 839–847). Cambridge, MA: MIT Press. Tulving, E. (2000). Concepts of memory. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 33–43). Oxford: Oxford University Press. Tulving, E. (2002). Episodic memory: From mind to brain. Annual Reviews of Psychology, 53, 1–25. Tulving, E. (2005). Episodic memory and autonoesis: Uniquely human? In H. Terrace & J. Metcalfe (Eds.), The missing link in cognition: Evolution of selfknowing consciousness (pp. 3–56). New York: Oxford University Press. Tulving, E., & Kim, A. (2007). The medium and the message of mental time travel. Behavioral and Brain Sciences, 30, 334–335. Ulrich-Lai, Y. M., & Herman, J. P. (2009). Neural regulation of endocrine and autonomic stress responses. Nature Reviews Neuroscience, 10, 397–409. Valentino, K., Toth, S. L., & Cicchetti, D. (2009). Autobiographical memory functioning among abused, neglected, and nonmaltreated children: The overgeneral memory effect. Journal of Child Psychology and Psychiatry, 50, 1029–1038. Van der Hart, O., & Dorahy, M. (2006). Pierre Janet and the concept of dissociation. American Journal of Psychiatry, 163, 1646 [Letter to the editor]. van der Hart, O., & Nijenhuis, E. (2001). Generalized dissociative amnesia: Episodic, semantic and procedural memories lost and found. The Australian and New Zealand Journal of Psychiatry, 35, 589–600. Vann, S. D. (2009). Gudden’s ventral tegmental nucleus is vital for memory: Re-evaluating diencephalic inputs for amnesia. Brain, 132, 2372–2384. Vargha-Khadem, F., Gadian, D. G., Watkins, K. E., Connelly, A., Van Paesschen, W., & Mishkin, M. (1997). Differential effects of early hippocampal pathology on episodic and semantic memory. Science, 277, 376–380. Vercammen, A., Knegtering, H., Bruggeman, R., & Aleman, A. (2010). Subjective loudness and reality of auditory verbal hallucinations and activation of the inner speech processing network. Schizophrenia Bulletin, 8, 4–23. Vermetten, E., Schmahl, C., Lindner, S., Loewenstein, R. J., & Bremner, J. D. (2006). Hippocampal and amygdalar volumes in dissociative identity disorder. American Journal of Psychiatry, 163, 630–636. Vogeley, K., & Kupke, C. (2007). Disturbances of time consciousness from a phenomenological and a neuroscientific perspective. Schizophrenia Bulletin, 33, 157–165. Vuilleumier, P., & Pourtois, G. (2007). Distributed and interactive brain mechanisms during emotion face perception: Evidence from functional neuroimaging. Neuropsychologia, 45, 174–194. Vythilingam, M., Heim, C., Newport, J., Miller, A. H., Anderson, E., Bronen, R., et al (2002). Childhood trauma associated with smaller hippocampal volume in women with major depression. American Journal of Psychiatry, 159, 2072–2080. Wang, Q. (2001). Culture effects on adults’ earliest childhood recollection and self-description: Implications for the relation between memory and the self. Journal of Personality and Social Psychology, 81, 220–233. Wang, F., Kalmar, J. H., He, Y., Jackowski, M., Chepenik, L. G., Edmiston, E. E., et al (2009). Functional and structural connectivity between the perigenual anterior cingulate and amygdala in bipolar disorder. Biological Psychiatry, 66, 516–521. Weaver, I. C., Champagne, F. A., Brown, S. E., Dymov, S., Sharma, S., Meaney, M. J., et al (2005). Reversal of maternal programming of stress responses in adult offspring through methyl supplementation: Altering epigenetic marking later in life. Journal of Neuroscience, 25, 11045–11054. Welch-Ross, M. K., Diecidue, K., & Miller, S. A. (1997). Young children’s understanding of conflicting mental representation predicts suggestibility. Developmental Psychology, 33, 43–53. Welzer, H., & Markowitsch, H. J. (2005). Towards a bio-psycho-social model of autobiographical memory. Memory, 13, 63–78. Wheeler, M. A., Stuss, D. T., & Tulving, E. (1997). Toward a theory of episodic memory: The frontal lobes and autonoetic consciousness. Psychological Bulletin, 121, 331–354. Williams, J. M., & Scott, J. (1988). Autobiographical memory in depression. Psychological Medicine, 18, 689–695. Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children’s understanding of deception. Cognition, 13, 103–128. Wood, H. (2003). Consolidating reconsolidation. Nature, 4, 774–775. Yasmin, H., Nakata, Y., Aoki, S., Abe, O., Sato, N., Nemoto, K., et al (2008). Diffusion abnormalities of the uncinate fasciculus in Alzheimer’s disease: Diffusion tensor tract-specific analysis using a new method to measure the core of the tract. Neuroradiology, 50, 293–299. Yehuda, R., Schmeidler, J., Siever, L. J., Binder-Brynes, K., & Elkin, A. (1997). Individual differences in posttraumatic stress disorder symptom profiles in Holocaust survivors in concentration camps or in hiding. Journal of Trauma and Stress, 10, 453–463. Zanarini, M. C., Frankenburg, F. R., Jager-Hyman, S., Reich, D. B., & Fitzmaurice, G. (2008). The course of dissociation for patients with borderline personality disorder and axis II comparison subjects: a 10-year follow-up study. Acta Psychiatrica Scandinavica, 118, 291–296.
Consciousness and Cognition 19 (2010) 802–815
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Retro- and prospection for mental time travel: Emergence of episodic remembering and mental rotation in 5- to 8-year old children q Josef Perner *, Daniela Kloo, Michael Rohwer Department of Psychology, University of Salzburg, Austria
a r t i c l e
i n f o
Article history: Available online 22 July 2010 Keywords: Episodic memory Mental rotation Development Prospection Theory of mind Preschool period Mental time travel
a b s t r a c t We investigate the common development of children’s ability to ‘‘look back in time” (retrospection, episodic remembering) and to ‘‘look into the future” (prospection). Experiment 1 with 59 children 5 to 8.5 years old showed mental rotation, as a measure of prospection, explaining specific variance of free recall, as a measure of episodic remembering (retrospection) when controlled for cued recall. Experiment 2 with 31 children from 5 to 6.5 years measured episodic remembering with recall of visually experienced events (seeing which picture was placed inside a box) when controlling for recall of indirectly conveyed events (being informed about the pictures placed inside the box by showing the pictures on a monitor). Quite unexpectedly rotators were markedly worse on indirect items than non-rotators. We speculate that with the ability to rotate children switch from knowledge retrieval to episodic remembering, which maintains success for experienced events but has detrimental effects for indirect information. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction We naturally speak of ‘‘looking back on past events” or ‘‘looking forward to the future.” Clearly, we cannot literally look back or forward in time (Martin, 2001). We cannot see the future or the past in the same way as we see an event unfolding in front of us or behind us. The best we can do to capture past or future experiences of an event is to re-experience the event (retrospection), or imagine experiencing a future event (prospection).1 The ability to retrospect in this sense has become a central feature for episodic memory with Tulving’s (1985) introduction of ‘‘autonoetic consciousness”, namely the awareness that remembering consists of ‘‘calling back into consciousness a seemingly lost state that is then ‘immediately recognized as something formerly experienced’ (Ebbinghaus, 1885, p. 1).” Philosophers spoke of ‘‘experiential memory,” at least since Locke (Owens, 1996). Although the different terms all capture the phenomenon adequately, and ‘‘retrospection” does so in nice juxtaposition to ‘‘prospection,” we prefer ‘‘episodic remembering” as our standard term. The choice of ‘‘remembering”, rather than ‘‘memory” is to emphasise that it is more than just
q
This article is part of a special issue of this journal on Self, Other and Memory. * Corresponding author. E-mail address: [email protected] (J. Perner). 1 In response to an anonymous reviewer we need to point out that the precise form of how the future is addressed in different prospection studies varies. Sometimes it is a concern about a specific future event (e.g., on the way to the puzzle room I have to think taking the needed implement with me; Suddendorf & Nielsen, 2009) but sometimes a timeless assumption suffices that one can treat as placed in the future (e.g., When(-ever) I go to a hibernal resort (tomorrow) I will take my coat and not the swimsuit; Atance & Meltzoff, 2005). In both cases the events one imagines experiencing (into which one projects oneself) are imagined possibilities (i.e., although one intends to do the puzzle, it is not a fact that one will) in contrast to retrospection, where one re-experiences (projects oneself into) a past event that has actually taken place. By precedent and due to this commonality we keep using the term ‘‘prospection” for experiencing imagined events in contrast to retrospection for re-experiencing actual events. 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.022
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
803
retrieval of knowledge about a past episode. It is a re-experiencing of that episode. This distinction is also brought out by the notion of ‘‘mental time travel” (MTT: Suddendorf & Corballis, 1997; Wheeler, Stuss, & Tulving, 1997): one has to not just retrieve information about the past or think about the likely future. It requires projecting oneself as an experiencing agent into the past or future. This distinction is also akin to the difference between having a theory of mind (Churchland, 1984; Gopnik & Astington, 1988) as opposed to simulating one’s own (or other people’s) mental processes (Goldman, 2006; Gordon, 1986; Heal, 1986). There is the strong and widespread, but not uncontested, intuition that these three abilities, theory of mind, episodic remembering, and prospection belong together. They may be uniquely human abilities (Gilbert & Wilson, 2007; Roberts, 2002; Roberts et al., 2008; Suddendorf & Busby, 2003; Suddendorf & Corballis, 1997, 2007; Wheeler et al., 1997). There is now also fast growing empirical evidence from different quarters that episodic remembering (often investigated under the label or as part of autobiographical memory), prospection, and theory of mind share a developmental schedule, and common neural substrate as shown in coactivation patterns and in common deficits in clinical cases. 1.1. Brain imaging Spreng, Mar, and Kim (2008) meta-analyzed, among other areas, all available brain imaging studies on theory of mind (n = 30, as a random selection of 50), autobiographical memory (n = 19), and prospection (n = 6). Direct overlap was observed in the medial temporal lobe (left parahypocampal gyrus: BA 36), medial parietal regions (precuneus, posterior cingulate, bilaterally: BA 31), left temporo-parietal junction (BA 39, touching on BA 19), medial prefrontal cortex (frontal pole: BA 10). Convergence within same Brodmann areas were also observed in right TPJ, left ventrolateral prefrontal cortex (BA 47), medial prefrontal cortex, and rostral anterior cingulate (BA 32), and lateral temporal lobe (BA 21, 22) especially left. All these regions tend to also be activated by navigation problems and default processing (areas that tend to be more strongly activated in the absence than in the presence of external stimulation). Two theories have been proposed as to the common denominator underlying these common activations. Hassabis and Maguire (2007) suggested that all these tasks require scene-construction and Buckner and Carroll (2007) that they all require projection of self into different time points or spatial locations. 1.2. Brain injury Patients without autonoetic consciousness in Tulving’s sense, i.e., amnesics with a special impairment of episodic remembering and autobiographical memory, have been reported to also have severe deficits of prospection, Patients K.C. (Tulving, 1985), R. (Stuss, 1991), M.L. (Levine et al., 1998), and D.B. (Klein, Loftus, & Kihlstrom, 2002). Loss of autonoetic consciousness does, however, not lead inevitably to an impairment in theory of mind (patients K.C. and M.L.: Rosenbaum, Stuss, Levine, & Tulving, 2007). This finding does not preclude theory of mind being necessary for autonoetic consciousness and episodic remembering, in particular, it may be crucial as a developmental requirement or linkage. In fact, there is growing evidence that theory of mind development around 4 and 5 years is linked to both, episodic remembering as well as prospection. 1.3. Development There are studies showing a specific relationship between advances in theory of mind and free recall as a measure of episodic remembering in relation to cued recall (Tulving, 1985). Perner and Ruffman (1995) were able to show that between 3 and 5 years, children’s improvement on free recall correlates significantly with their understanding of how knowledge depends on experience. Even when cued recall and verbal intelligence were partialled out, correlations stayed above .30. Tasks used included children’s ability to explain why they know the contents of a box (How-do-you-know test: Wimmer, Hogrefe, & Perner, 1988a; Wimmer, Hogrefe, & Sodian, 1988b), to distinguish a lucky guess from proper knowledge (Miscione, Marvin, O’Brien, & Greenberg, 1978), and to understand which sense modality to use to find out about colour or weight of an object (O’Neill et al., 1992). These results were largely replicated by Naito (2003) on a Japanese sample. She also found a relationship between free recall and children’s ability to understand when they had learned a fact (Taylor, Esbensen, & Bennett, 1994). Perner, Kloo, and Gornik (2007) used a different measure of episodic remembering. They contrasted recall of experienced events with recall of indirectly conveyed events. In the experience condition children put cards with drawings of simple objects into a box. In the indirect-information condition they were blindfolded and so could not see which cards they put inside. They were afterwards shown on a monitor the pictures that were on these cards (information about individual cards was thus indirectly conveyed). The reasoning was that children can have an episodic memory of putting a particular card inside the box only when they experienced putting that card into the box. Only with this experience can they later re-experience their action. When blindfolded, they cannot experience which card they had put inside. When later shown what was on these cards, they can only infer that they put that card inside the box. Hence there is no experience of putting that card inside, which they could re-experience. Free recall of experienced events correlated with performance on various theory of mind tasks (How-do-you-know, When-did-you-learn, and the modality-specificity test). Sprung (2008) and Sprung and Harris (2009) found that theory of mind abilities (especially introspective understanding) modulates children’s ability to report
804
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
on their intrusive thoughts, which in case of traumatic experiences (e.g., Hurricane Katrina victims) provides a basis for outgrowing the trauma. Rapidly growing evidence suggests that children’s prospective abilities develop at about the same age. Moore, Barresi, and Thompson (1998, Experiment 1) reported some correlations between 3- and 4-year-old children’s understanding of desire and belief and of the benefits of increased but delayed reward. Bischof-Köhler (2000) investigated several abilities in a group of 3- to 4.5-year-olds with a large battery of tasks including understanding of duration, theory of mind (false belief and deception), planning (shopping, need for preparation), delay of gratification, and motivational conflict. These abilities all underwent a noticeable improvement from 3 to 4.5 years and correlated significantly even when children’s age was partialled out. Atance and O’Neill (2005) investigated 3- to 4-year-olds’ ability to think about what to pack for a trip. About half of these children went beyond packing typical (scripted) or attractive things. They also thought of providing for possible eventualities (e.g., pack telephone to contact someone in case of an emergency). In another task children were given the beginning of a drawing, e.g., a straight line, and then were asked what they wanted to draw on the basis of this first element. There was a strong correlation r = .65 between children’s ability to anticipate events in the trip task and their choice of a drawing that was feasible with the given element (e.g., a sun but not a ladder starting from a circle) even when language ability was controlled. Atance and Meltzoff (2005) tested for mental time travel and found that 3 year olds above chance chose a suitable item (e.g., winter coat) for going to a snowy place (as shown in a photograph) and they were able to explain their choice by reference to their future need created by that environment. By 4 and 5 years all children could do this. It remains unclear whether this shows children’s ability to project themselves into the future state of going to a wintery landscape. They might just know about the need for winter coats in a cold environment. Atance and Meltzoff (2006) overly satiated children on salty pretzels, which made them change their preference from pretzels to water. Even 5 year old children showed little sign of understanding that the next day their preference would return to pretzels. They predicted a lasting preference for water. Busby and Suddendorf (2005) simply asked 3- to 5-year-old children what they had done yesterday and what they were likely to do tomorrow. Discounting general, scripted answers (e.g., ‘‘I played”) there was a marked improvement in children’s ability to report past and likely future episodes. Suddendorf and Nielsen (2009) tested 3- and 4-year-olds on a simple problem (e.g., use a triangular key to open a box). When the key was not available children went to another room and played for 15 min. Before returning to the first room they were given the choice among three different keys (one triangular) and asked which one they would like to bring with them to the room. Three-year-olds chose at random, while 4-year-olds chose the triangular key above chance (with plenty of room for improvement). Atance and Jackson (2009) assessed different aspects of future thinking in 3- to 5-year-old preschoolers, including mental time travel (Atance & Meltzoff, 2005; Busby & Suddendorf, 2005), delay of gratification, planning (simple version of Tower of Hanoi, Carlson, Moses, & Claxton, 2004), and prospective memory (remembering to do something after finishing something else: Kvavilashvili, Messer, & Ebdon, 2001). Children improved on all these tasks and performance between all tasks was correlated but correlations were dependent on age and receptive vocabulary. Ford, Shum, and Driscoll (2009) found that prospective memory performance in 4- to 6-year-olds was related to inhibitory ability and understanding false beliefs. Finally, Russell, Alexis, and Clayton (2010) let 3–5 year old children play blow-football for which one needed to bring a straw for blowing and as player on the blue (as opposed to the red) side a box to stand on as that side of the table was otherwise too high for children at this age. Children were asked questions framed in the past, present and future about what they themselves or another child needed to play on the blue side. The variable was whether they thought of both required items or not. In case children gave different answers for themselves than for their peer, the authors reasoned, this is a likely sign that children bring to bear their own perspective when imagining how they had played, or were going to play, i.e., project themselves into the role of player. Systematic comparison of self and other was only carried out for present (Experiment 4) and future (Experiments 2 and 3). In particular, 4-year-olds gave similar answers for self and other in the present and for other in the future, but ignored their own needs for the future. This was interpreted as evidence that between 3 and 5 years children become able to project themselves into future situations. Supportive evidence for this age trend comes from Prencipe and Zelazo (2005), who found that 3 year olds make wise choices of delaying an immediate small reward for a larger one later when choosing for another person but not for themselves. This discrepancy ceases around 4 years. 1.4. Autism The spectrum of autism consists of a developmental disorder that affects the very triad of theory of mind, retrospection, and prospection. Children with autism spectrum disorder tend to have problems with basic theory of mind tasks (Baron-Cohen, Leslie, & Frith, 1985; Perner, Frith, Leslie, & Leekam, 1989) or the least impaired cases (Asperger’s syndrome) have deficits appreciating subtle mental interactions like sarcasm, jokes, etc. (Happé, 1994). They also have executive deficits (Ozonoff, Pennington, & Rogers, 1991; for a review, see, Hill, 2004), which tend to be more pronounced for planning (prospection) than for inhibition. When tested for memory their free recall seems to be specifically impaired over cued recall and recognition (Boucher & Lewis, 1989; Boucher & Warrington, 1976; Bowler, Gardiner, & Grice, 2000; Tager-Flusberg, 1991) and even the highest functioning individuals with Asperger syndrome give fewer ‘‘remember” judgments in recognition tasks (Bowler, Gardiner, & Gaigg, 2007) than unimpaired people.
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
805
1.5. Sharpening the issues The neurophysiological and developmental evidence leaves little doubt that there is some common ground for theory of mind, episodic remembering and prospective abilities. Whether the co-emergence of these different abilities and their sharing of common cerebral structures is good evidence for mental time travel, i.e., for the ability to re- and pre-experience past or future events as the common basis remains an open question. The evidence for that claim is weak in two respects: few of the reviewed studies attempt to isolate specific cases of episodic remembering (retrospection: re-experiencing a past situation by projecting oneself into the past) as opposed to merely retrieving knowledge about the past, and of projecting oneself into a future experience (prospection) as opposed to knowing what is likely to happen in the future or knowing what is needed in an imagined future situation. Consequently, the developmental synchrony could easily be explained by quite general features, e.g., thinking about the past and future. The contrast between free and cued recall does provide some measure for teasing out the retrospective aspects of memory. Its usefulness is based on Tulving’s (1985) argument that free recall is more dependent on the availability of episodic traces than is cued recall. Hence, if free recall correlates specifically with progress in theory of mind when controlling for cued recall, then that correlation ought to be due to changes in episodic remembering (Perner, 1990) and not some more general ability to think about or retrieve information about the past. The link between free recall and episodic remembering, one has to admit, is rather weak. Free recall does not depend completely on episodic remembering (even in free recall a few items can come to mind automatically). Moreover, free recall is not exclusively helped by episodic remembering, which can also enhance cued recall. The contrast between memory of experienced events and indirectly conveyed events (Perner et al., 2007) is somewhat sharper in this respect. It takes advantage of the fact that only experienced events can possibly be reexperienced. Indirectly conveyed events should, therefore, not profit at all from the developing ability to remember episodically. The self-other contrast used by Russell et al. (2010) helps isolate the prospective aspects of anticipating future needs. It also allows for projection into the past, unfortunately, not for remembering any past event but only for figuring out what was needed for a past action as can be seen from their example test question: ‘‘. . . point to the two things you think the little girl had to have to play blow-football on the blue side?” (Section 2.1.3). Even for prospection this method depends on the risky background assumption that the difference between predicting one’s own future needs and predicting someone else’s future needs reflects an inability of projection, which is primarily applicable to oneself and not to others. For believers in simulation theory projection is also applied in the case of others. More generally, the difference can be due to factors that tend to interfere more strongly in the case of self than other. In sum, the claim for a specific developmental relationship between the ability to retrospect (episodic remembering) and prospection rests on evidence that can be given a more general interpretation (e.g., thinking about different times) or fails to provide the needed correlations: Perner and Ruffman (1995), Naito (2003) and Perner et al. (2007) only provide evidence for retrospection and ToM but not prospection. Russell et al. only provide evidence for prospection. Our prime objective is to provide evidence for a specific developmental relationship between retrospection (episodic remembering) and prospection. To capture retrospection we use the free-cued recall contrast in our first experiment and the experienced-indirectly conveyed contrast in our second experiment. For measuring children’s prospective abilities (the study by Russell et al. was published well after our research was conceived and conducted) we relied on children’s ability to engage in mental rotation. We used the child appropriate simplification by Estes (1998) of the original tasks by Shepard and Metzler (1971). Children were shown pairs of teddy bears (monkeys in Estes’ original study) with one arm raised. Children had to judge whether the monkeys had raised the same side arm (e.g., both their left arm) or an arm on different sides (one left the other right). This is easy if the creatures stand side by side but increasingly difficult if one of them is rotated sideways (see Fig. 1). An easy solution is to take the tilted picture and rotate it back to upright. However, children are not allowed to do this physically. In that case one can rotate it mentally by forming an image of what one sees and rotate the image and then compare the rotated image of the tilted figure with the perception of the other figure. This procedure is an instance of prospection: one pre-experiences an actual rotation of the tilted figure. And the reaction times tell us whether a child used this method because mental rotation results in a linear increase of reaction time with the degree of rotation (Shepard & Metzler, 1971). Estes (1998) reported that children start using this method between 4 and 6 years.
Fig. 1. Examples of stimuli used for the mental rotation task in both experiments.
806
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
2. Experiment 1 2.1. Method 2.1.1. Participants Fifty-nine children (25 girls and 34 boys) participated in the study. Children came from a nursery school in Upper Austria and two after-school care clubs, one in the city of Salzburg and one in Upper Austria. Most participants came from a middle-class background. Children’s ages ranged from 4, 11 (years, months) to 8, 7 (M = 6, 8, SD = 13.67 months). To analyze and display age trends, we divided the children into three approximately same sized age-groups: Twenty children from 4, 11 to 6, 0 (M = 5, 5, SD = 3.99 months), 17 children ranging in age from 6, 1 to 7, 4 (M = 6, 8, SD = 5.51 months), and 22 children ranging in age from 7, 5 to 8, 7 (M = 7, 11 SD = 4.4 months).
2.1.2. Design Each child was tested individually in a quiet room of the nursery school or after-school care club. Children were given two memory tasks (one with free and one with cued recall). In addition, children received a computerized mental rotation task based on Estes (1998) and an age-appropriate measure of verbal intelligence (verbal subtest of KISTE, Häuser, Kasielke, & Scheidereiter, 1994; or HAWIK-III, Tewes, Rossmann, & Schallberger, 1999). All tasks were administered in two sessions a few days apart. Each session started with presentation of items for the memory task and ended with recall of memory items. Between presentation and recall of memory items, half of the children were given the mental rotation task in the first session and the verbal intelligence measure in the second session; the other half received these tasks in the opposite order. Half of the children received the free recall task in the first session and the cued recall task in the second session; the other half started with the cued recall task.
2.1.3. Procedure and materials 2.1.3.1. Mental rotation task (Estes, 1998). In a warm-up phase, children were presented with an odd one out game. They were shown three bears in upright position (two of them both raising the same arm and one of them raising a different arm). Children were asked to select the bear that was different from the other two. Then, children were asked to play a computerized mental rotation task presented on a notebook computer using the software package Presentation (Neurobehavioral Systems Inc., http://www.neuro-bs.com). Children were told to press the button with the smiling face if two bears both raising the same arm appeared on the screen and to press the button with the sad face if two bears each raising a different arm appeared. After experimenter and child had jointly completed six training trials, children were asked to play on their own. The test phase comprised 56 trials. Each trial consisted of a pair of bears. The bear on the left was always upright, whereas the bear on the right appeared in seven different orientations. He was either upright (i.e., 0° rotation) or rotated clockwise in 30° increments up to 180°. In addition, each bear raised one of his arms. For half of the trials, both bears raised the same arm (i.e. both right or both left). For the other half of trials, the bears raised different arms (i.e. the left bear raised its right arm and the right bear raised its left arm, or the reverse). In total, there were 28 different stimulus pairs: two ‘‘same” and two ‘‘different” pairs for each of the seven different orientations. Each child was given these 28 different stimulus pairs twice in a fixed random order. After the first 28 trials, there was a short 2 min break. Children had to press as fast as possible the button marked with a smiling face if the two bears had raised their same side arm and the button marked with a sad face if the bears had raised different side arms. Children had to keep their index fingers positioned on the response buttons throughout the trial block. Children were asked if they were ready before the experimenter initiated the next trial. Correct responses were followed by a brief tune, errors by silence. No other feedback was given. After Trials 5, 28, and 56, children were asked how they could tell if the two bears had raised their same arms or not. Sample stimuli are shown in Fig. 1.
2.1.3.2. Memory tasks. For the memory tasks two sets of 20 coloured pictures (21 29.7 cm) of familiar objects or animals were created (similar to Perner & Ruffman, 1995, Exp. 1). Each set comprised five categories, with four items in each category (see Appendix A for list of items). Pictures were presented in a fixed random order. Each child received both sets. Half of the children were given Set 1 for the free recall task and Set 2 for the cued recall task. For the other half the two sets were exchanged. Children were told to look carefully at each picture they are going to see, because they will be asked about them later. Each picture was shown for 5 s and the child had to name it. Slight misidentifications (e.g., calling the farm a barn) were accepted. Completely wrong answers were corrected. After the intervening task (mental rotation or verbal intelligence measure), children were reminded: ‘‘I’ve shown you some pictures earlier on. Can you remember?” In the free recall condition, children were then simply asked ‘‘What was in these pictures?” In the cued recall condition, they were asked for each category, e.g.: ‘‘There were some animals, what were they?”
