No title

Consciousness and Cognition EDITOR-IN-CHIEF Bruce Bridgeman University of California, Santa Cruz ASSOCIATE EDITORS Axel ...

Author: Consciousness and Cognition - Elsevier

476 downloads 554 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Consciousness and Cognition EDITOR-IN-CHIEF Bruce Bridgeman University of California, Santa Cruz ASSOCIATE EDITORS Axel Cleeremans

Leah Light

Universite´ Libre de Bruxelles

Pitzer College

Antti Revonsuo

Daniel Smilek

University of Turku

University of Waterloo

EDITORIAL BOARD Jackie Andrade

Stevan Harnad

W. Trammell Neill

University of Plymouth

Princeton University

University at Albany

Bernard J. Baars

Steven A. Hillyard

Keith Oatley

The Neurosciences Institute San Diego

University of California San Diego

Ontario Institute for Studies in Education

Talis Bachmann

J. Allan Hobson

Steven Palmer

University of Tartu

Massachusetts Mental Health Center

University of California Berkeley

MRC Applied Psychology Unit

Larry L. Jacoby Washington University St. Louis, MO

John Pani

W. Banks

Alan Baddeley

Pomona College

John Bargh

E. Roy John

New York University

New York University Medical Center

Arthur L. Blumenthal

John F. Kihlstrom

The New School University

University of Louisville

Ernst Pçppel Ludwig-MaximiliansUniversitt, Mnchen

William Prinzmetal University of California Berkeley

Gordon H. Bower

University of California Berkeley

Stanford University

Christof Koch

Deborah Burke

California Institute of Technology

Brooklyn College of CUNY

Pomona College

Stephen M. Kosslyn Harvard University

Eyal Reingold

Wallace Chafe University of California Santa Barbara

Alfred B. Kristofferson

David Chalmers

David LaBerge

University of Arizona Tucson

University of California Irvine

Antonio Damasio

Stephen LaBerge

University of Iowa

Stanford University

Meredyth Daneman

Donald G. MacKay

University of California Los Angeles

University of Toronto

University of California Los Angeles

Jonathan W. Schooler

Richard Davidson

Ontario, Canada

Arthur Reber

University of Toronto

David Rosenthal Graduate School of CUNY

Daniel Schacter Harvard University

Arnold Scheibel

University of Pittsburgh

University of Wisconsin

George Mandler University of California San Diego

Tim Shallice

Daniel C. Dennett

Bruce Mangan

Jerome L. Singer Yale University

Hamburg University

University of California Berkeley

David Spiegel

Matthew Erdelyi

Anthony Marcel MRC Applied Psychology Unit

Stanford University School of Medicine

Hazel R. Markus

Petra Stoerig

Tufts University

Andreas K. Engel

Brooklyn College of CUNY

Owen Flanagan Duke University

David Galin Langley Porter Psychiatric Institute, San Francisco

University College, London

University of Michigan

Heinrich-Heine-Universitat

Philip M. Merikle

Giulio Tononi

University of Waterloo

The Neurosciences Institute

Thomas Metzinger

Geoffrey Underwood University of Nottingham

Dartmouth College

Johannes GutenbergUniversitt Mainz

Anthony G. Greenwald

Jeff Miller

Harvard University

University of Washington

University of Otago

Charles Yingling

Henk J. Haarman

Michael C. Mozer

University of Maryland

University of Colorado

University of California San Francisco

Michael S. Gazzaniga

Daniel M. Wegner

Consciousness and Cognition 20 (2011) 1

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Editorial Consciousness and Cognition is undergoing several transitions. In the summer of 2010, after our esteemed Editor-in-Chief Prof. William Banks fell ill, a backlog of manuscripts developed. Bill’s Pomona colleagues Prof. Leah Light and Prof. Debby Burke identiﬁed the backlog and donated many hours of their time to working through it with skill and dedication. I don’t know what would have happened to the journal without their efforts. With the help of the editorial board, the backlog was resolved, but Bill was unable to resume his post. As of this writing he remains gravely ill. He asked me to take over as editorin-chief, and I was honored to accept. Bill had shepherded me through learning the ropes of editing in my years as an associate editor of the journal, and I owe him a debt of gratitude for his guidance. As founding co-editor of the journal along with Bernie Baars, Bill set the tone and standards, guiding the journal to the status it has today. He negotiated its association with Elsevier, a major publisher of top journals. That association has provided professional staff, worldwide distribution, and a structure that has served the journal well. I thank the journal staff in the Elsevier San Diego ofﬁce, and especially Ann Barajas, for working beyond the call of duty during our summer overload and for continually providing the highest level of professional support. Publisher Sam Hodder in England has also provided unwavering support. The journal has seen other changes recently. Longtime associate editor Jim Enns has moved on to become Editor-in-Chief of a ﬂagship APA journal, the Journal of Experimental Psychology: Human Perception and Performance. Even with this new assignment, he volunteered to see the manuscripts in his portfolio through to ﬁnal decisions, and performed this task admirably. Leah Light has now taken on the Associate Editor position opened by the departure of Jim Enns, to my delight. The journal continues to prosper, with 324 submissions in the calendar year 2010, nearly one per day. The editorial policy of the journal will not change, continuing to emphasize empirical contributions and also accepting reviews and critiques that advance theory. Interest in consciousness as the big unanswered question in cognitive science, partly facilitated by Bill Banks and his role in developing this journal, continues to grow. Bruce Bridgeman Editor-in-Chief