807
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
The number of correctly recalled items and of false alarms (items that had not been on the learning list) were recorded. The difference between the number of correct recalls and the number of false alarms was used as a measure of recall accuracy. 2.2. Results and discussion First, children’s performance on the memory tasks and on the mental rotation task was analyzed separately, and then we looked at the relationship between these tasks. 2.2.1. Memory tasks Recall accuracy (hits minus false alarms) on the two memory tasks was analyzed by an analysis of variance with agegroup (younger, middle, older) as between participants factor and recall (free, cued) as a within participants factor. There was a significant main effect of recall, F(1, 56) = 49.61, p < .001, partial g2 = .47. No other effect was significant p > .50. Correlation between free and cued recall accuracy was r = .17, p > .19. 2.2.2. Mental rotation We first looked at children’s number of correct answers. There was a clear bimodal distribution. One group of 24 children had values from 24 to 38 normally distributed around the mean of 30.21 correct, which is just two items more than the guessing level of 28 (of 56 items): t(23) = 3.35, p < .003. Presumably these are the children who did not resort to rotation, without which they could only judge about two items consistently correctly. There were no children with values 39–41 and the remaining 35 children attained values from 42 to 55 with a mean of 49.89 correct. To see how children managed to give correct answers they were classified as rotators or non-rotators according to the two criteria used by Estes (1998). For each child a regression analysis was computed for the median reaction times for each angle as dependent and angle of rotation as independent variable. When the regression coefficient (slope) was significantly different from zero the child was classified as a ‘‘rotator according to RT”. Children were also asked at the end how they had approached the task. If they indicated that they had rotated the stimuli in their mind (e.g., ‘‘I rotated the bear in my head”) they were classified as ‘‘rotators according to explanation”. The other children gave no insightful answer (e.g., ‘‘I just know it”). Correlation of these measures with the number correct solutions showed r = .64 for explanation, r = .66 for reaction time, and r = .67 for rotation by reaction time and explanation. We use this binary classification as ‘‘rotators” (rotators by RT and by explanation) vs. ‘‘non-rotators” for all further computations (we checked that the interpretations stay the same if anyone of the other two is used). This way of classifying children into rotators and non-rotators had a marginally significant relation with age (r = .25, p = .052) but a clear relationship with verbal intelligence (r = .40, p = .002). 2.2.3. Interrelations between retrospection and prospection The argument about the relationship between type of recall and episodic remembering is that both kinds are (can be) helped by the availability of episodic traces. This can be equally strong, so that no interactive effect of the availability of episodic traces on free vs. cued recall can be expected. However, free recall is more dependent on episodic traces than cued recall (which profits to a large degree from the availability of the cues). Availability of episodic traces should, therefore, correlate more strongly with free than with cued recall, since more alternative factors influence cued than free recall. The theoretical claim is that the ability for episodic remembering requires the ability to project oneself into the past and that the same projective ability is required in the mental rotation task by projecting oneself as an observer of a future or hypothetical action (rotating one of the items to be compared). On those grounds we expect that mental rotation should show a significant correlation with free recall even when any correlation with cued recall has been partialled out. Table 1 shows the correlations between relevant variables in the upper right corner above the main diagonal. Partial correlations with age and verbal intelligence taken into account are shown for the remaining variables in italics below the main diagonal. The correlation between rotation and free recall remains at least marginally significant but the correlation between rotation and cued recall reaches only about half its size. More importantly, when cued recall is partialled out the partial
Table 1 Correlations and partial correlations with age and verbal intelligence controlled.
1. 2. 3. 4. 5.
Age Verbal intelligence Mental rotation Free recall: accuracy Cued recall: accuracy
1
2
3
4
5
–
.17 –
.25+ .40** – .26+ .14
.10 .29* .35** – .06
.18 .39** .29* .17 –
Note: Numbers in italics – partial correlations after partialling out age and verbal intelligence. + p < .06. * p < .05. ** p < .01.
808
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
correlation of rotation with free recall remains clearly significant (pr = .31, p = .016). This is as predicted when retrospection (as measured by free recall beyond cued recall) is developmentally linked with prospection (as measured by mental rotation). In contrast, when free recall is partialled out the partial correlation of rotation with cued recall is weaker but does stay marginally significant (pr = .25, p = .06), which attests to the fact, pointed out earlier, that both free and cued recall profit from the ability of episodic remembering to project oneself back into the past. This highlights the need for a more discerning measure of episodic remembering in the next experiment. 3. Experiment 2 Experiment 1 provided some evidence that the development of prospection, assessed by the ability to engage in mental rotation, is specifically linked to retrospection (free recall) when controlling for general ability to retrieve information about the past (cued recall). The differential performance on free over cued recall is a very weak measure of episodic remembering (retrospection) since both types of recall can benefit to equal amounts from the availability of episodic traces. The only difference is that the overall variance of free recall should be more strongly dominated by episodic abilities than the variance of cued recall. This weakness is also reflected in the controversy over Tulving’s (1985) claim that relatively more items recalled in free recall should be judged as ‘‘remembered” than in cued recall (Jones & Roediger, 1995; Roediger & McDermott, 1995; vs. Hamilton & Rajaram, 2003). In this second experiment we try to get stronger evidence for a specific developmental link by employing a contrast that depends more directly on the availability of episodic traces. A suitable contrast is between (directly) experienced events and events about which one was indirectly informed (e.g., through verbal or pictorial media). Only experienced events can be reexperienced. Though imagined experiences can be mistaken for re-experiences (false memories) this should be relatively rare if children are not instructed to vividly imagine the event about which they are indirectly informed. Under these premises we can predict that development of retrospective abilities (episodic remembering) should uniquely improve recall of experienced events but not at all (except for the occasional false memory) recall of indirectly conveyed events. Following the method used by Perner et al. (2007) we had children put cards with different pictures into a box. In the direct-experience condition the children saw the picture on each card as they put it into the box. In the indirect-information condition they put the cards into the box when being blindfold and afterwards they were shown on a monitor the pictures on those cards. We expect that children’s ability to engage in mental rotation should coincide with improved recall of directly experienced items but not of indirectly conveyed items. 3.1. Method 3.1.1. Participants Thirty-one children (16 girls and 15 boys) from two nursery schools in towns near the city of Salzburg participated in the study. Most participants came from a middle-class background. Children’s ages ranged from 5, 0 (years, months) to 6, 4 (M = 5, 10; SD = 4.72 months). For later analysis children were divided into a younger (n = 13, M = 5, 5; ranging from 5, 0 to 5, 10) and an older group (n = 18, M = 6, 1; ranging from 5, 11 to 6, 4). 3.1.2. Design Each child was tested individually in a quiet room of the nursery school. Children were given two memory tasks (one with direct experience and one with indirect information). In addition, children received the mental rotation task based on Estes (1998) described in Experiment 1. All tasks were administered in three sessions a few days apart. In the first session, children received the mental rotation task and a modality-specificity task in counterbalanced order.2 In the second and third session, one of the two memory tasks was given. In each session, presentation and recall of memory items were separated by a 10-min delay. Half of the children started with the direct-experience condition and were then given the indirect-information condition. For the other half, the order was reversed. 3.1.3. Procedure and materials The mental rotation task (Estes, 1998) was administered as described in Experiment 1. 3.1.3.1. Memory tasks. For the memory tasks, four sets of 12 coloured pictures (21 29.7 cm) of familiar objects, animals, or human beings were used. Set 1 and Set 2 were administered in the first session. Set 3 and Set 4 were given in the second session. In each session, half of the children were given a particular set as test items and the other set as distractor items. For the other half, the two sets were exchanged, so that each set was equally often used as test or distractor set in the direct experience and in the indirect-information memory task. The experimenter presented the 24 pictures (test and distractor items), one after the other, alternating between items from each set. As each picture was produced, the child was asked to name it. If necessary (which was rarely the case), the experimenter provided the correct label. 2 This task was included to make good use of the retention time. It was part of a project of a student who helped testing the children. This task will not be further analysed.
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
809
In the direct-experience memory task, children were then asked to place the 12 test items into a (26 40 4 cm) box. They were allowed to look at each picture for 2 s and were instructed to keep these pictures in mind. In the indirect-information memory task, children were also asked to place the 12 test items into a box but they were blindfolded so that they could not see the pictures. After having placed the 12 items into the box, they were shown ‘‘what these 12 pictures had depicted” by means of a computerized presentation. Each picture was shown for two seconds, and children were instructed to keep these pictures in mind. After a 10-min delay, children were asked in both memory tasks, ‘‘Do you remember the pictures you put into the box?” If children answered with an item that was not put inside the box it was scored as a false alarm. On both memory tasks, the number of correct recalls (hits) and the number of false alarms (if children mentioned items that were not put inside the box) were recorded. Mostly, false alarms comprised items from the distractor set used in the familiarization phase. The distractor set was introduced to get a more precise measure of remembering the pictures being put into the box as opposed to mere familiarity with the pictures. The contrast between recall of items placed inside the box with false alarms (mostly items that children were familiarized with but did not put inside) sharpens the detection of episodic memories in contrast to mere familiarity answers. For instance, a child who recalls two correct items and no distractors is comparable with one who recalls 12 targets and 10 distractors. Both children vastly differ in recall of familiar items (pictures) but are similar in terms of memory for items placed inside the box. This is critical, because our manipulation of direct experience and indirect knowledge pertains to the placing of cards into the box, not to familiarity with the pictures (they are directly experienced in both conditions). For this reason it is essential to rely not only on the quantity of pictures recalled (number of placed pictures recalled in relation to all pictures placed into the box) but also check for accuracy (number of placed pictures recalled in relation to all pictures recalled; as this terminology is used by Koriat and Goldsmith (1996, p. 177)). A measure that captures both these aspects common in signal detection theory is d0 which is based on the difference between hits and false alarms. This is typical for forced choice recognition tasks and rarely used with recall (Koriat & Goldsmith, 1996, p. 183). Following this approach, we use the difference between number correct items recalled minus number false alarms as our critical indicator of episodic recall and refer to it, for want of a shorter label and in line with signal detection theory, as: recall accuracy. 3.2. Results First, children’s performance on the memory tasks and on the mental rotation task was analyzed separately, and then we looked at the relationship between these tasks. 3.2.1. Memory tasks Recall accuracy (hits minus false alarms) was subjected to an analysis of variance with age-group as between participants factor and experience (direct, indirect) as a within participants factor. There was a significant interaction between age-group and experience, F(1, 29) = 4.31, p < .047, partial g2 = .13. Whereas the recall of directly experienced items increased with age from 2.69 to 3.56 items, recall of indirect items declined with age from 3.46 to 2.50 items. This unexpected decline will be discussed below when looking at the relationship with mental rotation (prospection). No other effect was significant p > .70. Correlation between direct and indirect recall accuracy was r = .42, p = .019. 3.2.2. Mental rotation The results mirror those of Experiment 1. The numbers of correct answers range from 28 to 54. There was again a strong correlation (r = .57, p = .001) between number correct solutions and rotation by explanation (r = .38, p = .033), rotation by reaction time (r = .46, p = .01), and rotation by reaction and explanation (r = .58, p = .001). As in Experiment 1 we use rotation by reaction time and explanation for our further analyses. This classification had a marginal correlation with age in Experiment 1 but this time it had no correlation with age at all (r = .00), which is probably due to the much narrower age range in this experiment. 3.2.3. The relation between retrospection and prospection The contrast between directly experienced versus indirectly conveyed events works differently than the contrast between free and cued recall in Experiment 1. While free and cued recall can profit from episodic remembering, only directly experienced events can do so, because only experienced events can be re-experienced. This means that the emerging ability for episodic remembering should enhance recall of directly experienced events but not recall of indirectly conveyed events. Moreover, re-experiencing past events (episodic remembering) and pre-experiencing imagined events for mental rotation require the same ability of projecting oneself as an observer. Hence, we expect that children who can use mental rotation will have better recall of directly experienced events than of indirectly conveyed events. Children, who fail to use rotation, will show no, or at least a much reduced, difference in recall. An analysis of variance of recall accuracy (number of hits minus false alarms) with rotation (rotation by RT and explanation: rotators vs. non-rotators) as a between subjects factor and experience (experienced vs. indirectly conveyed events) showed significant main effects of rotation (F(1, 29) = 5.16, p = .031, partial g2 = .15) and experience (F(1, 29) = 4.51, p = .042, partial g2 = .13), and a highly significant rotation experience interaction (F(1, 29) = 7.83, p = .009, partial g2 = .21). This interaction was predicted. However, the form of the interaction as displayed in the left panel of Fig. 2 harbours
810
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
Fig. 2. Results for rotators and non-rotators on three different memory measures in Experiment 2.
a surprise. The ability to mentally rotate does not lead to enhanced memory for experienced events but has a detrimental effect on trying to remember indirectly conveyed events. This decline in accuracy for indirectly conveyed events persists for rotators-by-RT but not for rotators-by-explanation (p = .016), although the pattern of results is still very similar. This sharp decline (by 3.2 items) in accuracy (hits minus false alarms) is, as the centre panel in Fig. 2 shows, less pronounced (1.6 items) for hits (items correctly recalled) but still significant (interaction: (F(1, 29) = 4.81, p = .037, partial g2 = .14)). At the same time the number of false alarms increases more sharply for indirectly conveyed items than experienced items (F(1, 29) = 4.45, p = .045, partial g2 = .13) as the right panel in Fig. 2 shows. 3.3. Discussion The ability to engage in mental rotation had a strong differential effect on children’s ability to recall directly experienced events and to recall indirectly conveyed events. The finding, however, contained a great surprise: the differential effect did not consist in rotators’ better recall of experienced events but was due to their worse recall of indirectly conveyed events. This unexpected decline in recall of indirectly conveyed items does, nevertheless, confirm a trend observed in two experiments by Perner et al. (2007), where recall of indirectly conveyed items declined with increased theory of mind competence. What seemed initially a curious and haphazard finding now looks increasingly robust: As children improve their theory of mind and develop mental rotation skills their recall of indirectly conveyed information declines. It is difficult to find an obvious explanation for this decline. One could suspect that the reason lies in the unusual procedure of the indirect condition. But why should the more sophisticated rotators become confused by this procedure, when adults find it unusual but clear and perform as well as in the more normal experience condition (Stöttinger, 2006). The only plausible but highly speculative explanation has been briefly alluded to by Perner et al. (2007, p. 480). The younger children use their non-episodic recall (knowledge retrieval) for both conditions more or less successfully. Then when their theory of mind competence increases and become able to project themselves as past perceivers of experienced events they switch to using episodic recall. This switch may (Perner et al., 2007) or may not (current Experiment 2) help them immediately to recall more items than with non-episodic recall. In any case, this switch does not work for indirectly conveyed items and children’s performance declines. Presumably some time after discovering episodic recall children will realize that it does not work in every case and will readjust their strategy for indirect information. Eventually they may also discover the possibility of indirect episodic recall for indirect information. That is, when told about an event they should not try to re-experience (episodically remember) the event (e.g., putting the crocodile into the box) but to remember the information event (i.e., remember the crocodile appearing on the monitor and the instructions that pictures on the monitor were on the cards put inside the box). It is left to future studies to explore this development. 4. General discussion We have presented some data showing that prospection (as measured by children’s ability to engage in mental rotation) is related to retrospection (episodic remembering: as measured by variance shared with free recall when controlling for cued recall and by the difference between recall of experienced events (which can be re-experienced) versus recalling indirectly conveyed events (which cannot be re-experienced). Our expectation was that recall of indirect information would be unaffected while recall of experienced events, which can be re-experienced, would increase with children’s ability to mentally rotate. Contrary to expectations recall of experienced events was unaffected while children’s recall of indirect information dropped drastically with their ability to engage in mental rotation. Although, retrospectively (sic!), this trend was already apparent in investigations of children’s theory of mind competence (Perner et al., 2007), the sharpness of this trend with
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
811
mental rotation came as a surprise. The only plausible explanation we found was that acquisition of mental rotation marks a change in recall strategy. Instead of retrieving knowledge as before, children expect to find answers by trying to re-experience the relevant events. This is a successful replacement for the old strategy in the case of experienced events but fails miserably for indirect information about events. Despite these unexpected results we were able to demonstrate a developmental link between retro- and prospection that is Difficult to reduce to a more general connection, e.g., understanding time, which could account for existing data as reviewed in the Introduction. The specific connection we identified is children’s ability to project themselves mentally as experiencing events. This characterisation is commensurable with the neurocognitive suggestions of self-projection (Buckner & Carroll, 2007) and scene-construction (Hassabis & Maguire, 2007). In fact, on our view these two abilities are necessary complements: To project oneself as an observer of events at different times one has to be able to project one self into different times AND be able to reconstruct the scenes that one is to observe. Our results are less compatible with the ages observed in earlier developmental studies. Results from many of the studies reviewed earlier that looked at prospection suggest that the main development takes place between 3 and 5 years, though not from all studies. For instance, Atance and Meltzoff’s (2006) report that 5-year olds still could not imagine/anticipate that their current desire for water induced by overfeeding on salty pretzels will have reverted next day to their usual desire for pretzels. The fact that prospection as measured with mental rotation points to later development might mean that the studies that see the development completed by 5 years did not assess children’s ability to project themselves as observers (experiencers) of events at different times but assessed their knowledge about events in the past or future. Research on children’s retrospection (episodic remembering) also finds development in this age bracket but clearly also points to improvements beyond the age of 5 years. The original studies by Perner and Ruffman (1995) and also Naito (2003) included children up to 7 years and their more difficult theory of mind tasks tended to show the stronger correlations with measures of episodic remembering. These findings are, therefore, more in line with the developments documented in the present studies. We want to end with more general considerations about the methodological difficulties in assessing young children’s and non-verbal creatures’ explanations and subjective phenomenology. The autonoetic awareness of re-experiencing a formerly experienced event is something very subjective, in the sense that evidence for it comes exclusively from introspective verbal reports. With adult participants in memory experiments the relevant experience has been assessed by the ‘‘R–K” test. People have to judge their subjective experience of recall or recognition as one of merely knowing (K) what had happened (presentation of an item) or remembering (R) this event. This test is less than straight forward with adults, its value highly controversial (Donaldson, 1996; Dunn, 2004; Gardiner & Richardson-Klavehn, 2000; Stöttinger, Aigner, Hanstein, & Perner, 2009; Stöttinger, Kaiser, & Perner, 2009) and its use for young children beyond contemplation. The only solution seems to be an objectively measurable indicator of these subjective phenomena. This enterprise requires a fine instrument in order to distinguish between knowing what happened and re-experiencing (remembering) the event retrospectively, and between knowing what will happen and pre-living the likely event in prospection. The instrument has to be precise because both kinds of mental processes can, in principle, produce the required information about the world. They are ‘‘computationally equivalent”: both can give the correct answer to the questions posed, e.g.: What happened? Which item was presented? The difference must primarily lie in how the system arrives at the answer. Cognitive Psychology can, supposedly, provide the relevant instruments. So, no wonder that one of its most impressive findings, mental rotation, has such appeal in this context. The critical point is not that mental rotation can provide the correct same–different judgments—there may be infinitely many computationally equivalent ways. It is the pattern of reaction times that tells us that the answer is arrived at by mental rotation as we subjectively experience it. However, even the reaction time pattern is not a fail safe indicator of prospection. Several species of animals can give correct responses, e.g., pigeons, but do not show the same linear pattern of reaction times with angle of rotation (Hollard & Delius, 1982). Their visual system directly provides rotation invariance and they do not have to mentally simulate perceiving a hypothetical rotation. However, the reaction times of baboons (Vauclair, Fagot, & Hopkins, 1993), a sea lion (Stich, Dehnhardt, & Mauck, 2003) and a lion-tailed macacque (Burmann, Dehnhardt, & Mauck, 2005) also show some signs of the linear relationship with angle of rotation, though their times deviate from the human pattern in ways adaptive for their respective perceptual environment. For argument’s sake let us assume their reaction times showed exactly the same pattern. We could still not conclude that these animals project themselves as perceivers of hypothetical rotation events (mental time travel). Their visual system might provide rotation invariance but processing time happens to depend on angle of rotation. Thus, the ultimate justification for interpreting the linear reaction time pattern in humans as a sign of mental time travel hinges on the fact that this linearity goes hand in hand with the subjective experience of mental rotation in adults. This relationship between observable answer and subjective experience in adults is not so tight for other measures. For instance, when asked what to pack for a trip to a place shown in a hibernal scene (Atance & Meltzoff, 2005) my answer ‘‘winter coat” could be given by projecting myself travelling to that place and noticing the need for a winter coat. However, it could equally well be based on just knowing that in places like this you need a winter coat. Mental rotation (provided the linear reaction time pattern does indicate it) is a much less vulnerable indicator of prospection. For investigating episodic remembering (retrospection) in animals and young children several proposals have been made. The technically most sophisticated has been developed by Yonelinas (1999) for distinguishing between recall based on famil-
812
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
iarity versus recollection, which can be profitably used even with rats (Sauvage, Fortin, Owens, Yonelinas, & Eichenbaum, 2008). Unfortunately the distinction between familiarity of items and recollection of associative information is not quite the same as between knowing and remembering. Associative information can be known or remembered. We need methods that drive a wedge between knowing the past and remembering the past. Perner and Ruffman (1995) followed Tulving’s criterion that free recall produces relatively more experiences of remembering than cued recall. Unfortunately the link to the one direct method of assessment of the remember–know paradigm remains tenuous (Hamilton & Rajaram, 2003; Jones & Roediger, 1995; Roediger & McDermott, 1995). Perner et al. (2007); our Experiment 2) relied on an objective prerequisite for episodic remembering, i.e., the direct experience constraint (Stöttinger, Aigner, et al. 2009; Stöttinger, Kaiser, et al., 2009): only an experienced event can be re-experienced. Neither of these methods is apt to produce positive demonstrations of a child enjoying a re-experience of a past episode. These methods only allow an objective test of developmental hypotheses: if some other ability (theory of mind, mental rotation) can be claimed to index the onset of re-experiencing then the method allows for a testable prediction. The development of the index ability should have a differential effect on recall conditions (free vs. cued; recall of directly experienced vs. indirectly conveyed events). Although the method falls short of giving a direct existence proof of episodic remembering it does provide an empirical test of hypotheses about episodic remembering that could also be used on animals. As it has been applied here it requires some linguistic proficiency for the indirect-information condition. It would not work without being able to tell children that the items shown on the monitor are the same as on the cards they could not see. Indirect knowledge through verbal communication is difficult to use with animals but there are dogs (Miklósi, Polgárdi, Topál, & Csányi, 2000) and goats (Kaminski, Riedel, Call, & Tomasello, 2005), who understand pointing gestures, and there are chimpanzees who can infer from seeing that one of two containers is empty that the object must be in the other container (Call, 2004). So, if we can find a theory of which individuals (ontogenetic development, enculturation, evolution) are able of episodic remembering, then we can provide testable predictions of how these individuals differ in recalling directly observed locations vs. gestured or inferred locations. Clayton and Russell (2009) made a new suggestion of how to differentiate knowledge of the past from remembering the past empirically without use of language. Their minimal requirement for remembering is evidence during re-experience of the subjective perspective of the original experience. This is clearly a highly relevant aspect—but an important limitation has to be pointed out. Like the direct experience constraint, this criterion cuts only one way: features of subjective perspective at recall do not licence claims about re-experiencing. For, perspective features could result from how knowledge of the past is encoded. To use an example from Davies, Russell, and Russell (2009), children observed how to fix a toy boat by performing on a left or right lever either a pumping or levering motion. By 3 years most children imitated after some delay a series of two consecutive actions in the correct spatio-temporal order. The authors suggested that this shows earlier episodic remembering than commonly expected, because children reproduced the temporal perspective of their original experience. The problem with this conclusion is that the study confounds the subjective order of experiences with the objective knowledge of the order in which the actions were performed. What one might be able to do is the following. Let us assume children below 3 years can remember sequences of events to some degree. One would then have to show that at the purported age of 3 years children show a marked change in the order in which they recall witnessed events but do not show any change in the order in which they report the events having happened. This method put together with the direct experience vs. indirect information contrast (Perner et al., 2007) may provide a more powerful combination. Reports of events need not follow the order of the events. So, particularly strong evidence would be forthcoming if at the critical age children began recalling events in the order they had been reported rather than the order in which they happened, e.g., when told ‘‘he ate after his nap” the younger ones recall, ‘‘he slept”–‘‘he ate,” while the older ones recall, ‘‘he ate”–‘‘he slept”. This kind of reversal would provide an interesting analogy to our finding that rotators started to have problems with indirectly conveyed events because—our suggestion—they switched from knowledge retrieval to episodic remembering. There has been a certain resentment among animal and infancy researchers against insisting on Tulving’s (1985) and Wheeler et al. (1997)) criteria for episodic memory because they seem to be impossible to meet for non- or low verbal creatures. With our research and this final discussion we want to make the case that to capture the phenomenon of episodic remembering we need to insist on these criteria but that there is a way of testing theories of when episodic remembering develops or evolves.
Acknowledgments The authors acknowledge the financial support from the Austrian Science Fund (FWF Grant P16215-G04 ‘‘Episodic Memory and Conscious Experience”). The data of Experiment 1 were collected by Michael Rohwer for his Diploma thesis (2006) ‘‘Zusammenhang zwischen episodischem Gedächtnis und mentaler Rotation.” We thank the Head and staff of the Kindergarten Seekirchen, Kindergarten Thalgau, Kindergarten Thalheim, Hort Taxham, and Hort Thalheim for their willing participation and Elisabeth Stöttinger for her guidance through the adult memory literature.
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
813
Appendix A Memory items used in Experiment 1, ordered by category. Original German words (English translation) Set 1
Set 2
Tiere (animals)
Pflanzen (plants)
– – – –
Schaf (sheep) Pferd (horse) Schwein (pig) Kuh (cow)
Gemüse (vegetables) – – – –
Karotte (carrot) Erbse (pea) Zwiebel (onion) Paprika (red pepper)
Möbel (furniture) – – – –
Bett (bed) Tisch (table) Stuhl (chair) Schrank (wardrobe)
Gebäude (buildings) – – – –
Kirche (church) Burg (castle) Haus (house) Scheune (barn)
Werkzeuge (tools) – – – –
Säge (saw) Schaufel/Spaten (spade) Zange (pliers) Hammer (hammer)
– – – –
Baum (tree) Sonnenblume (sun flower) Rose (rose) Getreide/Weizen (wheat)
Obst (fruit) – – – –
Birne (pear) Weintrauben (grapes) Kirschen (cherries) Apfel (apple)
Kleidung (clothes) – – – –
Pullover (sweater) Schuhe (shoes) Socken (socks) Hose/Jeans (trousers/Jeans)
Fahrzeuge (vehicles) – – – –
Bus (bus) Auto (car) Lastwagen (lorry/ truck) Motorrad (motorbike)
Musikinstrumente (musical instruments) – – – –
Trompete (trumpet) Klavier/Piano (piano) Geige/Violine (violin) Trommel (drum)
References Atance, C. M., & Jackson, L. K. (2009). The development and coherence of future-oriented behaviors during the preschool years. Journal of Experimental Child Psychology, 102(4), 379–391. Atance, C. M., & Meltzoff, A. N. (2005). My future self: Young children’s ability to anticipate and explain future states. Cognitive Development, 20(3), 341–361. Atance, C. M., & Meltzoff, A. N. (2006). Preschoolers’ current desires warp their choices for the future. Psychological Science, 17(7), 583–587. Atance, C. M., & O’Neill, D. K. (2005). The emergence of episodic future thinking in humans. Learning and Motivation, 36(2), 126–144. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘‘theory of mind”? Cognition, 21, 37–46. Bischof-Köhler, D. (2000). Kinder auf Zeitreise: Theory of mind, Zeitverständnis und Handlungsorganisation. Bern ua: Hans Huber. Boucher, J., & Lewis, V. (1989). Memory impairments and communication in relatively able autistic children. Journal of Child Psychology and Psychiatry and Allied Disciplines, 30(1), 99–122. Boucher, J., & Warrington, E. (1976). Memory deficits in early infantile autism: Some similarities to the amnesic syndrome. British Journal of Psychology, 67, 73–87. Bowler, D. M., Gardiner, J. M., & Gaigg, S. B. (2007). Factors affecting conscious awareness in the recollective experience of adults with Asperger’s syndrome. Consciousness and Cognition, 16(1), 124–143. Bowler, D. M., Gardiner, J. M., & Grice, S. (2000). Episodic memory and remembering in adults with Asperger syndrome. Journal of Autism and Developmental Disorders, 30, 305–316. Buckner, R. L., & Carroll, D. C. (2007). Self-projection and the brain. Trends in Cognitive Sciences, 11(2), 49–57. Burmann, B., Dehnhardt, G., & Mauck, B. (2005). Visual information processing in the lion-tailed macaque (Macaca silenus): Mental rotation or rotational invariance? Brain, Behavior and Evolution, 65(3), 168–176. Busby, J., & Suddendorf, T. (2005). Recalling yesterday and predicting tomorrow. Cognitive Development, 20(3), 362–372. Call, J. (2004). Inferences about the location of food in the great apes (Pan paniscus, Pan troglodytes, Gorilla gorilla, and Pongo pygmaeus). Journal of Comparative Psychology Copyright, 118, 232–241. Carlson, S. M., Moses, L. J., & Claxton, L. J. (2004). Individual differences in executive functioning and theory of mind: An investigation of inhibitory control and planning ability. Journal of Experimental Child Psychology, 87, 299–319. Churchland, P. M. (1984). Matter and consciousness: A contemporary introduction to the philosophy of mind. Cambridge, MA: MIT Press. A Bradford book. Clayton, N. S., & Russell, J. (2009). Looking for episodic memory in animals and young children: Prospects for a new minimalism. Neuropsychologia, 47(11), 2330–2340. Davies, J., Russell, C., & Russell, J. (2009). Making deferred imitation a measure of episodic memory. Unpublished manuscript, Department of Experimental Psychology, Cambridge University. Donaldson, W. (1996). The role of decision processes in remembering and knowing. Memory and Cognition, 24, 523–533. Dunn, J. C. (2004). Remember–know: A matter of confidence. Psychological Review, 111(2), 524–542. Ebbinghaus, H. (1885). Über das Gedächtnis. Leipzig: Duncker und Humblot. Estes, D. (1998). Young children’s awareness of their mental activity: The case of mental rotation. Child Development, 69(5), 1345–1360.
814
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
Ford, R., Shum, D., & Driscoll, T. (2009). Event-based prospective memory in young children: An individual difference analysis. Paper presented as part of the symposium on prospective memory at the BPS Cognitive Section 26th annual meeting, University of Hertfordshire, 1–3 September 2009. Gardiner, J. M., & Richardson-Klavehn, A. (2000). Remembering and knowing. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 229–244). Oxford, New York: Oxford University Press. Gilbert, D. T., & Wilson, T. D. (2007). Prospection: Experiencing the future. Science (New York, NY), 317(5843), 1351–1354. Goldman, A. I. (2006). Simulating minds: The philosophy, psychology, and neuroscience of mindreading. Oxford: Oxford University Press. Gopnik, A., & Astington, J. W. (1988). Children’s understanding of representational change and its relation to the understanding of false belief and the appearance-reality distinction. Child Development, 59, 26–37. Gordon, R. M. (1986). Folk psychology as simulation. Mind and Language, 1, 158–171. Hamilton, M., & Rajaram, S. (2003). States of awareness across multiple memory tasks: Obtaining a ‘‘pure” measure of conscious recollection. Acta Psychologica, 112, 43–69. Happé, F. (1994). An advanced test of theory of mind: Understanding of story characters’ thoughts and feelings by able autistic, mentally handicapped, and normal children and adults. Journal of Autism and Developmental Disorders, 24, 129–154. Hassabis, D., & Maguire, E. A. (2007). Deconstructing episodic memory with construction. Trends in Cognitive Sciences, 11(7), 299–306. Häuser, D., Kasielke, E., & Scheidereiter, U. (1994). Kindersprachtest für das Vorschulalter (KISTE). Bern, Göttingen: Hogrefe. Heal, J. (1986). Replication and functionalism. In J. Butterfield (Ed.), Language, mind, and logic (pp. 135–150). Cambridge: Cambridge University Press. Hill, E. L. (2004). Executive dysfunction in autism. Trends in Cognitive Sciences, 8, 26–32. Hollard, V., & Delius, J. (1982). Rotational invariance in visual pattern recognition by pigeons and humans. Science, 218(4574), 804–806. Jones, T. C., & Roediger, H. L. I. (1995). The experiential basis of serial position effects. European Journal of Cognitive Psychology, 7, 65–80. Kaminski, J., Riedel, J., Call, J., & Tomasello, M. (2005). Domestic goats, Capra hircus, follow gaze direction and use social cues in an object choice task. Animal Behaviour, 69, 11–18. Klein, S. B., Loftus, J., & Kihlstrom, J. F. (2002). Memory and temporal experience. The effects of episodic memory loss on an amnesic patient’s ability to remember the past and imagine the future. Social Cognition, 20, 353–379. Koriat, A., & Goldsmith, M. (1996). Memory metaphors and the real-life/laboratory controversy: Correspondence versus storehouse conceptions of memory. Behavioral and Brain Sciences, 19, 167–228. Kvavilashvili, L., Messer, D. J., & Ebdon, P. (2001). Prospective memory in children: The effects of age and task interruption. Developmental Psychology, 37, 418–430. Levine, B., Black, S. E., Cabeza, R., Sinden, M., Mcintosh, A. R., Toth, J. P., et al (1998). Episodic memory and the self in a case of isolated retrograde amnesia. Brain, 121(10), 1951–1973. Martin, M. G. F. (2001). Out of the past: Episodic recall as retained acquaintance. In C. Hoerl & T. McCormack (Eds.), Time and memory: Issues in philosophy and psychology (pp. 257–284). Oxford: Clarendon Press. Miklósi, A., Polgárdi, R., Topál, J., & Csányi, V. (2000). Intentional behaviour in dog-human communication: An experimental analysis of ‘‘showing” behaviour in the dog. Animal Cognition, 3, 159–166. Miscione, J. L., Marvin, R. S., O’Brien, R. G., & Greenberg, M. T. (1978). A developmental study of preschool children’s understanding of the words”know” and ‘‘guess”. Child Development, 49, 1107–1113. Moore, C., Barresi, J., & Thompson, C. (1998). The cognitive basis of future-oriented prosocial behavior. Social Development, 7, 198–218. Naito, M. (2003). The relationship between theory of mind and episodic memory: Evidence for the development of autonoetic consciousness. Journal of Experimental Child Psychology, 85, 312–336. O’Neill, D. K., Astington, J. W., & Flavell, J. H. (1992). Young children’s understanding of the role that sensory experiences play in knowledge acquisition. Child Development, 63, 474–490. Owens, D. (1996). A Lockean theory of memory experience. Philosophy and Phenomenological Research, LVI, 319–332. Ozonoff, S., Pennington, B. F., & Rogers, S. J. (1991). Executive function deficits in high-functioning autistic children: Relationship to theory of mind. Journal of Child Psychology and Psychiatry, 32, 1081–1105. Perner, J. (1990). Experiential awareness and children’s episodic memory. In W. Schneider & F. E. Weinert (Eds.), Interactions among aptitudes, strategies, and knowledge in cognitive performance (pp. 3–11). New York, Berlin, Heidelberg, London, Paris, Tokyo, Hong Kong: Springer-Verlag. Perner, J., Frith, U., Leslie, A. M., & Leekam, S. R. (1989). Exploration of the autistic child’s theory of mind: Knowledge, belief and communication. Child Development, 60, 689–700. Perner, J., Kloo, D., & Gornik, E. (2007). Episodic memory development: Theory of mind is part of re-experiencing experienced events. Infant and Child Development, 16, 471–490. Perner, J., & Ruffman, T. (1995). Episodic memory an autonoetic consciousness: Developmental evidence and a theory of childhood amnesia. Journal of Experimental Child Psychology, 59(3), 516–548 (special issue: early memory). Prencipe, A., & Zelazo, P. D. (2005). Development of affective decision making for self and other: Evidence for the integration of first- and third-person perspectives. Psychological Science, 16(7), 501–505. Roberts, W., Feeney, M., MacPherson, K., Petter, M., McMillan, N., & Musolino, E. (2008). Episodic-like memory in rats: Is it based on when or how long ago? Science, 320(5872), 113–115. Roberts, W. A. (2002). Are animals stuck in time? Psychological Bulletin, 128(3), 473–489. Roediger, H. L. I., & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory and Cognition, 21, 803–814. Rosenbaum, R. S., Stuss, D. T., Levine, B., & Tulving, E. (2007). Theory of mind is independent of episodic memory. Science, 318, 1257. Russell, J., Alexis, D., & Clayton, N. (2010). Episodic future thinking in 3- to 5-year-old children: The ability to think of what will be needed from a different point of view. Cognition, 114(1), 56–71. Sauvage, M. M., Fortin, N. J., Owens, C. B., Yonelinas, A. P., & Eichenbaum, H. (2008). Recognition memory: Opposite effects of hippocampal damage on recollection and familiarity. Nature Neuroscience, 11(1), 16–18. Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional objects. Science, 171, 701–703. Spreng, R. N., Mar, R. A., & Kim, A. S. (2008). The common neural basis of autobiographical memory, prospection, navigation, theory of mind, and the default mode: A quantitative meta-analysis. Journal of Cognitive Neuroscience, 21(3), 489–510. Sprung, M. (2008). Unwanted intrusive thoughts and cognitive functioning in kindergarten and young elementary school-age children following Hurricane Katrina. Journal of Clinical Child and Adolescent Psychology, 37(3), 575–587. Sprung, M., & Harris, P. (2009). Intrusive thoughts and young children’s knowledge about thinking following a natural disaster. Journal of Child Psychology and Psychiatry. Stich, K. P., Dehnhardt, G., & Mauck, B. (2003). Mental rotation of perspective stimuli in a California sea lion (Zalophus californianus). Brain, Behavior and Evolution, 61(2), 102–112. Stöttinger, E. (2006). Episodisches Gedächtnis – Erinnern als wirkliches Wiedererleben. Unpublished thesis, University of Salzburg. Stöttinger, E., Aigner, S., Hanstein, K., & Perner, J. (2009). Grasping the diagonal: Controlling attention to illusory stimuli for action and perception. Consciousness and Cognition, 18(1), 223–228. Stöttinger, E., Kaiser, W., & Perner, J. (2009). ‘‘Remember” judgments and the constraint of direct experience. Psychological Research – Psychologische Forschung, 73, 623–632. Stuss, D. T. (1991). Disturbance of self-awareness after frontal system damage. In G. P. Prigatano & D. L. Schacter (Eds.), Awareness of deficit after brain injury (pp. 63–83). Oxford: Oxford University Press.
J. Perner et al. / Consciousness and Cognition 19 (2010) 802–815
815
Suddendorf, T., & Busby, J. (2003). Like it or not? The mental time travel debate: Reply to Clayton et al.. Trends in Cognitive Sciences, 7, 437–438. Suddendorf, T., & Corballis, M. C. (1997). Mental time travel and the evolution of the human mind. Genetic, Social, and General Psychology Monographs, 123, 133–167. Suddendorf, T., & Corballis, M. C. (2007). The evolution of foresight: What is mental time travel, and is it unique to humans? The Behavioral and Brain Sciences, 30(3), 299–313 (discussion 313-51). Suddendorf, T., Nielsen, M. (2009). The development of foresight: Children’s capacity to remember a novel problem and to secure its future solution. Unpublished manuscript, University of Queensland. Tager-Flusberg, H. (1991). Semantic processing in the free recall of autistic children: Further evidence for a cognitive deficit. British Journal of Developmental Psychology, 9, 417–430. Taylor, M., Esbensen, B., & Bennett, R. T. (1994). Children’s understanding of knowledge acquisition: The tendency for children to report they have always known what they have just learned. Child Development, 65, 1581–1604. Tewes, U., Rossmann, P., & Schallberger, U. (1999). Hamburg-Wechsler-Intelligenztest für Kinder III (HAWIK-III). Handbuch und Testanweisung (3. überarbeitete und ergänzte Auflage). Bern: Verlag Hans Huber. Tulving, E. (1985). Memory and consciousness. Canadian Psychology, 26, 1–12. Vauclair, J., Fagot, J., & Hopkins, W. D. (1993). Rotation of mental images in baboons when the visual input is directed to the left cerebral hemisphere. Psychological Science, 4(2), 99–103. Wheeler, M. A., Stuss, D. T., & Tulving, E. (1997). Toward a theory of episodic memory: The frontal lobes and autonoetic consciousness. Psychological Bulletin, 121(3), 331–354. Wimmer, H., Hogrefe, J., & Perner, J. (1988a). Children’s understanding of informational access as source of knowledge. Child Development, 59, 386–396. Wimmer, H., Hogrefe, J., & Sodian, B. (1988b). A second stage in children’s conception of mental life: Understanding sources of information. In J. W. Astington, P. L. Harris, & D. R. Olson (Eds.), Developing theories of mind (pp. 173–192). New York: Cambridge University Press. Yonelinas, A. P. (1999). The contribution of recollection and familiarity to recognition and source-memory judgments: A formal dual-process model and an analysis of receiver operating characteristics. Journal of Experimental Psychology: Learning, Memory and Cognition, 25(6), 1415–1434.
Consciousness and Cognition 19 (2010) 816–828
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
ROC in animals: Uncovering the neural substrates of recollection and familiarity in episodic recognition memory q Magdalena M. Sauvage * Functional Architecture of Memory unit (www.rub.de/fam), Mercator Research Group, Faculty of Medecine, Ruhr University Bochum, Universitätsstraße 150, 44 801 Bochum, Germany Center for Memory and Brain, Boston University, 2 Cummington St., Boston, MA 02215, USA
a r t i c l e
i n f o
Article history: Available online 5 August 2010 Keywords: ROC Recollection Familiarity Episodic memory Hippocampus MEC
a b s t r a c t It is a consensus that familiarity and recollection contribute to episodic recognition memory. However, it remains controversial whether familiarity and recollection are qualitatively distinct processes supported by different brain regions, or whether they reflect different strengths of the same process and share the same support. In this review, I discuss how adapting standard human recognition memory paradigms to rats, performing circumscribed brain lesions and using receiver operating characteristic (ROC) methods contributed to solve this controversy. First, I describe the validation of the animal ROC paradigms and report evidence that familiarity and recollection are distinct processes in intact rats. Second, I report results from rats with hippocampal dysfunction which confirm this finding and lead to the conclusion that the hippocampus supports recollection but not familiarity. Finally, I describe a recent study focusing on the medial entorhinal cortex (MEC) that investigates the contribution of areas upstream of the hippocampus to recollection and familiarity. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction The contribution of two memory types to recognition memory function has been discussed since the time of Aristotle. One of these processes is described as a vague feeling of familiarity or ‘reminiscence’; for example when you recognize somebody, but cannot identify this person by name. The second, the recollection process, involves additional dimensions of memory; for example the spatial or temporal context in which this person was encountered (for a review see Yonelinas (2002)). Interestingly, the recollection and the familiarity processes are differentially affected in aging and in patients with amnesia. Indeed, the recollection process is strongly impaired in aging and amnesia, while the familiarity process is relatively spared (Barbeau et al., 2005; Brandt, Gardiner, Vargha-Khadem, Baddeley, & Mishkin, 2008; Daselaar, Fleck, Dobbins, Madden, & Cabeza, 2006; Duverne, Habibi, & Rugg, 2008; Düzel, Vargha-Khadem, Heinze, & Mishkin, 2001; Howard, Bessette-Symons, Zhang, & Hoyer, 2006; Peters & Daum, 2008; Prull, Dawes, Martin, Rosenberg, & Light, 2006; Quamme, Yonelinas, Widaman, Kroll, & Sauve, 2004; Turriziani, Serra, Fadda, Caltagirone, & Carlesimo, 2008; Vann et al., 2009; Yonelinas, Kroll, Dobbins, Lazzara, & Knight, 1998; but see Knowlton & Squire, 1995; Schacter, Verfaellie, & Anes, 1997). Hence, uncovering the neural substrates of the recollection and the familiarity processes could contribute to the characterization of new targets to rescue at least part of these deficits. A standard method to analyze human recognition memory performance is to use receiver q
This article is part of a special issue of this journal on Self, Other and Memory. * Functional Architecture of Memory unit (www.rub.de/fam), Mercator Research Group, Faculty of Medecine, Ruhr University Bochum, Universitätsstraße 150, 44 801 Bochum, Germany. Fax: +49 (0) 23432 14504. E-mail address: [email protected] 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.023
817
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
operating characteristic (ROC) functions (for a review see Yonelinas and Parks (2007)). In humans, episodic recognition memory is usually assessed by presenting a study list of items to a subject (for example a list of words appearing on a screen one at a time), and after a delay, presenting a longer list of items composed of the same items intermixed with an equal number of new items, also appearing one at a time. The probability of correct recognition of a study list item (p‘hit’) is plotted as a function of the probability of incorrect recognition of a ‘new’ item (p‘false alarm’) across confidence or bias levels, and the best fitting curve is defined to generate an ROC function (Yonelinas, 1999; see Fig. 1A and C for idealized human ROC curves for item and associative recognition memory, respectively).
B
1
Familiarity Index (F)
0.6
P(Hit)
0.8
l
ve
ce
le
an
ch
0.4
F
0.2
Recollection Index (R)
0.6 F
0.4
R
Probability Estimates
P(Hit)
0.8
1
Probability Estimates
A
0.2
0 0
0.2
0.4
0.6
0.8
1
0
0.2
D
1
0.6
0.8
1
1
0.8
0.6
0.6
P(Hit)
0.8
F
0.4
R
Probability Estimates
F
Probability Estimates
0.4
0.2
0.4
P(Fa)
P(Fa)
P(Hit)
0.2 0
0
C
R
0.4
0.2
R
0.4 0.2 0
0
0 0
0.2
0.4
0.6
0.8
1
P(Fa)
E
0
0.2
0.4
0.6
0.8
1
P(Fa)
1
P(Hit)
0.8
0.6
F Probability Estimates
0.4
0.2
R
0.4 0.2 0
0 0
0.2
0.4
0.6
0.8
1
P(Fa) Fig. 1. ROC functions for humans and rats recognition memory. Recollection (R) and familiarity (F) indices are shown as bar diagrams (see Yonelinas & Parks, 2007 for detailed calculations). (A) Ideal human ROC curve for single item recognition memory: the ROC curve is asymmetrical and curvilinear, reflecting the contribution of recollection and familiarity to recognition memory. (B) Similar ROC function observed in rats for single odor recognition memory (graph from Fortin et al., 2004). (C) Ideal human ROC curve for associative recognition memory: the ROC is asymmetrical and linear suggesting that recognition performance is based essentially on recollection. (D) Comparable ROC function observed in rats (graph from Sauvage, Fortin, Owens, Yonelinas, & Eichenbaum, 2008). (E) ROC function observed in rats for single odor recognition with a speeded response deadline demand (data from Sauvage, Beer, & Eichenbaum, 2010). The ROC is symmetrical and curvilinear, suggesting recognition relies primarily on familiarity.
818
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
Multiple models of recognition memory are based on the contribution of the familiarity and the recollection processes to recognition memory function (for a review see Yonelinas (2002)). Among them, two have been used extensively to study episodic recognition memory in humans: the dual-process model and the one-process model (for reviews see Wixted (2007) and Yonelinas and Parks (2007) respectively). The dual-process model describes familiarity and recollection processes as qualitatively distinct processes. In light of this model, the familiarity process is described as a rapid and continuous signal detection process sensitive to perceptual manipulations, while the recollection process is described as a slower and conceptually driven threshold process (Atkinson & Juola, 1973, 1974; Mandler, 1980; see for a review Yonelinas (2002)). Studies using the dual-process model of recognition memory also report that familiarity and recollection have different neural substrates: the hippocampus for recollection, the parahippocampal region for familiarity (Aggleton et al., 2005; Bowles et al., 2007; Yonelinas et al., 2002, 2007; see for reviews Diana, Yonelinas, and Ranganath (2007) and Eichenbaum, Yonelinas, and Ranganath (2007)). According to this model, two distinct indices can be generated from the analysis of ROC functions. The recollection index (R), y-intercept of the ROC function, which reflects the contribution of the recollection process to recognition memory performance, and the familiarity index (F), reflected by the degree of curvilinearity of the function, which reveals the contribution of familiarity to recognition memory performance (see Yonelinas, 1994; Yonelinas & Parks, 2007 for calculation of the indices). Hence, within the frame of the dual-process model, ROC functions are asymmetrical and curvilinear when familiarity and recollection contribute to recognition memory performance (Fig. 1A), asymmetrical and linear when the recollection process is primarily involved (Fig. 1C), or symmetrical and curvilinear when recognition memory is essentially based on the familiarity process (Fig. 5; Aged group). A major alternative to this model is the one-process theory that describes familiarity and recollection as qualitatively similar processes differing only in the strength of memory they reflect (see for a review Wixted (2007)). Familiarity would reflect weak memory, while recollection would reflect stronger memory or memory involving more information. As a consequence, recollection and familiarity would have a unique neural substrate, the hippocampus and could not vary independently (Manns, Hopkins, Reed, Kitchener, & Squire, 2003; Stark & Squire, 2000; Wais, Wixted, Hopkins, & Squire, 2006; see for a review Squire, Stark, and Clark (2004)). According to this view, the degree of curvilinearity of the ROC function reflects the sum of the strengths of memory components, and its asymmetry reflects greater variability in strength for the ‘old’ than for ‘new’ items. Among human ROC studies, some report that damage restricted to the hippocampus impairs specifically the recollection process (Aggleton et al., 2005; Quamme et al., 2004; Yonelinas et al., 2002), while others report that both processes are affected in patients with damage thought to be circumscribed to the hippocampus (Manns et al., 2003; Reed & Squire, 1997; Stark & Squire, 2000; Wais et al., 2006). This discrepancy resides mainly in the fact that the hippocampus and the parahippocampal region are adjacent brain structures. Indeed, identifying the precise extent of brain damage in patients with amnesia, or the precise source of brain activity within adjacent brain regions is beyond the spatial and neuropsychological resolution of standard techniques used currently in humans (e.g. functional and structural MRI imaging, and psychological tests). Thus, it has been suggested that the familiarity impairments accompanying the recollection deficits reported in some studies in patients with damage to the hippocampus resulted from additional damage to areas adjacent to the hippocampus (e.g. the parahippocampal region), rather than from damage to the hippocampus per se. Given the medial temporal lobe is exceptionally conserved across species (Manns & Eichenbaum, 2006), one way to clearly define whether the hippocampus supports familiarity as well as recollection is to perform lesion restricted to the hippocampus in animals, and assess the effect of this circumscribed lesion with behavioral paradigms that allow for the generation of distinct recollection and familiarity indices. We discuss this approach in the present review. A second major issue in human recognition memory is whether the parahippocampal region is functionally segregated in terms of its contribution to the recollection and the familiarity processes. Indeed, recent human and animal studies suggest that specific areas of the parahippocampal region, which are adjacent and strongly interconnected, contribute to different aspects of memory function (see Eichenbaum et al. (2007) for a review). The perirhinal cortex (PRc) and the lateral entorhinal cortex (LEC) would process information about the familiarity of individual items. In contrast, the postrhinal cortex (POR; parahippocampal cortex in humans) and the medial entorhinal cortex (MEC) would support recollection by representing the spatial and temporal context (whether the items are new or old) in which items have been experienced. However, this hypothesis could not be thoroughly tested in humans principally because of two reasons. One: it is not possible to determine the precise source of brain activity during recognition memory tasks in humans when areas are adjacent. Two: because cases showing restricted lesions to a single area of the parahippocampal region are extremely rare. In this review I will show how we addressed these two controversial issues by developing behavioral animal ROC paradigms that allow for recognition memory performance to be evaluated in a similar manner to the way that it is in humans (e.g. translational paradigms). Moreover, given these tasks are performed with animals, they present a key advantage over human ROC studies in that they can be combined with stereotactic surgery, which allows for brain areas to be damaged in a very restricted manner, while preventing additional damage to the adjacent brain structures. Using this approach, we first aimed at investigating the contribution of the hippocampus to recollection and familiarity by performing lesions circumscribed to the hippocampus, and defined whether the hippocampus supports recollection and familiarity, or recollection only. Second, we investigated the contribution of areas upstream of the hippocampus, more specifically the contribution of the MEC, to recollection and familiarity by performing lesion circumscribed to the MEC.
819
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
In the first part of this review, I will discuss how animal ROC paradigms contributed to bridging human and animal recognition memory, and brought evidence that recollection and familiarity are qualitatively distinct processes using intact rats. In a second part, I will show that this finding was confirmed in rats with impaired hippocampal function, and will show that animal ROC paradigms brought compelling evidence that the hippocampus supports recollection but not familiarity. In the last part of the review, I will report the first step of a series of studies investigating the functional segregation of the parahippocampal region, which focuses on the characterization of the contribution of the MEC to recollection and familiarity. 2. A translational model of episodic recognition memory A prerequisite for animal ROC paradigms to be appropriate translational models is to yield results comparable to those observed in humans, e.g. the contribution of familiarity and recollection to recognition memory performance in rats would be expected to be comparable to that of humans under the same experimental conditions. Within the frame of the dual-process model, a typical standard human ROC function for single item recognition memory reflects the contribution of the recollection and the familiarity processes, as shown by the asymmetry of the ROC function
A
Test list: 20 odors (10 old, 10 new)
Study list: 10 odors presented one at a time
1
3
2
, coriander
, sage
10
, paprika
30 min delay
... lemon
cumin
1
3
11
,
paprika
'old'
'new'
20
, ...
,
coriander
coffee
'old'
'new'
B
(2)
The odor is *new*: dig and retrieve reward
C
Bias levels
Cup sizes
(1)
The odor is *old*: repress (1) digging and retrieve reward in the back (2)
Reward for 'new' odors (buried in the cup for a correct response)
Reward for 'old' odors (at the back of the cage for a correct response)
5 4 3 2
1
Fig. 2. Example of ROC paradigm in rats. (A) Sequence of odor presentations on the sample and test phases (Fortin et al., 2004). (B) Non-matching-tosample rule. (C) Bias levels obtained by varying cup heights and payoff ratios of cereal rewards.
820
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
(y-intercept: R different from 0) and its curvilinearity (F different from 0) respectively (Fig. 1A; Yonelinas, 1994; see for a review Yonelinas and Parks (2007)). In 2004, Fortin and colleagues laid out the ground work for ROC studies in animals by readily adapting a standard human episodic recognition memory task to rats using their innate ability to discriminate odors and to forage (Fortin, Wright, & Eichenbaum, 2004). The animal behavioral ROC paradigm was designed to be as similar as possible to the human paradigm to minimize the potential effect of methodological differences on the interpretation of the data. In human and rat ROC paradigms, items are presented during a study phase, and after a delay the same items are presented again intermixed with new items. Correct recognition of the items presented during the study phase (a hit) and incorrect recognition of new items (a false alarm; fa) are assessed across five decision criteria, and the probability of a hit (phit) is plotted as a function of the probability of false alarm (pfa). Subsequently, the exact same analytical methods are used to evaluate the shape of human or rat ROC functions, and to generate recollection and familiarity indices (see Yonelinas, 1999 and Yonelinas & Parks, 2007 for details). Understandably though, stimulus modalities differed as they were defined to yield an optimal performance for each species. In humans, stimuli used are usually visual, while in rats stimuli used are olfactory. In addition, human subjects reported that they recognized a stimulus item as ‘old’ or ‘new’ verbally or by pressing appropriate keyboard keys, while rats were trained on a non-matching to sample rule to ‘show’ that they recognized the test stimuli as being ‘new’ or ‘old’. In more detail, the animal protocol was performed as follows: every day, a unique study list of 10 odors mixed in sand contained in cups would be presented to a given animal. Odors were common household spices (cumin, coriander etc.) chosen from a pool of 40 odors. Only 20 odors were used per session (10 ‘study’ odors, 10 ‘new’ odors; one session a day), and the study list changed every day because episodic recognition memory was investigated. Sampling of the odors during the study phase was ensured by placing a small piece of cereal buried in the cup. After a 30 min delay, recognition memory was tested as animals were presented with the same odors (‘old’ odors) intermixed with additional odors that had not been presented that day (‘new’ odors), but which the animal was highly familiar with (Fig. 2A). Animals were trained on a delay non-matching to sample rule, following which they had to dig in the cup to retrieve a food reward if the odor presented was ‘new’, or repress digging if the odors were ‘old’ and go to the back of the cage to collect a food reward (Fig. 2B). To prevent that rats solve the task by smelling the presence of the reward buried in the cups containing the ‘new’ odors, cups containing ‘old’ odors were baited with food rewards that were not accessible to the animals. In addition, spatial information could not be used to solve the task because all stimulus cups were presented at the same location (in the front of the cage), and the reward location was experienced only after each trial ended, e.g. after the recognition judgment was completed. Once animals were trained on the delayed non-matching to sample rule, recognition performance was assessed across five bias levels, ranging from conservative to liberal, by manipulating the amount of reward and the cup sizes (each bias type once a week in a pseudorandom order; Fig. 2C). Recognition memory performance was assessed by collecting the number of hits and the number of false alarms (fa) over 20 trials per session. Subsequently, the probability of hits and fas were calculated, and data averaged over four sessions for each bias level to plot and generate ROC functions for odor recognition memory (Fortin et al., 2004). In this study, Fortin and colleagues showed for the first time that the recollection and the familiarity processes contributed to single item recognition memory in rats, as it is the case in humans (Yonelinas, 1994). Indeed, performance in rats resulted in an asymmetrical ROC function, reflecting the contribution of the recollection process. In addition, the ROC function for odor recognition memory in rats was curvilinear, reflecting the contribution of the familiarity process (Fig. 1B). This finding brought the first concrete evidence that analyzing single item recognition memory performance with ROC methods yielded comparable results in humans and rats with familiarity and recollection contributing to single item recognition memory, thereby suggesting that the animal ROC paradigm for odor recognition was an appropriate translational tool to study single item recognition memory. With this in mind, we pursued the validation of the animal ROC paradigms by testing whether rat ROC functions varied in a similar manner to human ROC functions when subjected to the same experimental conditions. In addition, this manipulation allowed us to test the hypothesis that recollection and familiarity were qualitatively different processes. 3. Are recollection and familiarity distinct processes? evidence from intact rats The first controversial issue that we addressed using animal ROC paradigms was whether recollection and familiarity were qualitatively distinct processes, or whether they reflected different strengths of the same process. Compelling evidence that these processes are distinct would be that one process could support recognition memory performance without a significant contribution of the other. To test this hypothesis, we investigated recognition memory performance of intact rats under memory demands reported to favor the contribution of the recollection process or the contribution of the familiarity process to recognition memory. 3.1. Can recognition memory performance solely rely on the recollection process? In humans, episodic recognition memory for pairs relies primarily on the recollection process, as revealed by an ROC function which is asymmetrical, reflecting a strong contribution of the recollection process, and linear, which reflects the lack of contribution of the familiarity process to recognition performance (Fig. 1C; see Yonelinas, 2002 for a review). Although linear ROC functions have been reported in a number of independent studies focusing on associative recognition memory (Arndt & Reder, 2002; Kelley & Wixted, 2001; Rotello, Macmillan, & Van Tassel, 2000; Slotnick, Klein, Dodson, & Shimamura, 2000),
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
821
the finding of linear ROCs remained controversial because they have been rarely reported within the frame of the one model process (Wais et al., 2006). To determine whether the main process contributing to associative recognition memory in rats is the recollection process, we examined recognition memory performance of rats in an associative ROC paradigm involving the recognition of odors paired with media (Sauvage et al., 2008). Given that rats can separately attend to odors and media as distinct dimension when presented as an odor-medium pair (Birrell & Brown, 2000), we adapted the ROC protocol described in the previous section (Fortin et al., 2004) by presenting odor-medium pairs instead of simple odors. For example, among the 10 study pairs, cumin was mixed with beads and thyme with cotton balls. During the recognition phase, the same study pairs (old pairs) would be presented intermixed with ‘new’ rearranged odor-medium pairs (for example: cumin with cotton balls, thyme with beads). Using model-independent parameters (polynomial and linear regression functions) to assess the shape of the ROC function, and subsequently analyzing the data within the frame of the dual-process model, we reported that the rat ROC function for associative recognition memory was asymmetrical and linear, reflecting a recollection-based performance with no significant contribution of the familiarity process, as observed in humans (Fig. 1D; Sauvage et al., 2008). This result clearly showed that recognition memory performance could be achieved without a significant contribution of the familiarity process, supporting the claim that recollection and familiarity are qualitatively distinct processes. Note that partisans of the one-process model suggested that the linearity of the ROC function for associative recognition memory stemmed from the use of differential reward payoffs to manipulate response biases, rather than from memory demands (Wixted & Squire, 2008). Indeed, Wixted and colleague suggested that using different reward payoffs led to a ‘differential outcomes effect’, which in turn, was responsible for the linearity of the ROC function. A differential outcomes effect occurs when animals learn faster (as measured by an increase in performance accuracy), after hundreds of repetitions, stimulus–response reward combinations that yield a large reward than those that yield a smaller one. There are multiple reasons why this effect is not relevant to our protocol (see Eichenbaum, Sauvage, Fortin, & Yonelinas, 2008 for details). First, a simple visual inspection of the rat ROC function for single odor recognition memory reveals that it is possible to obtain a curvilinear ROC function using the exact same differential reward payoffs but different memory demands (compare the ROC for item recognition: Fig. 1B to the ROC for associative recognition: Fig. 1D). This clearly suggests that the shape of ROC functions is tied to memory demands and not to the use of differential payoff rewards. Second, the ‘differential outcomes effect’ requires hundreds of stimulus–response reward pairings for the preferential learning to take place. However, in our paradigm, stimuli are showed only once and the reward is received only after the recognition judgment is completed. Hence, there is no possibility for the animal to predict the amount of reward that will be received, and this to affect its performance. Last but not least, opposite to the predictions of the ‘differential outcomes effect’, accuracy is actually the lowest for the bias which yields the largest difference between rewards for the ‘new’ stimulus and the ‘old’ stimuli (Bias 1: 1=4 compared to 3 froot loops), while it is the highest for the bias for which there is NO difference between rewards for ‘old’ and ‘new’ stimuli (Bias 5: 1/2 froot loop in both cases). In conclusion, the differential outcomes effect is clearly not at work in our study, and the linearity of the rat ROC function for associative recognition memory is tied to memory demands and not to the use of differential payoff rewards. In the present study, we have reported that the ROC function for associative recognition in rats is linear, reflecting a strong contribution of the recollection process to the memory for pairs, without significant contribution of the familiarity process, which suggests that the recollection and the familiarity processes are qualitatively distinct processes. Moreover, given that linear ROC functions for associative recognition memory have been reported in the human literature (Arndt & Reder, 2002; Rotello et al., 2000; Slotnick et al., 2000; Yonelinas, 1997), even by the principal detractor of the dual-process model (Kelley & Wixted, 2001; see ROC functions for associative recognition with rearranged pairs), obtaining comparable ROC functions for associative recognition memory in rats constitutes a second step in the validation of animal ROC paradigms as proper translational tools to investigate the contribution of recollection and familiarity to recognition memory. In the next section, we focused on the missing piece of the puzzle, and studied whether recognition memory performance in rats could be achieved on the basis of the familiarity process only. 3.2. Can recognition memory performance solely rely on the familiarity process? A counterpart to the previous experiment was to examine recognition memory performance under conditions that favor the familiarity process. Familiarity is usually described as a rapid process that is based on pattern matching, and is sensitive to perceptual modulations, whereas recollection is characterized as a slower and conceptually driven process (see Mandler (2008) and Yonelinas (2002) for reviews). Indeed, studies involving process dissociation procedures (Yonelinas & Jacoby, 1994), response deadlines (Hintzman, Caulton, & Levitin, 1998; Hintzman & Curran, 1994; McElree, Dolan, & Jacoby, 1999) and evoked response potentials or electroencephalograms (Curran, 2004; Duarte, Ranganath, Trujillo, & Knight, 2006; Duzel, Yonelinas, Mangun, Heinzem, & Tulving, 1997; Smith, 1993; Woodruff, Hayama, & Rugg, 2006) reveal a two-component temporal function that includes a rapid familiarity process and a slower recollective process. Consistent with this view, limiting appropriately the latency to respond in a recognition memory task should allow for the familiarity process to be fully completed but prevent a significant contribution of the recollection process to recognition memory performance. To test this hypothesis, we first assessed odor recognition memory performance in rats without a deadline, and obtained a typical ROC function for odor recognition memory: asymmetrical and curvilinear, reflecting the contribution of the recollection and the familiarity processes to recognition memory performance. However, when a deadline was subsequently applied, by giving the animals half the time to respond, the ROC function remained curvilinear, reflecting the contribution of the familiarity process to recognition memory,
822
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
but became symmetrical (y-intercept not different from zero) suggesting that the recollection process did not contribute to recognition memory performance in a significant manner (Fig. 1E; Sauvage et al., 2010). These results suggested that under speeded conditions recognition memory performance relies essentially on the familiarity process, while the contribution of recollection is negligible, giving further support to the claim that recollection and familiarity are qualitatively distinct processes. Furthermore, this last experiment completed the validation of animal ROC paradigms as appropriate tools to study the contribution of familiarity and recollection to recognition memory in animals, given the animal ROC function under speeded conditions mirrored results obtained in the human literature under comparable conditions. In conclusion, in this part of the review, we reported a double dissociation of the contribution of familiarity and recollection to episodic recognition memory in intact rats. This finding reveals that the contribution of each process depends on memory demands, and strongly supports the hypothesis that recollection and familiarity are qualitatively distinct processes. Moreover, we validated animal ROC paradigms as appropriate translational tools to investigate the contribution of recollection and familiarity to recognition memory, given that animal and human ROC studies yielded comparable results. In the following sections, we report findings that emerged from animal studies using ROC paradigms combined with lesions restricted to target areas to address issues that remained controversial in humans because of the difficulty of defining the extent of brain damage in patients with amnesia, or because of the impossibility of dissociating the precise source of activity in adjacent brain areas in healthy subjects. 4. Does the hippocampus support the recollection and the familiarity processes? evidence from rats with hippocampal dysfunction A major controversy in recognition memory is whether the hippocampus supports the recollection process only or whether it also supports the familiarity process. Previous studies report a predominant role for the hippocampus in the recollection process while the parahippocampal region is suggested to primarily contribute to the familiarity process (Brown & Aggleton, 2001; Eichenbaum, Otto, & Cohen, 1994). However, conflicting results emerged from the ROC literature in humans. Evidence from studies on amnesic patients with damage restricted to the hippocampus suggests that the hippocampus specifically contributes to the recollection process but not to the familiarity process. Indeed, the y-intercept of the ROC function of patients with amnesia is significantly reduced compared to healthy subjects, while the curvilinear shape of the ROC function is maintained (Aggleton et al., 2005; Turriziani et al., 2008; Yonelinas et al., 1998, 2002; see for a review Eichenbaum et al. (2007)). In addition, studies focusing on specific areas of the parahippocampal region, the perirhinal cortex (PRc) and the entorhinal cortex (EC), report a preponderant role of these regions in the familiarity process (Bowles et al., 2007; Haskins, Yonelinas, Quamme, & Ranganath, 2008; Yonelinas et al., 2007; for reviews see Diana, Yonelinas, and Ranganath (2010) and Diana et al. (2007)). In striking contrast, other human studies report that both the recollection and the familiarity processes are affected following damage thought to be restricted to the hippocampus (Manns et al., 2003; see for a review Wixted and Squire (2004)), reflected by an alteration in both the asymmetry and the curvilinear shape of the ROC function (Wais et al., 2006). In addition, studies from the same group suggest that activity in the hippocampus and the PRc is correlated to the memory strength of remembered items, rather than to the specific contribution of these brain areas to the recollection or the familiarity processes to recognition memory (Shrager, Kirwan, & Squire, 2008; for a review see Squire, Wixted, and Clark (2007)). These findings left the field of human recognition memory divided regarding the contribution of the hippocampus to the recollection and the familiarity processes. The controversy principally stems from the fact that using standard functional or structural MRI imaging techniques, it is impossible to determine with precision whether brain activity (or brain damage) is truly circumscribed to the hippocampus, or whether it also extends to areas adjacent to the hippocampus (the parahippocampal region). A clear advantage of animal studies over human studies is that brain damage can be generated in a very controlled and restricted manner using stereotactic surgery techniques. Hence, as a first step to study the contribution of the hippocampus to recollection and familiarity, we performed lesions unequivocally circumscribed to the hippocampus, and assessed the effect of this selective lesion on the contribution of the recollection and the familiarity processes to item and associative recognition memory using animal ROC paradigms. In addition, we report later animal ROC findings related to the ethological model of reduced hippocampal function that is aging. Of note, data from these experiments have been analyzed with model-independent parameters (linear and polynomial regressions) to evaluate the shape of the ROC functions, with the dual-process model and with the one-process model (see Fortin et al., 2004; Robitsek, Fortin, Koh, Gallagher, & Eichenbaum, 2008; Sauvage et al., 2008 for details). In each experiment, analysis with the one-process model confirmed the dual-process findings by revealing that reducing hippocampal function selectively eliminates one parameter of the ROC function, specifically the inequality of variances between ‘old’ and ‘new’ items. By contrast, no change was observed in the other parameter of the ROC function, d0 , which reflects the difference in memory strength between ‘old’ and ‘new’ items. In summary, analysis with the one-process model confirmed that hippocampal lesions and aging selectively eliminated one parameter supporting the identification of old items, which is comparable to altering the recollection process in the dual-process theory, and left intact another parameter comparable to familiarity. 4.1. Contribution of the hippocampus to recollection and familiarity in item recognition memory? In 2004, Fortin and colleagues performed lesions restricted to the hippocampus in rats and reported that the ROC function of hippocampal-damaged rats remained curvilinear, reflecting the contribution of the familiarity process to recognition
823
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
A
B
1
Sham Hippocampus
0.6
F
0.2
0
0.2
0.4
F
0.2 0 S
0
R
0.4 Probability Estimates
0.4 Probability Estimates
P (Hit)
0.8
0.6
P(Fa)
H
S
0.8
H
1
0
0.2
0.4
R
0.4 0.2 0
0.6
0.8
1
P(Fa)
Fig. 3. Item recognition ROC in rats with circumscribed hippocampal lesion (graph from Fortin et al., 2004). (A) Hippocampal lesion eliminates recollectionbased performance while sparing the familiarity process. (B) ROC function in sham rats with extended memory delay. Extending the delay significantly affects familiarity not recollection.
memory. In addition, the ROC function became fully symmetrical suggesting that the recollection process did not contribute significantly to recognition memory performance of hippocampal-lesioned rats (Fig. 3A, Hippocampus). This finding was the first evidence that damage unequivocally restricted to the hippocampus significantly impaired the recollection process without affecting the familiarity process, which suggested that the hippocampus supports recollection but not familiarity. Of note, opponents of the dual-process model argued that hippocampal lesions did not specifically affect the recollection process but rather reduced overall memory performance because the accuracy level of the hippocampus group was slightly but significantly reduced compared to that of sham rats. To test this hypothesis, Fortin and colleagues lowered the overall memory performance of sham rats by increasing the delay between study and recognition phases (Yonelinas et al., 2002), and studied the effect of this manipulation on the contribution of recollection and familiarity to recognition memory. Against the predictions of the one-process model, the ROC function of sham rats became linear and asymmetrical (Fig. 3B) instead of curvilinear and symmetrical as observed following hippocampal lesion (Fig. 3A, Hippocampus). This manipulation clearly stated that reducing overall recognition memory performance affected the familiarity process and not the recollection process, and therefore could not account for the recollection deficit observed after hippocampal lesion. In summary, this study showed for the first time that the hippocampus supports recollection and not familiarity, given that lesions were specifically restricted to the hippocampus and only recollection was affected. Moreover, these data confirmed the results obtained in intact rats suggesting that recollection and familiarity are qualitatively distinct processes since the recollection and the familiarity processes were not affected in a similar manner following hippocampal lesion. 4.2. Contribution of the hippocampus to recollection and familiarity in associative recognition memory? To study further the contribution of the hippocampus to the familiarity and the recollection processes, we studied recognition memory performance for pairs using the associative ROC paradigm described in Sauvage et al., 2008. Rats were presented with a study list of odors paired with media, and recognition memory performance was subsequently assessed by presenting the same pairs intermixed with rearranged odor-medium pairs. As previously described, the ROC function for associative recognition memory was asymmetrical and linear, reflecting a recollection-based memory performance with no significant contribution of the familiarity process (Fig. 4, Sham). In striking contrast, recollection was significantly impaired following hippocampal lesion as reflected by a significant drop of the y-intercept (index of recollection, R), suggesting that the hippocampus is critical for recollection in associative recognition memory. Even more interesting, the ROC function of hippocampal- lesioned rats was now curvilinear reflecting a significant contribution of the familiarity process to recognition memory following hippocampal damage (Fig. 4, Hippocampus). This finding was the first clear evidence in the human and animal literature that the hippocampus did not support the familiarity process since the contribution of the familiarity process to recognition memory increased following hippocampal damage. An alternative interpretation of these data, in line with the predictions of the one-process model, was recently voiced. It has been argued that if the hippocampus supports familiarity and recollection, and if recollection judgements are ‘more demanding’ than familiarity judgments, and the lesion of the hippocampus is only partial, it could be possible that hippocampal lesions significantly affect the recollection process, but that enough of the hippocampus remains to complete the ‘less demanding’ familiarity judgments. As much as this could theoretically be the case in humans, for example in mild cases of stroke (assuming that damage remains in the vicinity of blood vessels or brain arteries, and potentially leaves part of the hippocampus and its connections intact), it is highly improbable in our animal studies given the type of lesion we perform. Indeed, our surgery aims at lesioning the hippocampus
824
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
1
P(Hit)
0.8
0.6 Sham Hippocampus F
Probability Estimates
0.4
0.2
R
0.4 0.2 0 S
H
S
H
0 0
0.2
0.4
0.6
0.8
1
P(Fa) Fig. 4. Associative recognition ROC of rats with restricted hippocampal lesion (graph from Sauvage et al., 2008). Hippocampal lesion affects recollection and familiarity in an opposite manner. Hippocampal damage decreases the recollection index (R: y-intercept) while increasing the familiarity index (F).
1
Young Aged
0.6 F
R
0.4 Probability Estimates
P(Hit)
0.8
0.2
0.4 0.2 0
Y
A
Y
A
0 0
0.2
0.4
0.6
0.8
1
P(Fa) Fig. 5. Item recognition ROC in aging (graph from Robitsek et al., 2008). Aging reduces the contribution of recollection to recognition performance while sparing familiarity.
by applying numerous small lesions (24 total; 12 per hemisphere) spread along the rostro-caudal, dorso-ventral and mediolateral axis. Hence, even if the lesion of the hippocampus is not total, the remaining hippocampal ‘bits’ are definitively disconnected and unlikely to be functional. Second, if in agreement with the single-process model, the hippocampus supports familiarity and recollection, the prediction of this model would be that hippocampal damage should also impair familiarity (even if it affects much more the recollection process), which is opposite to the results obtained in the present study, since a significant increase of the contribution of familiarity to recognition memory was reported. Finally, it is also important to underline that the specific recollection impairment observed in the Hippocampus group cannot be explained by a difference in memory strength between the Hippocampus and the Sham groups, since overall recognition memory performance did not significantly differ between the Hippocampus group and the Sham rats (Sauvage et al., 2008). Thus, the ‘memory strength hypothesis’ could not apply in the present study, and our data suggested that rats without a functional hippocampus recognized pairs to a similar level as sham rats but used an alternate strategy to solve the task. Interestingly, damage to the hippocampus in rats was reported to increase the tendency to unitize stimulus elements of a pair into a single stimulus; for example, lemon and sand could be encoded as lemon-scented sand by rats with a hippocampal lesion (Eichenbaum et al., 1994), and such a strategy was also observed in patients with amnesia. Indeed, amnesic patients who performed very poorly on memory tasks that use unrelated items as stimulus pairs, were reported to perform better on associative recognition memory tasks when given the possibility to rely more heavily on familiarity. For example, when compound words such as fireman, hardware or sleepwalk were used as stimulus pairs instead of unrelated words, and patients could consider both elements of a pair as a single item (Giovanello, Keane, & Verfaellie, 2006; Quamme, Yonelinas, & Norman, 2007; Turriziani, Fadda, Caltragirone, & Carlesimo, 2004). In summary, this study provided the first clear evidence that the hippocampus did not support familiarity, because the contribution of familiarity to recognition memory performance was enhanced following damage restricted to the hippocampus. It also revealed that the hippocampus supports the recollection process in associative recognition memory, as it is the case in item recognition memory. Furthermore, it strongly suggests that familiarity and recollection are qualitatively distinct processes because they can vary in an opposite manner.
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
825
4.3. Contribution of the hippocampus to recollection and familiarity in aging? Finally, the finding of a selective contribution of the hippocampus to recollection and not familiarity was further supported by results from a ROC study with aged rats, used as a model of reduced hippocampal function (Daselaar et al., 2006; Rosenzweig & Barnes, 2003; Wilson, Gallagher, Eichenbaum, & Tanila, 2006). In this study, the ROC function for item recognition of memory-impaired rats was found to be symmetrical and curvilinear, as it was the case following hippocampal lesion, suggesting that reducing hippocampal function limits the contribution of recollection to recognition memory while sparing the familiarity process (Fig. 5, Robitsek et al., 2008). Interestingly, as observed in associative recognition memory, the contribution of the familiarity process to recognition memory performance was increased in a subset of the memory-impaired rats, confirming in an ethologically relevant model that the hippocampus supports recollection and not familiarity, and that the two processes are qualitatively distinct since they can vary in an opposite manner. In conclusion, data from ROC studies in rats with reduced hippocampal function (lesion and aging studies) confirmed results in intact rats by showing that the recollection and the familiarity processes are qualitatively distinct, because they can vary independently when hippocampal function is compromised. In addition, these data brought convincing evidence that the hippocampus supports recollection but not familiarity, given the contribution of the recollection process to single item and associative recognition memory is significantly reduced following hippocampal lesion, whereas the contribution of familiarity is either unaltered or enhanced. In this section, I have discussed how animal ROC paradigms contributed to characterize the role of the hippocampus in recollection and familiarity in recognition memory by studying animals with hippocampal dysfunction. In the following section, I report how animal ROC paradigms were used to identify brain structures, upstream of the hippocampus, that provide information necessary to complete recollection-based-judgments to the hippocampus. 4.4. What areas provide information required for recollection-based-judgments to the hippocampus? Here, I will address a second issue that considerably fuels the current debate in recognition memory, which is the investigation of the functional segregation of the parahippocampal region in terms of recollection and familiarity. More precisely, I report here the first step of a series of studies which aim at characterizing the specific contribution of each area of the parahippocampal region to recollection and familiarity, starting with the medial entorhinal cortex (MEC), which has recently drawn a lot of attention for its dedicated role in spatial navigation and path integration (Fyhn, Molden, Witter, Moser, & Moser, 2004; Hafting, Fyhn, Molden, Moser, & Moser, 2005, see for reviews McNaughton, Battaglia, Jensen, Moser, and Moser (2006) and Moser and Moser (2008)). The parahippocampal region includes the perirhinal cortex (PRc), the parahippocampal cortex (PHc), the lateral entorhinal cortex (LEC) and the medial entorhinal cortex (MEC), which share intricate and bidirectional projections (see for a review Van Strien, Cappaert, and Witter (2009)). Despite this complex network of connections, data emerging from the human literature suggest that some of the areas of the parahippocampal region support the recollection process, while others contribute the familiarity process (for reviews see Diana et al. (2007, 2010) and Eichenbaum et al. (2007)). According to these studies, the PRc and the LEC would process information about the familiarity of individual items, whereas the PHc and the MEC would support recollection by representing spatial and temporal contexts (whether items are old or new) in which items have been experienced. Subsequently, both the item and contextual information would be combined within the hippocampus (see for a review Lipton and Eichenbaum (2008)). Thus, human fMRI studies report that activity in the PRc is correlated to familiarity-based judgments in recognition memory tasks (Davachi, Mitchell, & Wagner, 2003; Haskins et al., 2008; Henson, Cansino, Herron, Robb, & Rugg, 2003; Montaldi, Spencer, Roberts, & Mayes, 2006; Ranganath et al., 2004; Suchan, Gayk, Schmid, Köster, & Daum, 2008). In addition, ablation of the same structure, in a patient suffering from an intractable case of epilepsy, led to deficits specifically tied to the familiarity process (Bowles et al., 2007). In contrast, evidence from imaging studies points towards a preponderant role of the PHc and the hippocampus in the recollection of single items and associations (Aminoff, Gronau, & Bar, 2007; Bar, Aminoff, & Schacter, 2008; Cansino, Maquet, Dolan, & Rugg, 2002; Eldridge, Knowlton, Furmanski, Bookheimer, & Engel, 2000; Suchan et al., 2008; Woodruff, Johnson, Uncapher, & Rugg, 2005; Yonelinas et al., 2005). The contribution of the MEC and LEC to recollection and familiarity has been essentially extrapolated from their anatomical connections to the PHc and the PRc, because it is not possible with fMRI imaging techniques to dissociate the activity that occurs in the medial part of the EC during a memory task from that occurring in the lateral part of the EC. Moreover, no patients with damage restricted to the MEC or LEC have been reported to this date. Thus, as a part of the PHc–MEC complex, the MEC is suggested to play a preponderant role in the recollection process. Conversely, the LEC would contribute more specifically to the familiarity process, as a part of the PRc–LEC complex. However, direct evidence of these selective contributions is missing. To investigate the involvement of the MEC in the recollection process, we studied the effect of damage restricted to the MEC on the contribution of recollection and familiarity to odor recognition memory (Sauvage, Beer, Ekovich, Ho, & Eichenbaum, submitted for publication). Following MEC lesions, the ROC function remained curvilinear, reflecting a strong contribution of the familiarity process (Fig. 6, MEC). However, in sharp contrast to the sham rats, the ROC function of MEC rats became symmetrical (y-intercept not different from 0), suggesting that the recollection process no longer significantly contributed to the recognition performance. These results provided the first evidence that the MEC contributes specifically to recollection-based-judgments, whereas its contribution to the familiarity process is minimal. These findings are consistent
826
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
1
P(Hit)
0.8 Sham
0.6
MEC
0.4 Probability estimates
0.4
0.2
0.2
0 S
0
0
0.2
0.4
M
0.6
S
0.8
M
1
P(Fa) Fig. 6. Item recognition ROC of rats with restricted lesion to the MEC (graph from Sauvage et al., submitted for publication). MEC lesion reduces the contribution of recollection to recognition memory while familiarity remained unaffected.
with a broader view of the MEC function in recognition memory than its dedicated role in spatial representation, since MEC damage significantly altered performance in our non-spatial (odor) recognition memory task. In agreement with this finding, a recent electrophysiological study reported that dorsocaudal MEC (dcMEC) neurons had distinct activity patterns depending on whether rats turned left or right on a T-maze in an alternation working memory task (Lipton, White, & Eichenbaum, 2007). This finding suggested that dcMEC cells signaled which of the two episodes was ongoing, and were encoding spatial and temporal contexts (Lipton & Eichenbaum, 2008). Similar findings regarding the representation of spatial and temporal contexts were reported for the PHc, which is adjacent and strongly connected to the MEC. Indeed, the PHc was commonly found to be activated during the viewing of spatial environments, of objects that evoke strong contextual associations and during the viewing of objects that have strong temporal associations (Aminoff et al., 2007; Bar et al., 2008; Epstein & Kanwisher, 1998). Thus, the PHc which is strongly connected with the MEC was also suggested to represent the spatial and temporal context of remembered objects. In light of previous studies focusing on the selective contribution of the PRc to the familiarity process, the present study brought further support to the hypothesis of a functional segregation within the parahippocampal region, by showing that MEC is not involved in familiarity for specific items but, is a part of the parahippocampal and medial entorhinal network that is essential to memory for the context in which events occur, a defining feature of episodic recollection (Eichenbaum et al., 2007). This perspective is consistent with the observations on the dcMEC in spatial representation (grid cells), but suggests a broader role in the representation of spatial and temporal contexts which is of importance to episodic memory. In summary, results from this study suggested once more that recollection and familiarity are qualitatively distinct processes, since damage restricted to the MEC affected recollection while sparing familiarity. This study also allowed us to firmly conclude that the MEC contributes selectively to the recollection process, possibly by providing information regarding the temporal context of the remembered items to the hippocampus (whether items are ‘old ‘or ‘new’). This study is the first of a series of ROC studies in animals focusing on the investigation of the functional segregation of the parahippocampal region in terms of recollection and familiarity, and calls for further studies to characterize the contribution of other areas of the parahippocampal region, for example the LEC, for which virtually no data are available. In conclusion, developing ROC paradigms in animals contributed to bridging recognition memory in humans and animals by assessing and analyzing recognition memory performance in a comparable manner. In addition, combining circumscribed brain damage in animals to the use of animal ROC paradigms allowed for major controversies in human recognition memory to be addressed, and led consistently to the conclusion that familiarity and recollection are qualitatively distinct processes, and that the hippocampus supports recollection but not familiarity. Finally, we started to apply this approach to the characterization of the specific contribution of the parahippocampal areas to recollection and familiarity in recognition memory and are confident that this translational approach, combined to more standard approaches, will contribute to significant progress in the elucidation of the neural substrates of recognition memory. Acknowledgments We thank Jarret Frank for his help in Graphic designs and Zachery Beer for proof-reading. These experiments were Funded by MH71702, MH51520, MH52090 and AG09973. References Aggleton, J. P., Vann, S. D., Denby, C., Dix, S., Mayes, A. R., Roberts, N., et al (2005). Sparing of the familiarity component of recognition memory in a patient with hippocampal pathology. Neuropsychologia, 43(12), 1810–1823.
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
827
Aminoff, E., Gronau, N., & Bar, M. (2007). The parahippocampal cortex mediates spatial and nonspatial associations. Cerebral Cortex, 17(7), 1493–1503. Arndt, J., & Reder, L. M. (2002). Word frequency and receiver-operating characteristic curves in recognition memory: Evidence for a dual-process interpretation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(5), 830–842. Atkinson, R. C., & Juola, J. F. (1973). Factors influencing speed and accuracy of word recognition. In S. Kornblum (Ed.). Attention and performance (Vol. 4, pp. 583–612). New York: Academic Press. Atkinson, R. C., & Juola, J. F. (1974). Search and decision processes in recognition memory. In D. H. Krantz, R. C. Atkinson, & R. D. Luce (Eds.), Contemporary developments in mathematical psychology (pp. 243–293). San Francisco: W.H. Freeman. Bar, M., Aminoff, E., & Schacter, D. L. (2008). Scenes unseen: The parahippocampal cortex intrinsically subserves contextual associations, not scenes or places per se. Journal of Neuroscience, 28(34), 8539–8544. Barbeau, E. J., Felician, O., Joubert, S., Sontheimer, A., Ceccaldi, M., & Poncet, M. (2005). Preserved visual recognition memory in an amnesic patient with hippocampal lesions. Hippocampus, 15(5), 587–596. Birrell, J. M., & Brown, V. J. (2000). Medial frontal cortex mediates perceptual attentional set shifting in the rat. Journal of Neuroscience, 20, 4320–4324. Bowles, B., Crupi, C., Mirsattari, S. M., Pigott, S. E., Parrent, A. G., Pruessner, J. C., et al (2007). Impaired familiarity with preserved recollection after anterior temporal-lobe resection that spares the hippocampus. Proceedings of the National Academy of Sciences of the United States of America, 104(41), 16382–16387. Brandt, K. R., Gardiner, J. M., Vargha-Khadem, F., Baddeley, A. D., & Mishkin, M. (2008). Impairment of recollection but not familiarity in a case of developmental amnesia. Neurocase, 15(1), 60–65. Brown, M. W., & Aggleton, J. P. (2001). Recognition memory: What are the roles of the perirhinal cortex and hippocampus? Nature Reviews Neuroscience, 2, 51–61. Cansino, S., Maquet, P., Dolan, R. J., & Rugg, M. D. (2002). Brain activity underlying encoding and retrieval of source memory. Cerebral Cortex, 12(10), 1048–1056. Curran, T. (2004). Effects of attention and confidence on the hypothesized ERP correlates of recollection and familiarity. Neuropsychologia, 42, 1088–1106. Daselaar, S. M., Fleck, M. S., Dobbins, I. G., Madden, D. J., & Cabeza, R. (2006). Effects of healthy aging on hippocampal and rhinal memory functions: An event related fMRI study. Cerebral Cortex, 16, 1771–1782. Davachi, L., Mitchell, J. P., & Wagner, A. D. (2003). Multiple routes to memory: Distinct medial temporal lobe processes build item and source memories. Proceedings of the National Academy of Sciences of the United States of America, 100(4), 2157–2162. Diana, R. A., Yonelinas, A. P., & Ranganath, C. (2010). Medial temporal lobe activity during source retrieval reflects information type, not memory strength. Journal of Cognitive Neuroscience, 22(8), 1808–1818. Diana, R. A., Yonelinas, A. P., & Ranganath, C. (2007). Imaging recollection and familiarity in the medial temporal lobe: A three-component model. Trends in Cognitive Sciences, 11(9), 379–386. Duarte, A., Ranganath, C., Trujillo, C., & Knight, R. T. (2006). Intact recollection memory in high performing older adults: ERP and behavioral evidence. Journal of Cognitive Neuroscience, 18(1), 33–47. Duverne, S., Habibi, A., & Rugg, M. D. (2008). Regional specificity of age effects on the neural correlates of episodic retrieval. Neurobiology of Aging, 29(12), 1902–1916. Düzel, E., Vargha-Khadem, F., Heinze, H. J., & Mishkin, M. (2001). Brain activity evidence for recognition without recollection after early hippocampal damage. Proceedings of the National Academy of Sciences of the United States of America, 98(14), 8101–8106. Duzel, E., Yonelinas, A. P., Mangun, G. R., Heinzem, H. J., & Tulving, E. (1997). Event-related brain potential correlates of two states of conscious awareness in memory. Proceedings of the National Academy of Sciences of the United States of America, 94(11), 5973–5978. Eichenbaum, H., Otto, T., & Cohen, N. J. (1994). Two functional components of the hippocampal memory system. Behavioral and Brain Sciences, 17(449–472), 472–518. Eichenbaum, H., Sauvage, M., Fortin, N., & Yonelinas, A. P. (2008). ROCs in rats? Response to Wixted and Squire. Learning and Memory, 15(9), 691–693. Eichenbaum, H., Yonelinas, A. R., & Ranganath, C. (2007). The medial temporal lobe and recognition memory. Annual Review of Neuroscience, 30, 123–152. Eldridge, L. L., Knowlton, B. J., Furmanski, C. S., Bookheimer, S. Y., & Engel, S. A. (2000). Remembering episodes: A selective role for the hippocampus during retrieval. Nature Neuroscience, 3(11), 1149–1152. Epstein, R., & Kanwisher, N. (1998). A cortical representation of the local visual environment. Nature, 392(6676), 598–601. Fortin, N. J., Wright, S. P., & Eichenbaum, H. (2004). Recollection-like memory retrieval in rats is dependent on the hippocampus. Nature, 431(7005), 188–191. Fyhn, M., Molden, S., Witter, M. P., Moser, E. I., & Moser, M. B. (2004). Spatial representation in the entorhinal cortex. Science, 305(5688), 1258–1264. Giovanello, K. S., Keane, M. M., & Verfaellie, M. (2006). The contribution of familiarity to associative memory in amnesia. Neuropsychologia, 44(10), 1859–1865. Hafting, T., Fyhn, M., Molden, S., Moser, M. B., & Moser, E. I. (2005). Microstructure of a spatial map in the entorhinal cortex. Nature, 436(7052), 801–806. Haskins, A. L., Yonelinas, A. P., Quamme, J. R., & Ranganath, C. (2008). Perirhinal cortex supports encoding and familiarity-based recognition of novel associations. Neuron, 59(4), 554–560. Henson, R. N., Cansino, S., Herron, J. E., Robb, W. G., & Rugg, M. D. (2003). A familiarity signal in human anterior medial temporal cortex? Hippocampus, 13(2), 301–304. Hintzman, D. L., Caulton, D. A., & Levitin, D. J. (1998). Retrieval dynamics in recognition and list discrimination: Further evidence of separate processes of familiarity and recall. Memory & Cognition, 26(3), 449–462. Hintzman, D. L., & Curran, T. (1994). Retrieval dynamics of recognition and frequency judgments: Evidence for separate processes of familiarity and recall. Journal of Memory and Language, 331, 1–18. Howard, M. W., Bessette-Symons, B., Zhang, Y., & Hoyer, W. J. (2006). Aging selectively impairs recollection in recognition memory for pictures: Evidence from modelling and receiver operating characteristic curves. Psychology and Aging, 21, 96–106. Kelley, R., & Wixted, J. T. (2001). On the nature of associative information in recognition memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 27(3), 701–722. Knowlton, B. J., & Squire, L. R. (1995). Remembering and knowing: Two different expressions of declarative memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(3), 699–710. Lipton, P. A., & Eichenbaum, H. (2008). Complementary roles of hippocampus and medial entorhinal cortex in episodic memory. Neural Plasticity, 2008, 258467. Lipton, P. A., White, J. A., & Eichenbaum, H. (2007). Disambiguation of overlapping experiences by neurons in the medial entorhinal cortex. Journal of Neuroscience, 27(21), 5787–5795. Mandler, G. (1980). Recognizing: The judgment of previous occurrence. Psychological Review, 87, 252–271. Mandler, G. (2008). Familiarity breeds attempts: A critical review of dual process theories of recognition. Perspectives in Psychological Science, 3(5), 392–401. Manns, J. R., & Eichenbaum, H. (2006). Evolution of declarative memory. Hippocampus, 16, 795–808. Manns, J. R., Hopkins, R. O., Reed, J. M., Kitchener, E. G., & Squire, L. R. (2003). Recognition memory and the human hippocampus. Neuron, 37, 171–180. McElree, B., Dolan, P. O., & Jacoby, L. L. (1999). Isolating the contributions of familiarity and source information to item recognition: A time course analysis. Journal of Experimental Psychology: Learning, Memory, and Cognition, 253, 563–582. McNaughton, B. L., Battaglia, F. P., Jensen, O., Moser, E. I., & Moser, M. B. (2006). Path integration and the neural basis of the ‘cognitive map’. Nature Reviews Neuroscience, 7(8), 663–678. Montaldi, D., Spencer, T. J., Roberts, N., & Mayes, A. R. (2006). The neural system that mediates familiarity memory. Hippocampus, 16(5), 504–520. Moser, E. I., & Moser, M. B. (2008). A metric for space. Hippocampus, 18(12), 1142–1156.
828
M.M. Sauvage / Consciousness and Cognition 19 (2010) 816–828
Peters, J., & Daum, I. (2008). Differential effects of normal aging on recollection of concrete and abstract words. Neuropsychology, 22(2), 255–261. Prull, M. W., Dawes, L. L., Martin, A. M., III, Rosenberg, H. F., & Light, L. L. (2006). Recollection and familiarity in recognition memory: Adult age differences and psychology. Journal of Experimental Psychology: Human Learning and Memory, 28(6), 1499–1517. Quamme, J. R., Yonelinas, A. P., & Norman, K. A. (2007). Effect of unitization on associative recognition in amnesia. Hippocampus, 17(3), 192–200. Quamme, J. R., Yonelinas, A. P., Widaman, K. F., Kroll, N. E. A., & Sauve, M. J. (2004). Recall and recognition in mild hypoxia: Using covariance structural modeling to test competing theories of explicit memory. Neuropsychologia, 42, 672–691. Ranganath, C., Yonelinas, A. P., Cohen, M. X., Dy, C. J., Tom, S. M., & D’Esposito, M. (2004). Dissociable correlates of recollection and familiarity within the medial temporal lobes. Neuropsychologia, 42(1), 2–13. Reed, J. M., & Squire, L. R. (1997). Impaired recognition memory in patients with lesions limited to the hippocampal formation. Behavioral Neuroscience, 111, 667–675. Robitsek, R. J., Fortin, N. J., Koh, M. T., Gallagher, M., & Eichenbaum, H. (2008). Cognitive aging: A common decline of episodic recollection and spatial memory in rats. Journal of Neuroscience, 28(36), 8945–8954. Rosenzweig, E. S., & Barnes, C. A. (2003). Impact of aging on hippocampal function: Plasticity, network dynamics, and cognition. Progress in Neurobiology, 69, 143–179. Rotello, C. M., Macmillan, N. A., & Van Tassel, G. (2000). Recall-to-reject in recognition: Evidence from ROC curves. Journal of Memory and Language, 43(1), 67–88. Sauvage, M., Beer, Z., & Eichenbaum, H. (2010). Recognition memory: Adding a response deadline eliminates recollection but spares familiarity”. Learning & Memory, 17(2), 104–108. Sauvage, M., Beer, Z., Ekovich, M., Ho, L., & Eichenbaum, H. (submitted for publication) The medial entorhinal cortex: A selective role in recollection-based memory. Journal of Neuroscience. Sauvage, M., Fortin, N. J., Owens, C. B., Yonelinas, A. P., & Eichenbaum, H. (2008). Recognition memory opposite effects of hippocampal damage on recollection and familiarity. Nature Neuroscience, 11(1), 16–18. Schacter, D. L., Verfaellie, M., & Anes, M. D. (1997). Illusory memories in amnesic patients: Conceptual and perceptual false recognition. Neuropsychology, 11(3), 331–342. Shrager, Y., Kirwan, C. B., & Squire, L. R. (2008). Activity in both hippocampus and perirhinal cortex predicts the memory strength of subsequently remembered information. Neuron, 59(4), 547–553. Slotnick, S. D., Klein, S. A., Dodson, C. S., & Shimamura, P. (2000). An analysis of signal detection and threshold models of source memory. Journal of Experimental Psychology: Human Learning and Memory., 28(6), 1499–1517. Smith, M. E. (1993). Neurophysiological manifestations of recollective experience during recognition memory judgments. Journal of Cognitive Neuroscience, 5(1), 1–13. Squire, L. R., Stark, C. E. L., & Clark, R. E. (2004). The medial temporal lobe. Annual Review of Neuroscience, 27, 279–306 (Review). Squire, L. R., Wixted, J. T., & Clark, R. E. (2007). Recognition memory and the medial temporal lobe: A new perspective. Nature Reviews Neuroscience, 8(11), 872–883. Stark, C. E., & Squire, L. R. (2000). Functional magnetic resonance imaging (fMRI) activity in the hippocampal region during recognition memory. Journal of Neuroscience, 20, 7776–7781. Suchan, B., Gayk, A. E., Schmid, G., Köster, O., & Daum, I. (2008). Hippocampal involvement in recollection but not familiarity across time: A prospective study. Hippocampus, 18(1), 92–98. Turriziani, P., Fadda, L., Caltragirone, C., & Carlesimo, G. A. (2004). Recognition memory for single items and for associations in amnesic patients. Neuropsychologia, 42, 426–433. Turriziani, P., Serra, L., Fadda, L., Caltagirone, C., & Carlesimo, G. A. (2008). Recollection and familiarity in hippocampal amnesia. Hippocampus, 18(5), 469–480. Van Strien, N. M., Cappaert, N. L., & Witter, M. P. (2009). The anatomy of memory: An interactive overview of the parahippocampal–hippocampal network. Nature Reviews Neuroscience, 10(4), 272–282 (Review). Vann, S. D., Tsivilis, D., Denby, C. E., Quamme, J. R., Yonelinas, A. P., Aggleton, J. P., et al (2009). Impaired recollection but spared familiarity in patients with extended hippocampal system damage revealed by 3 convergent methods. Proceedings of the National Academy of Sciences of the United States of America, 106(13), 5442–5447. Wais, P. E., Wixted, J. T., Hopkins, R. O., & Squire, L. R. (2006). The hippocampus supports both the recollection and the familiarity components of recognition memory. Neuron, 49(3), 459–466. Wilson, I. A., Gallagher, M., Eichenbaum, H., & Tanila, H. (2006). Neurocognitive aging: Prior memories hinder new hippocampal encoding. Trends in Neurosciences, 29, 662–670. Wixted, J. T. (2007). Dual-process theory and signal-detection theory of recognition memory. Psychological Review, 114(1), 152–176. Wixted, J. T., & Squire, L. R. (2004). Recall and recognition are equally impaired in patients with selective hippocampal damage. Cognitive, Affective & Behavioral Neuroscience, 4(1), 58–66 (Review). Wixted, J. T., & Squire, L. R. (2008). Constructing receiver operating characteristics (ROCs) with experimental animals: Cautionary notes. Learning & Memory, 15(9), 687–690. Woodruff, C. C., Hayama, H. R., & Rugg, M. D. (2006). Electrophysiological dissociation of the neural correlates of recollection and familiarity. Brain Research, 1100(1), 125–135. Woodruff, C. C., Johnson, J. D., Uncapher, M. R., & Rugg, M. D. (2005). Content-specificity of the neuralcorrelates of recollection. Neuropsychologia, 43, 1022–1032. Yonelinas, A. P. (1994). Receiver-operating characteristics in recognition memory: Evidence for a dual-process model. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20(6), 1341–1354. Yonelinas, A. P. (1997). Recognition memory ROCs for item and associative information: The contribution of recollection and familiarity. Memory & Cognition, 25(6), 747–763. Yonelinas, A. P. (1999). The contribution of recollection and familiarity to recognition and source memory judgments: A formal dual-process model and an analysis of receiver operating characteristics. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25, 1415–1434. Yonelinas, A. P. (2002). The nature of recollection and familiarity: A review of 30 years of research. Journal of Memory and Language, 46, 441–517. Yonelinas, A. P., & Jacoby, L. L. (1994). Dissociations of processes in recognition memory: Effects of interference and of response speed. Canadian Journal of Experimental Psychology, 48(4), 516–535. Yonelinas, A. P., Kroll, N. E., Dobbins, I., Lazzara, M., & Knight, R. T. (1998). Recollection and familiarity deficits in amnesia: Convergence of remember-know, process dissociation, and receiver operating characteristic data. Neuropsychology, 12, 323–339. Yonelinas, A. P., Kroll, N. E., Quamme, J. R., Lazzara, M. M., Sauvé, M. J., Widaman, K. F., et al (2002). Effects of extensive temporal lobe damage or mild hypoxia on recollection and familiarity. Nature Neuroscience, 5, 1236–1241. Yonelinas, A. P., & Parks, C. M. (2007). Receiver operating characteristics (ROCs) in recognition memory: A review. Psychological Bulletin, 133(5), 800–832. Yonelinas, A. P., Widaman, K., Mungas, D., Reed, B., Weiner, M. W., & Chui, H. C. (2007). Memory in the aging brain: Doubly dissociating the contribution of the hippocampus and entorhinal cortex. Hippocampus, 17(11), 1134–1140.
Consciousness and Cognition 19 (2010) 829–837
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
It just felt right: The neural correlates of the fluency heuristic q Kirsten G. Volz a,b,*, Lael J. Schooler a, D. Yves von Cramon b,c a b c
Max Planck Institute for Human Development, Berlin, Germany Max Planck Institute for Neurological Research, Cologne, Germany Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany
a r t i c l e
i n f o
Article history: Received 30 November 2009 Available online 16 June 2010 Keywords: Fluency heuristic Decision making fMRI Claustrum
a b s t r a c t Simple heuristics exploit basic human abilities, such as recognition memory, to make decisions based on sparse information. Based on the relative speed of recognizing two objects, the fluency heuristic infers that the one recognized more quickly has the higher value with respect to the criterion of interest. Behavioral data show that reliance on retrieval fluency enables quick inferences. Our goal with the present functional magnetic resonance imaging study was to isolate fluency-heuristic-based judgments to map the use of fluency onto specific brain areas that might give a better understanding of the heuristic’s underlying processes. Activation within the claustrum for fluency heuristic decisions was found. Given that claustrum activation is thought to reflect the integration of perceptual and memory elements into a conscious gestalt, we suggest this activation correlates with the experience of fluency. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Simple heuristics can be used to exploit basic human abilities, such as recognition memory, to make decisions based on sparse information. One such heuristic is the fluency heuristic. Building on a long tradition of research on fluency (Jacoby & Dallas, 1981; Kelley & Jacoby, 1998), Schooler and Hertwig (2005) defined their fluency heuristic this way: If two objects are recognized, and one of the objects is more fluently retrieved, then infer that this object has the higher value with respect to the criterion, where retrieval fluency is defined as how long it takes to retrieve a trace from long-term memory. The fluency heuristic can help us make good inferences when all relevant objects are recognized and when there is a substantial correlation— in either direction—between the criterion and the retrieval fluency. For convenience, we assume a positive correlation for the remainder of this paper. For example, when choosing which of two recognized cities, say, the Japanese cities of Yokohama and Kyoto, has more inhabitants, one could achieve a reasonable level of accuracy if there is a substantial correlation between the ease of retrieving a city’s name and its population. The fluency heuristic works well to the extent that it can exploit relevant characteristics of the environment that are encoded in the relative accessibility of memory traces. The rationale behind the fluency heuristic is that memory performance reflects the patterns with which stimuli appear and reappear in the environment (Anderson & Schooler, 1991; Schooler & Anderson, 1997). Accordingly, the retrieval fluency associated with the recognition of stimuli correlates to a large extent with how frequently and recently relevant stimuli have been experienced (Hertwig, Herzog, Schooler, & Reimer, 2008). Since it is not currently possible to measure retrieval fluency directly, it has been operationalized by recognition latency, that is, how long it takes people to judge whether they recognize q
This article is part of a special issue of this journal on Self, Other and Memory. * Corresponding author. Address: Max Planck Institute for Human Development Research, Lentzeallee 94, 14195 Berlin, Germany. Fax: + 49 (0)30 82406 394. E-mail address: [email protected] (K.G. Volz). URL: http://www.cin.uni-tuebingen.de (K.G. Volz). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.05.014
830
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
an object or not (e.g., Hertwig et al., 2008). In short, differences in recognition latencies are taken as differences in retrieval fluencies. In a series of experiments, Hertwig et al. (2008) pursued several basic questions about how the fluency heuristic operates. More precisely, the authors showed that (a) recognition latencies are indeed indicative of criteria, such as city population or the wealth of athletes, (b) people can discriminate the recognition latencies of two objects that exceed a difference of 100 ms, (c) people’s judgments adhere more to the predictions of the fluency heuristic when differences between recognition latencies are large, which tends to happen when fluency is most predictive, (d) inferences in line with the fluency heuristic take less time than those that conflict with the heuristic, and (e) people’s inferences frequently accord with those predicted by the fluency heuristic, given that it is applicable. It can be argued, however, that accordance to the fluency heuristic is spurious, perhaps depending on a correlation between fast retrieval and knowledge about an object. So while people’s behavior may correspond to the fluency heuristic, the underlying process may be an entirely different knowledge-based strategy (Marewski & Schooler, 2010). Through a priming study, Hertwig et al. provided evidence that fluency does indeed guide inferences irrespective of other factors associated with fluency, such as the amount of knowledge known about the objects. Hertwig et al.’s main goal was to establish that the fluency heuristic does, in fact, guide decisions. Our aim was to uncover more about the processes underlying fluency-based decisions. One hypothesis draws on previous neuroscientific results on phenomenologically similar kinds of decisions. A second hypothesis focuses on an account of retrieval fluency that is based on successfully binding memory traces. The phenomenological experience of decisions based on the fluency heuristic has not been studied. The experience may be one of familiarity, the ‘‘felt-rightness” of a specific response, or an intuitive feeling about which option to choose. We start with Volz et al.’s (2006) investigation of the neural correlates of the recognition heuristic, which can be stated as follows: ‘‘If one of two objects is recognized and the other is not, then infer that the recognized object has the higher value with respect to a criterion” (Goldstein & Gigerenzer, 2002, p. 76). For this functional magnetic resonance imaging (fMRI) study, Volz et al. adapted Goldstein and Gigerenzer’s city judgment task in which participants had to infer which of two cities was larger. There were three critical types of trials, depending on whether both cities were recognized (RR trials), one of the cities was recognized (RU), or both cities were unrecognized (UU). After the fMRI experiment, participants were asked about what strategies they used in the three types of critical trials. For RR trials participants said they relied on knowledge if available, yet they said that such knowledge-based inferences were possible in only a fraction of these trials. When participants said they could not reach a decision based on knowledge, 17 of the 18 participants reported choosing the city that felt larger or more familiar or that they had made an informed guess. Together, these anecdotal data suggest that if knowledge was unavailable, participants relied on some other information. Their descriptions suggest that their judgments resemble the ‘‘intuitive assessment of the felt-rightness of a memory” (Schnyer, Nicholls, & Verfaellie, 2005, p. 837). The idea of an intuitive assessment of the felt-rightness of a memory is found in the literature on the ventromedial prefrontal cortex (VMPFC). For example, Schnyer et al. (2005) investigated feeling-of-knowing (FOK) judgments in an episodic memory task. Participants had to indicate the probability that they would recognize the final word of a sentence from a list of previously studied sentences. Prior to being shown the final word, the participants made an FOK judgment about whether they could retrieve the word on a 5-point scale. The instructions for using the rating scale were as follows: ‘‘Only press 5 if the answer pops effortlessly to mind. If you feel the answer is just there under the surface, press 4. . . . If you have no recollection of ever having seen the sentence, then press 1.” The authors found the VMPFC to be specifically engaged during accurate FOK judgments, defined as the four ratings on the scale that did not indicate certain knowledge of successful retrieval (i.e., 1–4 on the scale). In addition, the authors found a positive correlation between VMPFC activation and FOK judgments. When ‘‘know” ratings (i.e., 5 on the scale) were added to this analysis of retrievability, the amplitude levels for these ‘‘know” ratings did not continue the linear trend in the VMPFC but rather dropped off. Accordingly, the authors concluded that the VMPFC monitors the output of retrieval processes. Supporting this hypothesis, patients with lesions to the prefrontal cortex encompassing the VMPFC showed a clear impairment when making predictions about their subsequent recognition performance (Schnyer et al., 2004). Since the patients’ familiarity-based assessment was demonstrated to be intact, the results were taken to suggest that the VMPFC crucially subserves the assessment of the accessibility of memory contents. In addition, clinical and experimental findings showed the VMPFC to be crucially involved in ‘‘emotional decision making,” for example, when responses were based on feelings of rightness when deciding which option to pursue to gain or lose money or which action to judge as moral or immoral (Bechara, Tranel, & Damasio, 2002; Damasio, 2004; Greene & Haidt, 2002). Accordingly, previous findings may point to a crucial role of the VMPFC when decisions follow fluency. Schooler and Hertwig (2005) defined retrieval fluency in terms of how long it takes to retrieve a trace form long-term memory. Besides retrieval latency, other cognitive processes could contribute to a sense of retrieval fluency (Oppenheimer, 2008). One such process could be the ease with which associated memories are bound together. There is evidence that such a binding process would be associated with activation in the claustrum. Although activation within this area has rarely been reported in fMRI studies, Volz and von Cramon (2006), in a study on intuitive perceptual decisions, observed activation within the claustrum. They presented participants with fragmented line drawings of common objects. Within 400 ms, the participants had to indicate whether they perceived a coherent gestalt. A functional connectivity analysis revealed that the perceived gestalt was correlated with activation in the claustrum and medial orbitofrontal cortex (OFC). Medial OFC activation has been associated with a positive affective valence known to bias decisions (Bar et al., 2006; Kringelbach & Rolls, 2004; Volz, Rübsamen, & von Cramon, 2008). Further support for the critical role of the claustrum in binding can be found in anatomical and connectivity data (e.g., Crick & Koch, 2005; Fernández-Miranda, Rhoton, Kakizawa, Choi, & Alvarez-Linera, 2008;
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
831
Kowianski, Dziewiatkowski, Kowianska, & Morys, 1999). Because the claustrum integrates information within and across various modalities, Crick and Koch (2005) argued that the claustrum plays a key role in conscious experience. Additionally, the two-way connections between the claustrum and most, if not all, parts of the cortex as well as subcortical structures, which in turn have been suggested to subserve emotional processes, suggest that the claustrum binds disparate events into a single percept. Thus, the goal of the present fMRI study was to isolate judgments based on the fluency heuristic, so that we could map the use of fluency onto specific brain areas, which might give us a better understanding of the processes underlying adherence to this heuristic. As outlined above, it is difficult to tell whether a particular answer is based on knowledge or on fluency. To distill fluency-based judgments, we calculated two interaction contrasts specifically designed to tap fluency-based judgments (as described below) and the conjunction of these two interaction contrasts (test for a positive AND; Nichols, Brett, Andersson, Wager, & Poline, 2005). By drawing on previous results found by Hertwig et al. (2008), that people rely on the fluency heuristic when differences in recognition latencies are large and that inferences in line with the fluency heuristic take less time than inferences conflicting with the fluency heuristic, the first interaction is between differences in recognition latencies (large vs. small differences) and response time (individually determined fast vs. slow responses). The second interaction focuses on the factors that lead to participants’ judgments adhering to the predictions of the fluency heuristic (decisions agreeing vs. conflicting with the fluency heuristic) and incorrectness (incorrect vs. correct responses). Assuming knowledge-based strategies are more often right than wrong, looking at trials in which participants’ judgments adhered to the fluency heuristic but were wrong lowers the chances that their judgments were based on knowledge-based strategies. A conjunction of the two interaction contrasts should pull out trials on which judgments are more likely to be based on fluency than on knowledge-based strategies. As we are reporting new analyses of largely unanalyzed data from Volz and von Cramon (2006), we provide details of their methods. 2. Method 2.1. Participants Healthy, right-handed volunteers participated in the fMRI experiment (10 women, 8 men, mean age 25.6 years, SD 3.4, range 20–32 years). Informed consent was obtained prior to the experiment from each participant according to the Declaration of Helsinki. The local ethics committee of the University of Leipzig approved the experimental standards. Data were handled anonymously. 2.2. Stimuli, task, and experimental session On the left and right side of a screen two city names were presented simultaneously (horizontal visual angle 11°; vertical visual angle 1.7°). Participants had their left and right index fingers on left and right response buttons spatially corresponding to the stimulus locations on the screen. Within each trial a cue was presented for 500 ms, indicating the beginning of the next trial, followed by the presentation of a fixation cross for 500 ms; thereafter the two city names were presented for a maximum of 4 s during which participants’ response and reaction time were recorded. As soon as participants indicated their choice with a button press, the city names disappeared and a fixation cross was presented until the next trial started. No performance feedback was delivered whatsoever. The participants’ task (i.e., the inference task) was to indicate which city in each pair had the larger population. Each session contained 218 trials, consisting of 140 critical trials plus 48 filler trials and 30 null events, in which no stimulus was presented and the Blood Oxygen Level-Dependent BOLD response was allowed to return to a baseline state. In the filler trials participants had to indicate which of two presented words contained more vowels. All trials lasted for 8 s (i.e., four scans at a repetition time (TR) of 2 s). The onset of each stimulus presentation relative to the beginning of the first of the four scans was randomly varied (0, 500, 1500 ms) to enhance the temporal resolution of the signal captured (Birn, Cox, & Bandettini, 2002; Miezin, Maccotta, Ollinger, Petersen, & Buckner, 2000). Participants were unaware of this modulation. Following the fMRI session (i.e., outside the scanner), participants completed a recognition test in which they were presented with each particular city name and had to indicate whether they knew each city already before the experimental session. It was emphasized that participants should declare as recognized only those cities that they had heard of before the functional session. The data of the recognition test were used first to individually determine trial types in the inference task, for example, whether both cities were recognized (recognize–recognize, or RR trials), neither city was recognized (unrecognized–unrecognized, or UU trials), or one of the two cities was recognized and the other not (recognized–unrecognized, or RU trials), and second to determine recognition latency for each stimulus in the inference task. On the basis of these data, we determined which of the RR trials in the inference task were solved as predicted by the fluency heuristic, that is, when the city that was recognized faster (as determined from the recognition test) was chosen as the larger city. Following the recognition test, participants were requested to fill out a questionnaire asking for strategies; subsequently they were debriefed and thanked. Our experimental design called for the specific order of city task and recognition test, rather then counterbalancing the task order. Having the recognition test before the city task could have biased participants by making salient that we (the
832
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
experimenters) were interested in whether they recognized the stimuli. In addition, by keeping this fixed task order, we were able to reliably measure the hemodynamic activity elicited by recognition judgments. Had the recognition test come first, the participants would have had to judge not only whether they recognized the city, but also whether the source of the recognition was just from the experiment or possibly from elsewhere. The additional demands of this discrimination task mean that the recognition judgments from the two task orders could draw on somewhat different brain structures and so involve brain structures that would not otherwise be involved in the application of the fluency heuristic. Nevertheless, we expected that in the recognition test, participants could reliably report whether they recognized the city from the experiment or from elsewhere. That is, they would rarely mis-categorize as recognized a city that they had only encountered in the experiment. Support for this assumption come from the studies by Pohl (2006) and Goldstein and Gigerenzer (2002). Neither study found any differences in the recognition rates that depended on the task order. However, we would expect that the recognition latencies would be faster, because the cities would have been recently seen in the inference task. As a result, we may be underestimating the absolute differences in retrieval fluency between items, but the relative differences should be preserved. For a detailed description of how the city pairs were generated, please see Volz and von Cramon (2006). 2.3. Data acquisition Imaging was performed on a 3T scanner (Siemens TRIO, Erlangen, Germany). Twenty-two axial slices (4 mm thickness, 20% spacing, field of view [FOV] 19.2 cm, data matrix of 64 64 voxels, and in-plane resolution of 3 3 mm) parallel to the bicommissural plane (AC–PC) covering the whole brain were acquired using a single-shot echo-planar imaging (EPI) sequence (TR 2 s, echo time [TE] 30 ms, flip angle 90°). One functional run with 872 time points was run with each time point sampling over the 22 slices. Prior to the functional runs, 22 anatomical T1-weighted modified driven equilibrium Fourier transform (MDEFT; Norris, 2000; Ugurbil et al., 1993) images (data matrix of 256 256 voxels, TR 1.3 s, TE 10 ms) were acquired as well as 22 T1-weighted EPI images with the same spatial orientation as the functional data. The latter were used to coregister the functional scans with previously acquired high-resolution full-brain 3-dimensional brain scans. 2.4. Data evaluation The functional imaging data were processed using the software package LIPSIA (Lohmann et al., 2001). Functional data were motioncorrected off-line with the Siemens motion–correction protocol. To correct for the temporal offset between the slices acquired in one scan, a cubic spline interpolation was applied. A temporal high-pass filter with a cut-off frequency of 1/160 Hz was used for baseline correction of the signal and a spatial Gaussian filter with 5.65 mm full-width half-maximum (FWHM) was applied. The anatomical slices were coregistered with the high-resolution full-brain scan that resided in the stereotactic coordinate system and then transformed by linear scaling to a standard size (Talairach & Tournoux, 1988). The transformation parameters obtained from this step were subsequently applied to the preprocessed functional slices so that the functional slices were also registered into the stereotactic space. This linear normalization process was improved by a subsequent processing step that performed an additional nonlinear normalization known as ‘‘demon matching.” In this type of nonlinear normalization, an anatomical 3-dimensional data set (i.e., the model) is deformed such that it matches another 3-dimensional anatomical data set (i.e., the source) that serves as a fixed reference image (Thirion, 1998). Voxel size was interpolated during coregistration from 3 3 4 mm to 3 3 3 mm. The statistical evaluation was based on a leastsquares estimation using the general linear model (GLM) for serially autocorrelated observations (random effects model; Friston, Frith, Turner, & Frackowiak, 1995; Worsley & Friston, 1995). The general linear regression performs a ‘‘precoloring” of the data; that is, it applies a temporal Gaussian smoothing with a user-specified kernel width given by the parameter FWHM. The smoothing imposes a temporal autocorrelation that determines the degrees of freedom. An event-related design was implemented; that is, the hemodynamic response function was modeled by means of the experimental conditions for each stimulus (event being onset of stimulus presentation). The design matrix was generated using a synthetic hemodynamic response function and its first and second derivative (Friston et al., 1998) and a response delay of 6 s. The model equation, including the observation data, the design matrix, and the error term, was convolved with a Gaussian kernel of dispersion of 4 s FWHM to deal with the temporal autocorrelation (Worsley & Friston, 1995). Contrast images, that is, estimates of the raw score differences between specified conditions were generated for each participant. The single-subject contrast images were entered into a second-level analysis based on Bayesian statistics (Neumann & Lohmann, 2003). In Neumann and Lohmann’s approach, posterior probability maps and maps of the effect size for the effects of interest in groups of participants are calculated on the basis of the resulting least-squares estimates of parameters for the GLM. The output of the Bayesian second-level analysis is a probability map showing the probability of the contrast being larger than zero. For visualization, a threshold of 99% was applied to the probability maps. For each participant all contrasts of interest were calculated. Reasons to use Bayesian second-level analysis for fMRI data are manifold: A comparison between the established analysis based on t statistics and Bayesian second-level analysis showed that the latter is more robust against outliers. Furthermore, the Bayesian approach overcomes some problems of null hypothesis significance testing, such as the need to correct for multiple comparisons, and this approach provides estimates for the size of an effect of interest as well as for the probability that the effect occurs in the population (Neumann & Lohmann, 2003).
833
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
The main purpose of the fMRI study was to reveal brain areas involved in fluency-heuristic-based decision processes. The first analysis therefore included regressors for trials with large differences in recognition latencies, trials with small differences in recognition latencies, trials that were answered quickly (individually determined as trials below the median response time (RT)), trials that were answered slowly (individually determined as trials above the median RT), RU trials, and UU trials. Concerning the factor differences in recognition latencies, we classified trials with latencies > 400 ms as trials with large differences and trials with latencies > 100 ms and 6400 ms as trials with small differences following Hertwig et al. (2008). A second analysis included regressors for trials that were and were not answered according to the fluency heuristic (factor fluency), correctly and incorrectly answered trials (factor incorrectness), RU trials, and UU trials. The analysis that was conducted as a manipulation check for recognition- and retrieval-related activations was conducted by building special contrasts within the two described models. Note that in all analyses only trials in which the difference in recognition latencies between two recognized cities equaled or exceeded a just noticeable difference (JND) of 100 ms were included (see below for the rationale). 3. Results 3.1. Behavioral results On average, participants faced 55 RR trials (range: 34–76), 46 RU trials (range: 38–61), and 39 UU trials (range: 19–62). Volz and von Cramon (2006) specifically investigated the RU trials, where by definition the recognition heuristic is applicable but the fluency heuristic is not. Here we focus on the RR trials, where the fluency heuristic is applicable but the recognition heuristic is not. As outlined above, in this contribution, we are exclusively interested in how people deal with situations in which they recognize both of two presented objects and have to arrive at a decision about a criterion, which is very likely unknown (e.g., city size). Hence, in the following section on behavioral data we concentrate on RR trials only. For an overview of the behavioral results, please refer to Table 1. 3.2. The validity of the fluency heuristic in the present sample To determine how ecologically valid retrieval fluency was in our sample, we quantified the strength of the relationship between the retrieval fluency and the criterion as the proportion of times a more quickly recognized city indeed had a higher criterion value than the city that required more time to be recognized. Thus, fluency validity is calculated as: vf = Rf/(Rf + Wf), where Rf is the number of correct inferences made by relying on the fluency heuristic, and Wf is the number of incorrect inferences made by relying on the fluency heuristic. In the present study, the mean fluency validity was .55 (95% confidence interval [CI] = 0.497, 0.588) and exceeded chance level (.50), t(17) = 2.13; p = .048. Note that in this analysis only trials in which the difference in recognition latencies between two recognized cities equaled or exceeded a JND of 100 ms were included. The JND of 100 ms was determined following Fraisse’s (1984) suggestion, based on a thorough review of the timing literature, that durations of less than 100 ms are perceived as instantaneous. Hertwig et al. (2008) provided supporting results by showing that when differences in recognition latencies were shorter than 100 ms, people’s ability to discriminate between recognition latencies (of the two objects) dropped close to chance level. To be able to directly compare our results with those of Hertwig et al. (2008), we categorized the objective difference in recognition latencies into four equal bins: 0–99 ms, 100–399 ms, 400–699 ms, and > 700 ms and calculated fluency validity as a function of these four bins (see Table 1). For the first three bins, we replicated Hertwig et al.’s findings showing that there is a tendency that the larger the objective difference in recognition latencies, the higher the fluency validity. Yet, the linear trend did not continue for the last bin (>700 ms), F(3, 14) = .508; p = .634. Given recent results on fluency validities in different environments, fluency validity in the present sample can be considered moderate and hence participants could at least theoretically infer the distal properties of the world (Hertwig et al., 2008). 3.3. Participants’ accordance with the fluency heuristic To determine to what degree people’s inferences agreed with the fluency heuristic in the present sample, we computed for each participant the percentage of inferences that were in line with the fluency heuristic among all cases in which it could Table 1 Means (SD) of fluency heuristic validity, fluency heuristic accordance, percent correct overall, and percent correct when applying the fluency heuristic.
Fluency heuristic validity Fluency heuristic accordance % Correct % Correct with the fluency heuristic a
Overalla
0–99 ms
100–399 ms
400–699 ms
>700 ms
0.55 0.68 62.8 81.3
0.49 0.48 60.4 54.2
0.53 0.64 65.3 79.3
0.58 0.75 63.6 83.7
0.53 0.74 59.5 80.9
(0.21) (0.05) (19.5) (14.8)
(0.13) (0.13) (20.4) (32.9)
Data for this category exclude trials with a just noticeable difference of 100 ms or less.
(0.13) (0.09) (11.1) (10.2)
(0.24) (0.17) (23.3) (14.9)
(0.26) (0.17) (24.2) (19.2)
834
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
Interaction difference in re cognition latencies & response time
A
x = 31 claustrum
Interaction fluency and incorrectness
B
RCZ
y=6 claustrum
x = 34 claustrum
Conjunction analysis of the contrasts shown in A & B
C
y=6
claustrum
x = 31
Fig. 1. Group averaged activations are shown on four slices taken from an individual brain normalized and aligned to the Talairach stereotactic space. For visualization a threshold of 99% was applied to all probability maps. (A) Results of the interaction contrast differences in recognition latencies (large vs. small) and response time (individually determined fast slow responses). To understand which factor combination is driving the effects, the mean percent signal changes with standard errors are reported for all four conditions. (B) Results of the interaction contrast fluency heuristic (decisions agreeing vs. conflicting with the fluency heuristic) and incorrectness (incorrect vs. correct responses). To understand which factor combination is driving the effects, the mean percent signal changes with standard errors are reported for all four conditions. (C) Result of the conjunction analysis of the two interaction contrasts. Abbreviations: amla: above median and large differences in recognition latencies; amsm: above median and small differences in recognition latencies; bmla: below median and large differences in recognition latencies; bmsm: below median and small differences in recognition latencies; RCZ: rostral cingulate zone; flc: fluency heuristic trials and correct; fli: fluency heuristic trials and incorrect; nflc: no fluency heuristic trials and correct; nfli: no fluency heuristic trials and incorrect.
835
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
be applied (i.e., among all RR trials), excluding trials with a JND of 100 ms or smaller. The mean fluency heuristic accordance in the present sample was .678 (CI: 0.649, 0.707). The interindividual variation in the proportion of judgments that agreed with the fluency heuristic was rather moderate. The rate of individuals’ fluency heuristic accordance ranged between .58 and .79. Hence, none of the participants appeared to have systematically decided against fluency (i.e., fluency accordance rate below .50), nor did anyone use the fluency heuristic all the time. The fluency heuristic applicability ranged between .24 and .54 with a mean of .39 (CI: 0.343, 0.437) and hence in more than a third of all inferences, the fluency heuristic was applicable. In Table 1, we report fluency heuristic accordance rates as a function of differences in recognition latencies. Accordance rates increased with larger differences in recognition latencies, F(3, 14) = 11.67; p = .0001, but did not differ much between the last (>700 ms) and the second to last (400–699 ms) bin, t(17) = .876. Again, except for the last bin, we replicated the results of Hertwig et al. (2008). 3.4. Participants’ performance in the inference task Overall, participants scored a median .63 accuracy (CI: 0.580, 0.683) calculated as percent correct of all RR trials irrespective of response strategy (trials with a JND of 100 ms or smaller were excluded). In Table 1, we report percent correct of all RR trials as a function of differences in recognition latencies. Overall, performance did not differ substantially subject to differences in recognition latencies, F(3, 14) = .322; p = .76. Yet, this pattern changes when we look at percent correct made by the fluency heuristic: There is a clear tendency that the larger the objective difference in recognition latencies, the higher the percent correct, F(3, 11) = 5.50; p = .010. Again, the linear trend did not continue for the last bin. Inferences in line with the fluency heuristic were made faster (mean RT: 2842 ms) than inferences conflicting with the fluency heuristic (mean RT: 2944 ms), t(17) = 3.47, p = .003. In all analyses where the variable of interest was split according to differences in recognition latencies, the last bin (>700 ms) seemed to be the odd one out. One reason for this finding might be the restriction of response time. Before the experimental session, participants were informed that the two city names would be presented for a maximum of 4 s during which their response would be recorded. If no response was given within the 4 s, then the trial would be counted as a noresponse trial. Participants’ mean response time was 3055 ms when differences in recognition latencies exceeded 700 ms. Thus, when it took participants fairly long to recognized one of the city names, the overall decision time was almost expended. Accordingly, this time pressure might have eventually led to guessing. 3.5. Imaging results To test for the specific neural correlates of fluency-heuristic-based decision processes, we calculated two interaction contrasts and their conjunction. Based on previous findings on fluency-heuristic-based decisions (Hertwig et al., 2008), the interaction contrasts were designed to capture fluency-based inferences, relatively uncontaminated by extensive knowledge about the options. 3.5.1. Interaction between differences in recognition latencies and response time The interaction contrast of the factors large versus small differences in recognition latencies and fast versus slow response times revealed activation bilaterally within the dorsal claustrum and the left anterior insula; bilaterally within the anterior thalamus region and amygdala; and within the anterior portion of the left superior temporal gyrus, left supramarginal gyrus, and right cuneus (see Fig. 1A and Table 2). When we plotted the mean percent signal change in the dorsal claustrum, results revealed specifically those trials with large differences in recognition latencies and fast response times to elicit activation within this area. 3.5.2. Interaction between fluency and incorrectness The interaction contrast of the factors fluency heuristic trials versus no fluency heuristic trials and incorrect versus correct responses revealed activation within the right dorsal claustrum and bilaterally within the posterior rostral cingulate zone Table 2 Anatomical specification, cluster size (mm3), and Talairach (x, y, z) coordinates of significantly activated voxels of the interaction contrast differences in recognition latencies (large vs. small differences) and response time (fast vs. slow responses). Anatomical specification
mm3
L dorsal claustrum R dorsal claustrum L anterior insula R anterior thalamus region L anterior thalamus region L superior temporal gyrus R amygdala L amygdala L supramarginal gyrus R cuneus (occipital lobe)
81 81 162 1755 621 324 108 81 162 351
Note: Only clusters of at least five contiguous voxels are reported. L: left, R: right.
x
y 35 31 26 13 11 50 19 23 47 4
z 6 4 15 3 12 36 6 9 48 84
3 3 12 12 9 9 9 9 27 27
836
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
(RCZ) and the left superior and middle temporal gyrus (see Fig. 1B and Table 3). When we plotted the mean percent signal change in the right dorsal claustrum, results revealed both incorrect fluency trials and correct nonfluency trials to specifically elicit activation within this area. 3.5.3. Conjunction analysis of the two interaction contrasts To test for those regions that are commonly activated across the two interaction contrasts, we calculated a conjunction analysis, that is, a test for a logical AND (Nichols et al., 2005). Activation was revealed solely within the right dorsal claustrum (x = 31, y = 6, z = 9; mm2 = 135; see Fig. 1C). 4. Discussion The present imaging study was conducted to investigate the cognitive processes underlying the fluency heuristic. Simply stated, when people use the fluency heuristic they will infer that the more easily retrieved item has the higher criterion value (Schooler & Hertwig, 2005). Our method was to isolate fluency-heuristic-based judgments, to facilitate mapping specific brain areas subserving the heuristic. We started with two hypotheses: The first one assumed that the VMPFC is a neural correlate of fluency-heuristic-based decisions and was derived from previous imaging findings on phenomenologically similar kinds of decisions. The second hypothesis suggested that the claustrum is a neural correlate of fluency-heuristic-based decisions and was derived from an account of retrieval fluency based on successfully binding memory traces. We ran two interaction contrasts specifically designed to pull out trials that were very likely based on the fluency heuristic. The first interaction included the factor difference in recognition latencies (large vs. small differences) and the factor response time (fast vs. slow responses). The second interaction included the factor adherence to the fluency heuristic and the factor correctness. Both specific interaction contrasts elicited activation within the dorsal claustrum, as did a conjunction analysis of the two interaction contrasts. No activation within the VMPFC was observed. We take the latter result to indicate that fluency-heuristic-based decisions are dissimilar to metacognitive judgments about the felt-rightness of a memory. Thus, the present results revealing the dorsal claustrum as the most consistent neural correlate of decisions that are most likely to rely on the fluency heuristic support our second hypothesis. Hitherto, comparative anatomical, tractographic, electrophysiological, tracing, histological studies and the like have dominated the investigation of the claustrum, yet claustral activation has only seldom been reported in fMRI studies (Volz & von Cramon, 2006). Recently, Crick and Koch (2005) speculated that the claustrum gives rise to integrated conscious percepts. Morphologically, the claustrum is a thin, irregular band of gray matter, hidden beneath the inner surface of the neocortex and in close proximity to the insula. Macroscopically, the claustrum is divided into a dorsal (or posterosuperior) part and a ventral (anteroinferior) part (Fernández-Miranda et al., 2008; Kowianski et al., 1999). The dorsal claustrum lies between the putamen, from which it is separated by the external capsule, and the insular cortex, from which it is separated by the extreme capsule (Edelstein & Denaro, 2004; Fernández-Miranda et al., 2008). Considering the claustrum’s tractography may hint at its function. Tractography studies have revealed extensive neocortical connections and a topographical organization in the dorsal claustrum, where posterior cortical areas project to the posterior part of the dorsal claustrum and more anterior parts of cortical areas converge onto the anterior part of the claustrum, thereby forming (overlapping) cortical projection zones (Fernández-Miranda et al., 2008; Morys, Narkiewicz, & Wisniewski, 1993). Almost all of the claustrum-to-cortex projections are reciprocated, with only a few exceptions (e.g., V1; Sherk, 1986). In addition, the claustrum maintains two-way connections with subcortical structures involved in the limbic system, such as the amygdala and prepiriform cortex (Fernández-Miranda et al., 2008). These data were taken to suggest that the claustrum is a structure that interconnects the senses, providing them direct access to each other (Ettlinger & Wilson, 1990). In doing so, the claustrum plays a crucial role as a polymodal structure engaged in the transfer of information to and from various cortical regions (Kowianski et al., 1999). Crick and Koch’s (2005) metaphor for the claustrum is that of a conductor coordinating the players in an orchestra, where the musicians are the various cortical regions. The conductor is responsible for binding the performances of individual musicians into an integrated, synchronous whole. Similarly, the claustrum rapidly combines the different attributes of objects both within and across modalities so that an integrated whole is experienced, rather than a collection of isolated attributes. Given the findings on the function of the claustrum so far, claustral activation during fluency-heuristic-based decisions may reflect the experience of an integrated signal associated with the retrieval of one or more memory records. For example, Table 3 Anatomical specification, cluster size (mm3), and Talairach (x, y, z) coordinates of significantly activated voxels of the interaction contrast fluency (decisions agreeing vs. conflicting with the fluency heuristic) and incorrectness (incorrect vs. correct responses). Anatomical specification
mm3
R claustrum L rostral cingulate zone, posterior part R rostral cingulate zone L superior temporal gyrus L middle temporal gyrus, posterior part
216 1107 216 297 243
Note: Only clusters of at least five contiguous voxels are reported. L: left, R: right.
x
y 34 10 12 53 42
z 6 6 3 30 62
9 39 42 21 9
K.G. Volz et al. / Consciousness and Cognition 19 (2010) 829–837
837
episodic, semantic, and visual information may all be rapidly combined and bound in the claustrum. This signal may then suffice to guide inferences, and accordingly, be experienced as fluency. This differs from the way fluency was treated by Schooler and Hertwig (2005), where it was taken to simply represent the speed with which a single memory record is retrieved. The present finding corresponds to findings on perceptual fluency: In the study by Volz and von Cramon (2006), participants had to judge the coherence of pictorial stimuli, and claustral activation was suggested to represent the temporal synchronization process during perceptual fluency judgments. Together, we take the present results to indicate that inferences that are made according to the fluency heuristic specifically rely on an integrated signal at the fast time scale. Acknowledgments We thank Thomas Dratsch for helping with the behavioral data analyses and Anita Todd for editing a draft of this article. References Anderson, J. R., & Schooler, L. J. (1991). Reflections on the environment in memory. Psychological Science, 2, 396–408. Bar, M., Kassam, K. S., Ghuman, A. S., Boshyan, J., Schmid, A. M., Dale, A. M., et al (2006). Top-down facilitation of visual recognition. Proceedings of the National Academy of Sciences of the United States of America, 103, 449–454. Bechara, A., Tranel, D., & Damasio, A. R. (2002). The somatic marker hypothesis and decision-making. In F. Boller & J. Grafman (Eds.), Handbook of neuropsychology (2nd ed., pp. 117–143). Amsterdam, Netherlands: Elsevier. Birn, R. M., Cox, R. W., & Bandettini, P. A. (2002). Detection versus estimation in event-related fMRI: Choosing the optimal stimulus timing. NeuroImage, 15, 252–264. Crick, F. C., & Koch, C. (2005). What is the function of the claustrum? Philosophical Transactions of the Royal Society of London B: Biological Sciences, 360, 1271–1279. Damasio, A. R. (2004). William James and the modern neurobiology of emotion. In D. Evans & P. Cruse (Eds.), Emotion, evolution, and rationality (pp. 3–14). Oxford, England: Oxford University Press. Edelstein, L. R., & Denaro, F. J. (2004). The claustrum: A historical review of its anatomy, physiology, cytochemistry and functional significance. Cellular and Molecular Biology, 50, 675–702. Ettlinger, G., & Wilson, W. A. (1990). Cross-modal performance. Behavioral processes, phylogenetic considerations and neural mechanisms. Behavioural Brain Research, 40, 169–192. Fernández-Miranda, J. C., Rhoton, A. L., Kakizawa, Y., Choi, C., & Alvarez-Linera, J. (2008). The claustrum and its projection system in the human brain: A microsurgical and tractographic anatomical study. Journal of Neurosurgery, 108, 764–774. Fraisse, P. (1984). Perception and estimation of time. Annual Review of Psychology, 35, 1–36. Friston, K. J., Fletcher, P., Josephs, O., Holmes, A., Rugg, M. D., & Turner, R. (1998). Event-related fMRI: Characterizing differential responses. NeuroImage, 7, 30–40. Friston, K. J., Frith, C. D., Turner, R., & Frackowiak, R. S. (1995). Characterizing evoked hemodynamics with fMRI. NeuroImage, 2, 157–165. Goldstein, D. G., & Gigerenzer, G. (2002). Models of ecological rationality: The recognition heuristic. Psychological Review, 109, 75–90. Greene, J., & Haidt, J. (2002). How (and where) does moral judgment work? Trends in Cognitive Sciences, 6, 517–523. Hertwig, R., Herzog, S. M., Schooler, L. J., & Reimer, T. (2008). Fluency heuristic: A model of how the mind exploits a by-product of information retrieval. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 1191–1206. Jacoby, L. L., & Dallas, M. (1981). On the relationship between autobiographical memory and perceptual learning. Journal of Experimental Psychology: General, 110, 306–340. Kelley, C. M., & Jacoby, L. L. (1998). Subjective reports and process dissociation: Fluency, knowing and feeling. Acta Psychologica, 98, 127–140. Kowianski, P., Dziewiatkowski, J., Kowianska, J., & Morys, J. (1999). Comparative anatomy of the claustrum in selected species: A morphometric analysis. Brain, Behavior and Evolution, 53, 44–54. Kringelbach, M. L., & Rolls, E. T. (2004). The functional neuroanatomy of the human orbitofrontal cortex: Evidence from neuroimaging and neuropsychology. Progress in Neurobiology, 72, 341–372. Lohmann, G., Muller, K., Bosch, V., Mentzel, H., Hessler, S., Chen, L., et al (2001). LIPSIA—A new software system for the evaluation of functional magnetic resonance imaging of the human brain. Computerized Medical Imaging and Graphics, 25, 449–457. Marewski, J. N., & Schooler, L. J. (2010). Cognitive niches: An ecological model of emergent strategy selection, submitted for publication. Miezin, F. M., Maccotta, L., Ollinger, J. M., Petersen, S. E., & Buckner, R. L. (2000). Characterizing the hemodynamic response: Effects of presentation rate, sampling, procedure, and the possibility of ordering brain activity based on relative timing. NeuroImage, 11, 735–759. Morys, J., Narkiewicz, O., & Wisniewski, H. M. (1993). Neuronal loss in the human claustrum following ulegyria. Brain Research, 616, 176–180. Neumann, J., & Lohmann, G. (2003). Bayesian second-level analysis of functional magnetic resonance images. NeuroImage, 20, 1346–1355. Nichols, T., Brett, M., Andersson, J., Wager, T., & Poline, J. B. (2005). Valid conjunction inference with the minimum statistic. NeuroImage, 25, 653–660. Norris, D. G. (2000). Reduced power multislice MDEFT imaging. Journal of Magnetic Resonance Imaging: JMRI, 11, 445–451. Oppenheimer, D. M. (2008). The secret life of fluency. Trends in Cognitive Sciences, 12, 237–241. Pohl, R. F. (2006). Empirical tests of the recognition heuristic. Journal of Behavioral Decision Making, 19, 251–271. Schnyer, D. M., Nicholls, L., & Verfaellie, M. (2005). The role of VMPFC in metamemorial judgments of content retrievability. Journal of Cognitive Neuroscience, 17, 832–846. Schnyer, D. M., Verfaellie, M., Alexander, M. P., Lafleche, G., Nicholls, L., & Kaszniak, A. W. (2004). A role for right medial prefrontal cortex in accurate feeling of knowing judgments: Evidence from patients with lesions to frontal cortex. Neuropsychologia, 42, 957–966. Schooler, L. J., & Anderson, J. R. (1997). The role of process in the rational analysis of memory. Cognitive Psychology, 32, 219–250. Schooler, L. J., & Hertwig, R. (2005). How forgetting aids heuristic inference. Psychological Review, 112, 610–628. Sherk, H. (1986). The claustrum and the cerebral cortex. In E. G. Jones & A. Peters (Eds.). Cerebral cortex (Vol. 5, pp. 467–499). New York, NY: Plenum Press. Talairach, P., & Tournoux, J. (1988). Co-planar stereotactic atlas of the human brain. Stuttgart, Germany: Thieme. Thirion, J. P. (1998). Image matching as a diffusion process. An analogy with Maxwell’s demons. Medical Image Analysis, 2, 243–260. Ugurbil, K., Garwood, M., Ellermann, J., Hendrich, K., Hinke, R., Hu, X. P., et al (1993). Imaging at high magnetic fields: Initial experiences at 4 T. Magnetic Resonance Quarterly, 9, 259–277. Volz, K. G., Rübsamen, R., & von Cramon, D. Y. (2008). Cortical regions activated by the subjective sense of perceptual coherence of environmental sounds: A proposal for a neuroscience of intuition. Cognitive, Affective, & Behavioral Neuroscience, 8, 318–328. Volz, K. G., Schooler, L. J., Schubotz, R. I., Raab, M., Gigerenzer, G., von Cramon, D. Y., et al (2006). Why you think Milan is larger than Modena: Neural correlates of the recognition heuristic. Journal of Cognitive Neuroscience, 18, 1924–1936. Volz, K. G., & von Cramon, D. Y. (2006). What neuroscience can tell about intuitive processes in the context of perceptual discovery? Journal of Cognitive Neuroscience, 18, 1–11. Worsley, K. J., & Friston, K. J. (1995). Analysis of fMRI time-series revisited—Again. NeuroImage, 2, 173–181.
Consciousness and Cognition 19 (2010) 838–846
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Memory and content q Gottfried Vosgerau * Institut für Philosophie, 23.21.00.46B, Heinrich-Heine-Universität, Universitätsstr. 1, 40225 Düsseldorf, Germany
a r t i c l e
i n f o
Article history: Available online 23 August 2010 Keywords: Memory Mental representation Content Behavior Functionalism
a b s t r a c t The paper argues that any theory of content has to adopt a ‘‘functionalistic core” to concord with the cognitive sciences. This functionalistic core requires that representations are defined as substitutes in functions that describe the flexible behavior to be explained by the representation. The content of a representation can thus only be determined if the representation is ‘‘in use”, i.e. if it is an argument in such a function. The stored entities in memory are not in use while they are stored, and hence cannot be assigned a specific content. The term ‘‘template” is introduced to describe stored entities in memory. The discussion of some implications of this result show that some deep philosophical problems follow from this argument as well as consequences for empirical research on memory. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Memory is commonly assumed to be preservative storage of content.1 Just like a story can be ‘‘stored” by writing it down, the contents of our mental states can be stored in memory. Naïvely speaking, we take a content as it is and put it somewhere safe to be retrieved whenever needed. Starting with this picture, we can proceed with differentiating between different kinds of memories that fulfill different functions—some examples are: short-term vs. long-term memory, episodic vs. semantic memory, implicit vs. explicit memory, procedural vs. declarative memory, working memory, visual buffers. Although this naïve picture of memory is somewhat neat, there are some deep problems with it. The general strategy of this paper is to argue that any theory of (representational) content has to adopt what I call a ‘‘functionalistic core” if it is to capture the explanatory role of representations in behavioral sciences. This means that if the content of a representation is taken to explain (flexible) behavior, then we have to take the target of the behavior as the starting point for determining the content. The very abstract characterization of this idea is formulated in terms of substitution in functions. This abstract formulation cannot be taken as a theory of representational content itself, but it is sufficiently general to be a frame for theories of representational contents. On the basis of this general frame, it is then argued that we cannot assign certain contents to stored entities (whatever they are). Thus, it is concluded, there is no content in memory (and so, strictly speaking, no representations). Based on this negative outcome of the discussion, I will introduce the term ‘‘template” to sketch—again in a general way— a strategy of overcoming the obvious contradiction with the classical intuition regarding memory. Thereby, the introduced notion cannot provide a theory but only some constraints for a theory of memory. However, already these general considerations have some important implications which will be discussed in the last section.
q
This article is part of a special issue of this journal on Self, Other and Memory. * Fax: +49 (0)211 81 14338. E-mail address: [email protected] URL: http://www.phil-fak.uni-duesseldorf.de/philo/personal/thphil/vosgerau/. 1 See, for example, Burge (1993) and, for a critical discussion and rebuttal of this picture: Matthen (2010).
1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.06.021
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
839
d
d nest Fig. 1. Homing behavior of ants.
The problem I want to discuss in this paper is, however, confined to long-term memory and does not apply to short-term memory or working memory, let alone ultra-short-term buffers (like the visual buffer). This has to be qualified even further: The discussion is confined to ‘‘entries” in long-term memory that are not actually remembered, i.e. to non-occurrent memory items.2 Indeed, the points made in the paper address all forms of non-occurrent ‘‘mental representations” (I assume that all of these, which include the beliefs in our ‘‘belief-system”, can be said to be ‘‘in memory”3). Thereby, ‘‘occurrent” is not understood in terms of consciously available (background-beliefs that play a role in the tokening of a certain chain of reasoning, e.g. do not have to be consciously available but are occurrent in this sense) but rather in the sense of being currently involved in some mental process (consciously or unconsciously). To be more precise: a thought is occurrent if it has to be mentioned in an explanation of the behavior. My ‘‘stored belief” that Paris is the capital of France, e.g. is non-occurrent most of the time, since most of the time my behavior can be explained without reference to this belief. Of course, sometimes the explanation of my behavior has to refer to this belief, and in these cases we have to assume that this belief is indeed occurrent in the sense that it is involved in my reasoning. However, this does not mean that it necessarily is a conscious thought in such cases—it is very well possible that it stays an unconscious background belief. To sum up, this paper is concerned with non-occurrent ‘‘stored” representations that are not involved in any way in actual mental processes.4 Furthermore, let me add some notes on what is not in the scope of the paper. Some may think that having content (being intentional, if you want) is the hallmark of being mental. Thus, if something does not have content, it cannot be mental either. So, for some it might seem that I argue that memory is not part of the mind. However, this issue is, in large parts, independent of the arguments presented here and thus cannot be discussed in this paper. Therefore, I will be completely silent about the definition of the mental and its boundaries. (Indeed, I think that the mental cannot be defined with the notion of intentionality, and I would even be tempted to regard the arguments presented here as a reductio of such definitions.) Moreover, memory has often been discussed in epistemological context where the main question is whether and how memory can justify an occurrent belief. My considerations are not carried out in such a context—this paper is interested in whether the notion of stored content is compatible with the notion of representation as it is used in the behavioral sciences. However, the arguments presented here have some implications for the epistemological discussion, but they are not based and thus independent of this discussion. 2. Theories of content The problem of determining content5 is as old as philosophy itself. Accordingly, many more or less detailed theories of content have been formulated, and I will not discuss any account here. In contrast, let me start with some reflections on why we talk about representations and contents at all which will lead to a very general constraint for every theory of representation, namely that it has to acknowledge that the content of representations (and thereby the status of something as a representation) can only be determined if the representation is ‘‘in use”. 2.1. The role of representations Consider the so-called homing behavior of the ant (Gallistel, 1993): After an unsystematic search for food, the ant is able to go back to its nest on a straight line (see Fig. 1 left). If the ant is displaced before returning, it will take a parallel path to a point where the nest would have been, if it had been displaced in the same direction and distance d (see Fig. 1 right).
2 I do not want to suggest that the term ‘‘memory” should not be used for occurrent memories, i.e. occurrent mental representations based on memory. However, for the sake of readability, I will refer to non-occurrent memories with the term ‘‘memory” if not otherwise explicitly mentioned. 3 Thus, the discussion concerns all kinds of memory, not just episodic memory. 4 Thus, if you have a holistic picture of the mind in the sense that you belief that every potential representation of a subject has to figure in every explanation of her or his behavior, then my distinction between occurrent and non-occurrent representations does not make any sense to you, and so the whole paper is worthless to you. However, I will argue for the role of representations in explanations of behavior below in a way that makes the holisitic picture very unplausible. 5 In this paper, I will use ‘‘content” to refer to representational content exclusively.
840
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
In order to explain this kind of behavior, we cannot use simple stimulus–response-patterns, because the ant obviously does not rely on nest-stimuli in finding its way home (otherwise, it would not take the parallel path in the displacement condition). Hence, the behavior can only be explained if some internal states of the ant are assumed. These internal states, whatever they are in detail, are what is called a (mental) representation. In this case, the ant is said to have a representation of the location of the nest. (In fact, detailed research on the homing behavior revealed that the representation of the ant consists of a registration of the path’s angle to the sun and the number of steps the ant has to make.) The reason why the concept of representation is introduced in cognitive science is, as demonstrated with the case of the homing behavior, to explain flexible behavior, i.e. behavior that cannot be explained by rigid stimulus–response-patterns. The explanation of flexible behavior requires the assumption that some internal state of the behaving system exists which is then said to be a representation. Contrast this with the behavior of the moth, which flies towards light sources. If the moth is displaced on its way, it will not take a parallel route but continue to fly directly to the light source. If the light source is removed, the moth will simply stop this kind of behavior and not search for it or continue to fly to the point where the light source had been. Thus, this behavior of the moth is not flexible since it can be explained by a simple rigid response (flying) to a stimulus (light). Therefore, in this kind of rigid behavior, no representations are involved. (The difference between rigid behavior and flexible behavior can also be described in terms of misrepresentation: Flexible behavior is the kind of behavior which allows for unsuccessful cases, e.g. the displacement condition above. Rigid behavior, on the other hand, does not allow for such cases—there is no way in which the moth can err in flying towards a light source. The possibility of explaining such unsuccessful cases depends on the possibility of misrepresentation. If there is no representation, there cannot be misrepresentation either. And—as argued for by Dretske (1986), there is no representation without the possibility of misrepresentation. In other words: The possibility of misrepresentation is a conceptual necessity for every representation.) Why is the internal state of the ant a representation of the location of the nest, and not, say, of trees? Or: How is the content of a representation determined? The internal state of the ant, the representation, is assumed to explain the homing behavior of the ant. Since the homing behavior is directed towards the nest (the target object of the behavior is the nest), we assume that the internal state represents this very target object (namely the nest). It is, to borrow a term from Cummins (1996), a proxy or a stand-in for the nest.6 Note that the unsuccessful behavior displayed in the displacement condition is still of the same type, namely of the type ‘‘homing behavior”. If it would not be, empirical research on the homing behavior could not draw inferences from the displacement conditions. In particular, the content of misrepresentation is the same as in the case of successful representations: Misrepresentations are characterized as cases where content and the represented reality fall apart.7 Thus, even in cases where the ant does not reach the nest, it still has a nest-representation, which is a misrepresentation, because the nest is not there (and thus content and represented world fall apart). Hence, the target object of the homing behavior is still the nest, even if the behavior is not successful and the nest is not reached.8 Let me illustrate this very important point with another example. In now classical experiments, Cheng (1986) trained rats in a rectangular arena to find food in one specific corner. In the experimental phase, the rats showed in most of the trails the following behavior: they went either to the correct corner or to the corner diagonally opposite of the correct corner. Cheng (1986) concluded that rats are able to navigate according to the geometrical layout of the arena. The rationale of this (typical behavioral) experiment is the following: after training, the setup is so modified that systematic cases of unsuccessful behavior are shown. These cases are then interpreted as cases of misrepresentation. Because of the systematic nature of these misrepresentations, such cases allow insight into the structure of the representations. Two things are important here: first, the behavior has to be interpreted as being of the same kind, namely being a search-for-food-behavior. If the cases of misrepresentation are not interpreted in this way, they could not give us any insight into the structure of the rats’ representations. Therefore, even in these unsuccessful cases, the target of the behavior still is the food (and this is the reason why we can assume that it is also the target or the content of the (mis-)representation). Thus, the individuation of behavior is already the crucial move for determining the content of the representation which ultimately explains the (kind of) behavior. Second, although this experiment is based on learning and so on memory, we cannot conclude from these data that this or that content is stored in memory. The representation that explains the behavior is the occurrent representation that the rat forms in the experimental trails. This representation is, surely, based on memory. However, this experiment does not tell us anything about what is stored and how the occurrent representation is based on memory. The explanation of flexible behavior thus runs along the following lines. A cognitive system shows a certain kind of behavior (e.g. homing behavior, food-search-behavior). The classification of a behavior token is made according to the criteria for behavior individuation, which are certainly not easy to determine, but which are not to be discussed here (see also footnote 8). Importantly, the individuation of behavior is independent from the question whether the target of the behavior is present or not: the ant, e.g. shows homing behavior even in cases where the nest has been removed (or the ant is displaced), and the 6 The use of the term ‘‘target” is also inspired by Cummins (1996), although there is the difference that Cummins speaks of the target of a representation and I speak of the target of behavior. These two are not the same, since an unsuccessful behavior can still have a well defined target. 7 Otherwise, the famous disjunction problem (Fodor 1987) could not even be formulated—it arises if some X causes a (mis-)representation with content Y. 8 Of course, the individuation criteria for behavior types are not at all clear or easy to describe. However, ‘‘behavior” is the more basic term since this is ultimately the explanandum of cognitive sciences. In this sense, the individuation of representations is dependent on the individuation of behavior. This is also the reason why Cummins (1996) general critique of use-based theories of representation does not apply to the present account: What Cummins rightly criticizes is theories that rely on a notion of successful use; however, ‘‘use” here is meant to refer to the use in the guidance of behavior—successful or not. For place limitations, I cannot discuss the problem of behavior individuation here.
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
841
rat shows food-search-behavior even if no food was hidden. Thus, the target of behavior is determined by identifying a behavior as being of a certain kind. This behavior can now be explained by claiming that the cognitive system has a representation of the target of behavior: the ant shows homing behavior because it has a representation with the content that can be linguistically expressed by ‘‘the location of its nest”; the rat shows search-food-behavior because it has an (occurrent) representation with the content that can be linguistically expressed by ‘‘the location of food”. The content of a representation thus has a specific explanatory role: it explains why cognitive systems are able to show behavior that is directed at targets about which the cognitive system does not have sensory information. If the content of the representation would not be ultimately determined by the target of the behavior, it could not fulfill this explanatory role. This kind of explanation might sound vacuous (cf. Cummins, 1996), because the content of the representation is dependent of the behavior which is to be explained. I do not think that this is a serious problem for two reasons9: First, there are other explanations that involve similar twists and which, nevertheless, seem unproblematic. The gravitational mass of an object, for example, is determined by the force between the object and other masses. Nevertheless, the following explanation is unproblematic: the object attracts other masses because it has this and that mass. Second, the explanations I gave above are not the end of the story, of course. Behavioral experiments try to further analyze the structure of the representation, and thereby provide more and more detailed explanations. For example, experiments revealed that the ant is able to represent the nest in terms of the path’s angle to the sun and the number of steps it has to make to reach the nest. However, such refinements in explanation do not touch the core, namely that this representation really represents the nest (and not just angles and steps), and that the ant reaches the nest because it has this nest-representation. 10 To sum up, the content of a representation is ultimately determined by the target object of the behavior to be explained. Of course, this kind of determination is quite coarse-grained. Therefore, further criteria are often introduced (e.g. the criterion of simultaneous believability; Frege, 1892); nevertheless, none of these criteria can explain why the ant has a nest-representation rather than a tree-representation. In other words: Every determination of content has to start with the target object of the behavior to be explained, whatever further criteria are then added. In this sense, the ‘‘core” of every content individuation is based on the target object of the behavior in question (see also Vosgerau, Schlicht, & Newen, 2008). This ‘‘core” of content determination can be formulated in functionalistic terms (Vosgerau, 2008, 2009): If the behavior is described in terms of functions (in the mathematical sense), representations can be defined as substitutes in these functions. For example, the behavior of the ant might be described by a function mapping the actual location of the ant and the location 0 of the nest onto a new location of the ant which is identical to the location of the nest f: hlocant, locnesti # locant , whereby (usu0 ally) locant ¼ locnest : The representation of the location of the nest, which guides the ant, can now be defined as a substitute for h i 0 nest : hlocant ; repnest i # locant : Note that this formulation is functionalistic in the widest sense, as it is not confined locnest : f rep locnest to inputs, states, and outputs of brains (as in ‘‘classical” functionalism; Putnam, 1975; Block, 1978), nor to teleosemantics (Dretske, 1986; Millikan, 1984). It is formulated in such an abstract way that it is possible to cover different cases, e.g. evolutionary functions like the mapping of the presence of danger onto fleeing behavior, where the presence of danger can be substituted by the beaver’s splash (see Millikan, 1984, 1994); or behavioristic functions like the mapping of red light onto the press of a button in an experiment, where the red light can be substituted by the firing of certain cells in the retina. Moreover, it is able to include cases of misrepresentation (as, e.g. the case in which the ant is displaced), since in such cases the same kind of behavior is displayed which is described by the same function.11 Of course, this abstract functionalistic core is not apt to determine content (see Vosgerau, 2008, 2009 for details), but it is a core without which representation cannot be defined—at least not in a way that it could fulfill its role in the explanation of behavior.12 However, the first and ultimate reason to introduce the talk about representations is to explain flexible behavior. Thus, any definition of representation that is not able to cover the explanatory role of representations for behavior cannot be adequate with respect to the concept of representation as it is used in the behavioral sciences. Therefore, I take this functionalistic characterization of representations to be a necessary core feature of every definition of representation and content. 2.2. The ‘‘use” of representations The basic idea of mental representation as captured in the ‘‘functionalistic core” implies that representations and their contents can only be determined when they are ‘‘in use”, i.e. when they are actually substituting the represented entity
9
Due to space limitations, I cannot discuss the nature of explanation in detail here. There might be cases, where further investigations do change the content we ascribe. However, in these cases the classification of the behavior is changed as well. Take, e.g. the snapping-behavior of the frog: Lettvin, Maturana, McCulloch, & Pitts (1959) showed that frogs do not react to flies but to black moving dots. Thus, their behavior cannot be described as snapping at flies, but has to be re-described as snapping at black moving dots. (You might then add a story of why this behavior makes evolutionary sense . . .) 11 As argued for above, it is crucial to the empirical investigation of representation that cases of unsuccessful behavior/misrepresentation are still classified as the same kind of behavior, since otherwise the systematic variation of conditions that produce such cases would not provide information about successful cases. 12 In this sense, it is not a theory of representation but rather a frame for theories of representation. Thus, it does not compete with other theories—rather, it presents constraints for theories such that theories that do not include, in some way or another, this functionalistic core are ruled out as good candidates for successful theories of representation. 10
842
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
in a function to guide behavior. As long as they (the vehicles) are not used, they cannot be identified as representations with this and that content; this is to say that they are not representations per se. Rather, the property of being a representation with a certain content is a relational property. There are many things that can be used to represent something, but which cannot be said to represent when they are not used. One famous example is the trace of an ant in the sand which incidentally resembles a picture of Churchill (Putnam, 1981). Likewise, I can use (a representation of) the postal address system to explain somebody the IP system used to identify computers in networks. In this case, I will use the postal address system to represent the IP system (which it does, of course, not by itself). The reason is that something that is not actually playing the role of a substitute in a certain function does not fall under the definition of representation. The fact that a given thing (a certain physical object, a certain structure, or a certain mental entity) can be used to represent very different things is the reason why dispositional definitions of content will not succeed. The crucial problem here is basically the disjunction problem (discussed foremost in relation to causal theories of mental representation; e.g. in Fodor, 1987): If we define the content of a representation to be what it possibly could stand for (in the dispositional sense), then we end up with a huge (maybe even infinite) disjunction of all different possible targets for which the representation could stand (e.g. the content expressed by ‘‘the postal address system or the IP system or . . .”). Therefore, a dispositional definition of content can only be successful if there is only one way the representation is disposed to be used, which would imply that the content was independent of usage. This, however, is—as argued for—not the case. Thus, dispositional definitions of contents cannot succeed. To sum up, the behavior of a system can be described with functions (in the mathematical sense), which are describing processes.13 In these functions, representations can substitute the entities they represent. This very abstract and general description of representations expresses the fact that representations have to be in use and so reflects the basic constraint on any theory of content that hopes to capture the explanatory role of representations played in behavioral sciences. In particular, it is not sufficient to be usable (not even to be disposed to be used) as a representation in order for something to be a representation. Therefore, we can assign a certain content to a representation only if the representation is in use, i.e. involved in a process. 2.3. No theory of stored content The lesson from the discussion so far is that we do not have a theory of content and representation that would allow for ‘‘stored contents”: As soon as a representation is stored, it is not involved in processes anymore (until it is retrieved from memory); as long as it is not involved in any process, it cannot be assigned a content and thus ceases to be a representation. Therefore, as soon as a representation is stored, it is not a representation anymore. In particular, we cannot assign contents to whatever it is that gets stored. Although it is intuitively appealing to speak of stored content, there is no theory of content that could justify this talk. Quite on the contrary, the investigation of memory has to be conducted in different terms—the representational picture of the mind is, I assume, very well suited to describe occurrent mental processes and states, but it fails to give us an account of non-occurrent states (memory). Thus, the representational picture of mind is severely limited in its explanatory value. Note that a dispositional reading does not help here neither (as argued for above, it does not help to determine content in general): A stored ‘‘representation” can be ‘‘reactivated” when retrieved from memory and can be used in a process where it represents the same as it did before storage. In this sense, we could say that it is disposed to be used in the same way during storage. However, it can be used in very different ways as well. If the disposition to be used in this and that way would determine their content, then we would be faced with the classical ‘‘disjunction problem” (cf. Fodor, 1987), i.e. we would have to say that their content is the disjunction of all the different contents which we would assign in all different possible usages. To illustrate this point, a little story that I heard somewhere might help (the story is so nice that it is probably not true): The NASA used to record data on huge magnetic bands which they stored in some place. In order to read these bands, big machines were necessary. At some time when the bands were not used anymore, they decided to get rid of the reading machines. Only later they discovered that with the reading machines the stored information was literally gone for them— although the bands were still there, nobody could read them and retrieve the stored information. The question here is: Is there really any sense in saying that a certain specific content (as opposed to information, which is, of course, on the band) is stored in the bands when all reading machines are gone? I do not think so. The bands could be, for example, used by a composer with a regular music tape player to create a specific sound which figures as a part of a symphony. The band, which the composer stores in his studio, has now the (at least additional if not sole new) musical content. Therefore, the content of some storing device cannot be determined without the retrieval mechanism, which means that the content can ultimately only be determined when the representation is in use, i.e., when it is actually retrieved. Applying these results to human memory leads to the conclusion that there is no content stored in memory. The content of memory can only be determined when the use after retrieval is known. Thus, to say that specific contents are stored in memory presupposes that the use of the representation is always known. This can only be the case if the use after retrieval is always the same. However, this presupposition is far away from being warranted. Above, I already gave some example for 13 Here, nothing is assumed about the processes; in particular, it is not assumed that they are computational processes. By saying that force is a function of mass and acceleration, we give a functional description and we assume that some process (namely the acting of the force on an object) takes place; however, nothing is implied about the nature of this process.
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
843
how stored ‘‘representations” can be used in new ways (the postal address system for representing the IP system): This kind of new ways of usage could be subsumed under the term ‘‘analogies”. Because this kind of usage is not only possible but often actual, I will speak of templates that are stored in memory and that can be used to construct mental representations with content. The templates are, however, not representations themselves because they lack content (just like the business letter template on the computer is not a business letter). Rather, they are used for the construction of representations. Retrieval from memory is then the construction of a representation with the help of a stored template. 3. Templates in memory In long-term memory, templates are stored. They cannot be assigned a specific content, because they are not in use and because they can be used to construct different representations. Of course, templates cannot be used to construct any kind of representation: For each template, there are constraints on what representations can be constructed on its basis. These constraints are best understood in terms of certain relations between the templates and the possible represented things. For example, the ant’s trace in the sand (see above) can be used to represent Churchill because of a certain similarity between the two. Likewise, the structure of the postal address system stands in a certain similarity relation to the IP system, which is the reason that it can be used to represent it. I have argued that this relation is (at least for conceptual representations) best understood as a partial isomorphism (Vosgerau, 2006, 2008, 2009). Partial isomorphism is thereby defined as an isomorphism that holds between the structure of the representation and a part of the structure of the represented object. Which part of the represented object is relevant for this relation is determined by the function in which the representation plays its functional role. The crucial point here is that partial isomorphism does not suffice to determine content or to make something a representation since there is always a multitude of isomorphic objects (and structures). As argued for above, we need the functionalistic core—or: the use of the representation—for this.14 Templates are thus not just representing abstract entities (like structures) or fuzzy things. They do not represent at all, although their structure might be well-defined. However, the structure alone does not make them representations, although the structure highly constrains possible usages as representations. Therefore, they stand in contrast to the notion of schemata, which is widely used in memory research (cf. Alba & Hasher, 1983). Schemata, as scripts (Schank & Abelson, 1977) or frames (Barsalou, 1992; Minsky, 1975), are conceived of as representing abstract entities, e.g. not a specific birthday party but the general procedure of a birthday party or not a specific car but the general (possible) properties of a car. In this sense, schemata are taken to have a defined content (which can be true or false), although they are not supposed to represent single events or objects. Templates, in contrast, do not represent anything—not even abstract entities or generalized objects. Another term that is often used for memory is the term ‘‘trace” (e.g. Bernecker, 2008, 2010; Matthen, 2010). A memory trace is said to exist between a mental state (e.g. a perception) and the mental state we are in when we remember the first one. It connects the two mental states in a causal manner, and so it is able to build the basis for justification. However, a trace itself is not a mental state but a sequence of (mental?) states in which one state is caused by its predecessor (cf. Bernecker, 2008, Fig. 3.1 on p. 43). Therefore, the notion of a memory trace can be fruitfully applied to problems discussed in epistemology (where the trace can build the basis for justification for the remembered state), but it is accompanied with assumptions regarding causal connections that are not relevant for the present argument. The notion of a template is, in contrast, introduced as a more general notion which does not rely on specific assumptions about causation or sequences. In fact, traces can be viewed as one possible form of templates (if they are interpreted as bearing no content, as done, e.g. by Matthen, 2010). However, there are many other possible forms or ‘‘realizations” of templates, e.g. weights in neural networks, symbolic strings, mental models, . . . Since templates can be used to construct very different representations, memory recall has to be understood as a constructive processes that creates content rather than a conservative processes that re-activates content. The fact that at least large parts of memory recall are indeed constructive is well documented in the empirical literature (see, e.g. Schacter, Norman, & Koutstaal, 1998; Schank & Abelson, 1977). The claim of this paper goes beyond the empirical claim in that it claims that all kinds of memory retrieval are constructive. It is to be understood in the sense that in principle, every stored template can be the basis of different representation constructions. Thus, the point is clearly a philosophical point: Even if a specific template could be empirically shown to yield always the same kind of representation (the same content), the thesis could still be true, since a de facto existing rigid connection between a specific template and a specific kind of content does not contradict the in principle possibility of connections to different contents. The crucial point here is that the stored template alone can never explain behavior; only when retrieval mechanisms are also taken into account, such that the resulting occurrent representation is in focus, explanation of behavior can be successful (see above). Nevertheless, the argument has implications not only for philosophy but also for theorizing about memory and thus for empirical research on memory. In the next section, some of the most important implications will be sketched.
14 Bartels and May (2009) argue that ‘‘structural resemblance” (which is spelled out in terms of homomorphism; Bartels, 2005) is an integral part of functionalism. In Vosgerau (in press, chap. 8) I argue that the functionalistic story and the structural resemblance story are two different stories that in principle come apart.
844
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
4. Implications Ramsey (2007) presents a detailed discussion of different understandings of the term ‘‘representation”. His powerful (and in my view successful) arguments against the coherence of the notion of ‘‘implicit” or ‘‘tacit” representations shows parallels to the arguments presented here. However, the notion of tacit representation is ‘‘characterize[d] as based on the idea that there is a close link between the dispositional properties of a cognitive system, and a type of knowledge or information that the system represents ‘inexplicitly’ or ‘tacitly.’” (Ramsey, 2007, p. 151). Thus, tacit or implicit contents are conceived of as being present in the system (as dispositions) without there being a specific vehicle to which the specific content could be assigned. In this sense, they are stored (unless the dispositions manifest themselves)15 and therefore build a special subclass of what I called non-occurrent representations. Therefore, one implication of this paper is the claim of Ramsey: There is no such thing as implicit or tacit representations, since no content can be assigned to such non-occurrent ‘‘representations”. Another central implication is that stored templates are neither true nor false. Since stored templates are not representations and have no contents, they cannot be assigned truth-values either. A memory can thus not be correct by itself, but only when it is remembered. This means that only ‘‘occurrent” mental states have truth-values, and that true memories have to be constructed as occurrent representations of past events. There is no doubt that memory and stored templates play a crucial role for the explanation of cognition (not only human cognition) because memories shape the way we represent the world. Depending of what we have learned, we will represent the world in different ways. This fact can be easily explained with the idea of stored templates: Since templates constrain the structure of the constructed representations, occurrent representations will differ with the templates that are used for their construction. If, e.g. I have never learned how motorbikes work, my representation of a specific motorbike will not contain very specific information about the different parts of it. If, however, I am an expert in motorbikes, my representation of the same motorbike will have a much richer structure that allows me to identify different functional units and so on. The reason for this difference lies in the differently organized templates (in memory) that are available to me to construct the specific representation. However, the consequence of this discussion is that this very important part of cognition—memory—cannot be described in ‘‘classical” truth-apt terms. Such notions as ‘‘proposition” or ‘‘misrepresentation” cannot be applied by a theory of memory. Therefore, there can be no causal connection between the content of a stored representation and the content of an occurrent (remembered) memory because the first just does not exist. Bernecker (2008, 2010) argues for memory traces that explain the causal relation between past mental states (that are stored) and present states of remembering the past event that was represented by the past mental state. He distinguishes between a physical and a ‘‘mental” description, the first referring to some neural states and processes and the second referring to ‘‘content causation”. While there is, from the point of view of this paper, nothing wrong with the idea that occurrent memories are physically caused (at least in part) by some physical states or processes that are causally linked to former states (they are linked so by templates in my terminology), the idea that the content of the occurrent state is causally connected to the content of the past state, however, cannot be successful. Indeed, it is not at all clear why we need such a form of ‘‘content causation” at all (i.e. I see no explanatory advantage of assuming such a causal connection): Contentful states can be caused by non-contentful states (e.g. the tree in front of me and my sense apparatus are not contentful but cause my percept which is contentful), and there is no reason to assume that memory retrieval should work another way. Note that I do not argue against ‘‘mental causation”: Of course, contents can have causal powers since they explain behavior (this was the starting point in the paper). And perhaps it is even useful to think that contents can cause other contents by triggering some mental processes that produce other contentful states, e.g. in inferences. In this way, the past occurrent content could be said to indirectly cause the present (remembered) content through a processes involving non-contentful states (e.g. traces). Nevertheless, it does not follow that templates or traces have content. More generally, the so-called ‘‘belief system” of a cognitive agent, which is traditionally assumed to be a fairly stable system of contents which are believed by the agent, does not exist according to the position defended here. Rather, the stored items are templates without contents. The often adopted view (e.g. Bernecker, 2010) that stored items have a dispositional content does not help here as argued for above.16 Taken together, in the epistemological literature on memory, the term ‘‘content” is used to apply to non-occurrent states as well as occurrent states. However, such a notion of content is not compatible with the explanatory role this notion plays in behavioral sciences. Thus, the epistemic status of memory cannot be captured within a representational framework, especially not in a framework that postulates truth-apt stored representations.17 This point is also related to the discussion about ‘‘knowing how” and ‘‘knowing that” going back to Ryle (1945): Knowinghow is often conceived of as a kind of implicit knowledge not being analyzable in terms of truth-apt (propositional) representations (e.g. Jung & Newen, 2010). However, knowing-how is certainly part of our memory since we have learned how to
15 It could be discussed whether the manifestation of such dispositions always involves some kind of representation with content or not—I would suspect that in many cases the contents of the occurrent representations that build the basis of the manifestation have to be assigned contents that are not reducible to the disposition. However, due to space limitations this has to remain a suspicion at the moment. 16 In fact, Bernecker (2010) makes a distinction between conceptual and non-conceptual contents, the first being explained in terms of dispositional beliefs and the second being explained in terms of ‘‘subdoxastic” or implicit states. The notion of implicit representation is, however, also not compatible with the view presented here as shown above. 17 This does, of course, not entail that everything is wrong about such epistemological accounts—however, since this paper is not concerned with the epistemology of memory, I will not discuss these accounts further.
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
845
do things. The present discussion yielded that a representational analysis of stored ‘‘knowledge” is not possible at all. Thus, even the so-called ‘‘knowing that” kind of knowledge cannot be described in terms of content. In this sense, it might turn out that the distinction between the two kinds of knowledge vanishes, although in quite another sense than often assumed: Knowing-how is not reducible to knowing-that (e.g. Stanley & Williamson, 2001) but knowing-that is nothing but a kind of knowing-how in the sense that it is a certain ability to produce something (a representation in the one case and a certain kind of behavior in the other case). This ability, however, cannot be described or explained in terms of content. Therefore, according to the argument of this paper large parts of the classical picture of the ‘‘representational mind” have to be changed: contents are not the kind of things that can be stored, and—although memory is central to cognition—memory cannot be described in terms of representations. This does not necessarily mean that theories of memory have to be re-written from the very start. Instead, the aim of this paper is to point to one important factor in explanations involving memory that is often completely left out: In fact, the arguments given in this paper try to show that arguments of the following form are no good explanations: the cognitive system shows this and that behavior because it has stored this and that content. They are no good explanations because they at least leave out a very important part, namely the reference to the constructive retrieval mechanisms. In this sense, the (full) explanation should read: the cognitive system shows this and that behavior because it constructed this and that content out of this and that template with the help of this and that retrieval mechanism. Probably, some explanations of the above form that are actually given in behavioral sciences can be reinterpreted as ‘‘abbreviations” of full explanations. However, this will not be the case for all such explanations. To illustrate this point, consider one of the central paradigms of empirical research in memory which is focused on false recall and false recognition (cf. Roediger, McDermott, & Robinson, 1998). Thereby, memory recall is shown to be often influenced by associative processes. This effect perfectly fits the picture presented here: Since templates only serve to construct representations, whereby the content of the constructed representation is not fully determined but only constrained by the template, so-called ‘‘false” memories are even likely to occur. This line of research explicitly makes reference to the retrieval mechanisms and shows that they cannot be assumed to be stable (i.e. they cannot be assume to involve the same associations every time they are employed). However, the focus on false memory is not justified if we want to learn about the structure of the stored templates, since they cannot be true or false. If an occurrent representation that is constructed on the basis of a stored template turns out to be false, this does not mean that something wrong has been stored (it could equally well be that a different retrieval mechanism would have generated a true representations). Thus, if investigations should be informative about the structure of the stored templates, the strong focus on ‘‘false” memories has to be abandoned: Both true and false occurrent representations based on memory are constructed with the same mechanisms (this means that ‘‘true” memories are no less based on associations). Therefore, the content and structure of both true and false occurrent memories equally contribute to an understanding of the underlying templates, since they constrain the results of the constructive retrieval mechanisms in the same way in every case. However, these constraints can only be studied if we focus on the whole range of contents that are constructed from memory, and not only on false contents. 5. Conclusion I have argued that each theory of representation has to have a ‘‘functionalistic” core, i.e. it has to define representations as substitutes for the represented objects in functions which describe the behavior to be explained. Only such theories of representation can do justice to the use of the notion of representation as it is used in the cognitive sciences. This implies that the content of a representation can only be determined if the representation is in use. The consequence of this view is that we do not have an adequate theory of stored content. Quite on the contrary, contents cannot be stored since they cease to be contents as soon as they are no longer in use. Because they can be used to construct different representations when retrieved, a dispositional theory of stored content cannot succeed either (it would run into the disjunction problem). I have proposed to use the term ‘‘template” to describe the entities that are stored in long-term memory. They are not representations and do not have contents, but they can be used to construct representations (including representations about past events = memories). Some implications from this conclusion are shortly discussed. Thereby, I hope to have shown that the usual way of talking about memory as storing content is central to philosophical issues about memory and to empirical methods of investigating memory. However, this usual way is fundamentally mistaken such that we need some detailed account of memory that can deal with the absence of content for stored templates. This paper is certainly not able to present such an account. It rather tries to set up the stage for the problem itself and for a possible direction for a solution. Acknowledgments This paper is based on a presentation given at the international conference ‘‘Memory and Self-Understanding” (Delmenhorst, Germany, June 2009); I would like to thank the organizers and participants for this stimulating event and fruitful discussions. Moreover, I would like to thank Christoph Michel and three anonymous reviewers for very helpful remarks which improved earlier versions of the paper dramatically. References Alba, J. W., & Hasher, L. (1983). Is memory schematic? Psychological Bulletin, 93, 203–231.
846
G. Vosgerau / Consciousness and Cognition 19 (2010) 838–846
Barsalou, L. W. (1992). Frames, concepts and conceptual fields. In Adrienne Lehrer & Eva Feder Kittay (Eds.), Frames, fields and contrasts (pp. 21–74). Hillsdale, Hove and London: Erlbaum. Bartels, A. (2005). Strukturale Repräsentation, Mentis, Paderborn. Bartels, A., & May, M. (2009). Functional role theories of representation and content explanation: With a case study from spatial cognition. Cognitive Processing, 10, 63–75. Bernecker, S. (2008). The metaphysics of memory. New York: Springer. Bernecker, S. (2010). Memory: A philosophical study. Oxford: Oxford University Press. Block, N. (1978). Troubles with functionalism. In C. W. Savage (Ed.), Perception and cognition: Issues in the foundation of psychology. Minneapolis: University of Minnesota Press. Burge, T. (1993). Content preservation. The Philosophical Review, 102, 457–488. Cheng, K. (1986). A purely geometric module in the rat’s spatial representation. Cognition, 23, 149–178. Cummins, R. (1996). Representations, targets, and attitudes. Cambridge, MA, London: The MIT Press. Dretske, F. (1986). Misrepresentation. In Radu J. Bogdan (Ed.), Belief, form, content, and function (pp. 17–36). Oxford: Clarendon Press. Fodor, J. (1987). Psychosemantics. Cambridge, MA, London: The MIT Press. Frege, G. (1892). Über Sinn und Bedeutung. Zeitschrift für Philosophie und philosophische Kritik, 100, 20–50. Gallistel, C. (1993). The organization of learning. Cambridge, MA: The MIT Press. Jung, E.-M., & Newen, A. (2010). Knowledge and abilities: The need for a new understanding of knowing-how. Phenomenology and the Cognitive Sciences, 9, 113–131. Lettvin, J. Y., Maturana, H. R., McCulloch, W. S., & Pitts, W. H. (1959). What the frog’s eye tells the frog’s brain. Proceedings of the IRE, 47(11), 1940–1951. Matthen, M. (2010). Is memory preservation? Philosophical Studies, 148, 3–14. Millikan, R. G. (1984). Language, thought, and other biological categories. Cambridge, MA, London: The MIT Press. Millikan, R. (1994). Biosemantics. In Stephen Stich (Ed.), Mental representation: A reader (pp. 243–258). Cambridge, MA, Oxford: Blackwell. Minsky, M. (1975). A framework for representing knowledge. In P. H. Winston (Ed.), The psychology of computer vision. New York: McGraw-Hill. Putnam, H. (1975), The Meaning of ‘‘Meaning”, in ‘Mind, Language, and Reality’ (pp. 215–271). Cambridge: Cambridge University Press. Putnam, H. (1981). Reason, truth, and history. Cambridge: Cambridge University Press. Ramsey, W. M. (2007). Representation reconsidered. Cambridge: Cambridge University Press. Roediger, H. L., McDermott, K. B., & Robinson, K. J. (1998). The role of associative processes in creating false memories. In M. A. Conway, S. E. Gathercole, & C. Cornoldi (Eds.), Theories of memory II (pp. 187–245). Hove, UK: Psychological Press. Ryle, G. (1945). Knowing how and knowing that. Proceedings of the Aristotelian Society, 46, 1–16. Schacter, D. L., Norman, K. A., & Koutstaal, W. (1998). The cognitive neuroscience of constructive memory. Annual Review of Psychology, 49(1), 289–318. Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals and understanding: An inquiry into human knowledge structures. Oxford, England: Lawrence Erlbaum. Stanley, J., & Williamson, T. (2001). Knowing how. The Journal of Philosophy, 98, 411–444. Vosgerau, G. (2006). The perceptual nature of mental models. In C. Held, M. Knauff, & & G. Vosgerau (Eds.), Mental models and the mind: Current developments in cognitive psychology, neuroscience, and philosophy of mind (pp. 255–275). Amsterdam: Elsevier. Vosgerau, G. (2008). Adäquatheit und Arten mentaler Repräsentationen. Facta Philosophica, 10, 67–82. Vosgerau, G. (2009). Mental representation and self-consciousness. From basic self-representation to self-related cognition, Mentis, Paderborn. Vosgerau, G. (in press). Varieties of representation. In A. Newen, A. Bartels, & E. Jung, (Eds.), Knowledge and representation. Mentis, Stanford, Paderborn: CSLI Publications. Vosgerau, G., Schlicht, T., & Newen, A. (2008). Orthogonality of phenomenality and content. American Philosophical Quarterly, 45, 309–328.