1053-8100/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2011.02.002

Consciousness and Cognition 20 (2011) 2–3

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Introduction

Brain and Self: Bridging the Gap q

1. Introduction Five years ago almost to the month, Consciousness and Cognition published a special issue devoted to the topic of the ‘‘self’’. That issue was entitled The Brain and Its Self, an homage to Popper and Eccles’ 1977 book The Self and Its Brain, one of the earlier efforts during the modern era of neuroscience to attempt the integration of consciousness, self and the brain. The twist on the title, of course, was that while Popper and Eccles had advocated an explicitly dualistic approach to the problem of consciousness and self, the editors of this particular C & C issue hinted that Popper and Eccles had put the ‘‘Cart (esian)’’ before the horse! That remarkable 2005 issue was based upon a workshop on the topic of the self that was held at Washington University in St Louis in April 2004, which at that time was something of a novel meeting. Although there had been a burgeoning interest in the neuroscience and philosophy of consciousness, the neuroscience of the self was less developed as a separate area of investigation. However, anyone who has watched the ﬁeld develop knows that this situation has radically changed, and even from the time of that conference to the present there has been an enormous increase in interest in the study of how the brain creates and shapes a self. It was in this context that in early 2009 I approached William Banks, editor-in-chief of this journal, and enquired whether he thought it would possible to do another special issue on this topic, a proposal that he enthusiastically supported. Thus ‘‘Brain and self: Bridging the Gap’’ was born. This issue brings together an eclectic group of superb investigators and writers on the topic of the self. These authors come from different countries, schools of thought, backgrounds, domains of investigation, and points of view and each has something important to contribute to the topic. The contributions are roughly grouped into four sections. The ﬁrst contains ﬁve articles that focus on the neurobiological underpinnings of the self. In my contribution I address the neurological basis of the self and propose that the self is constructed upon three hierarchical systems, the interoself, the integrative self, and the exterosensorimotor systems. The nestedness of the neural hierarchy as subjectively experienced allows mental uniﬁcation and other unique features of consciousness and the self. Markowitsch and Staniloiu discuss how memory functions contribute to the self. Based upon neuroanatomical, neuroimaging, neurodevelopmental and evolutionary considerations they examine how the self, autonoetic consciousness and episodic-autobiographical memory are intimately interlocked and play an essential role in the creation of trans-temporal cohesiveness of self across contexts. Devue and Brédart focus on the neural correlates of self-recognition from facial stimuli. They ﬁrst address the issue of hemispheric lateralization of self-recognition and then consider more speciﬁc correlates of self-recognition derived from neuroimaging studies. They review the functional role of each region speciﬁcally activated during self-recognition and discuss the use of self-recognition tasks to investigate the neural correlates of self-awareness. Northoff, Qin, and Feinberg consider conceptual–experimental approaches to the self and its neuroanatomical substrate of the self. They ﬁrst distinguish content- and processed-based concepts of the self and argue that these entail different experimental strategies and anatomical substrates. They propose a novel view on the anatomy of an integrated subcortical–cortical midline system and discuss evidence that suggests that the anterior paralimbic and midline regions do indeed seem to be speciﬁc for self-speciﬁc stimuli. Finally, Sinigaglia and Rizzolatti describe the mirror neuron system and consider how it impacts the sense of self and others. They argue that the distinction between the self and the others is not a general sub-case of the distinction between the self and world but rather that the mirror neuron mechanism reveals an intrinsic link between sense of the self and others with both being rooted in their common ‘‘motor potentialities’’. The next four articles focus on how brain dysfunction creates clinical disorders of the self. I refer to these disorders as neuropathologies of the self, and in my article I address perturbations of the bodily, relational, and narrative self in which right, especially medial-frontal and orbitofrontal lesions play a leading role. I discuss how these brain lesions create ego q

This article is part of a special issue of this journal on Brain and self: Bridging the Gap.

1053-8100/$ - see front matter Ó 2010 Elsevier Inc.. All rights reserved. doi:10.1016/j.concog.2010.12.001

Introduction / Consciousness and Cognition 20 (2011) 2–3

3

disequilibrium, a disturbance of ego boundaries and ego functions that encourages the emergence of developmentally immature styles of thought, ego functioning, and psychological defense. A four-tiered hierarchical model of these conditions is proposed that emphasizes a multifactorial approach including both negative and positive, bottom up and top down, and neuropsychological and psychological variables. de Vignemont addresses the concept of embodiment in relationship to body ownership and disownership. She considers recent investigations of artiﬁcial embodiment of allograft, prostheses, rubber hands, and virtual avatars and considers how these relate to the notion of the plasticity of the representation of the body. She proposes a deﬁnition and analysis of embodiment as a ‘‘speciﬁc information processing system that entails the representation of one’s body.’’ Uddin discusses self-related cognition in cerebral commissurotomy patients and considers what these patients teach us about the role the commissures play in maintaining the sense of self and self-representation. She concludes that an intact corpus callosum enables interhemispheric transfer that is necessary for some but not all types of self representations. Finally, at the conclusion of this section, Sierra and David discuss the clinical syndrome of depersonalization as characterized by disembodiment and subjective emotional numbing. These authors consider the combined roles that fronto-limbic (particularly anterior insula), parietal and prefrontal regions play in the etiology of this disorder. The third group of articles focus on the self in relationship to early child development. First, Rochat explores the interaction among genes, brain, and the environment in the development of the self. He discusses how brain development, some of it prenatal, ensures that normally developing newborns already have subjective experience and minimal self-awareness that rapidly – within the ﬁrst years of life – evolves into explicit self-consciousness that by 3 years of age evolves into a sense of moral agency. Lewis posits that the ‘‘idea of a self’’ is made up of at least two major aspects, the machinery of the self that is comprised of unconsciousness, unreferenced action of the body, including its physiology and its processing of information that in turn includes cognitions and emotional states, and the mental state of the idea of ‘‘me’’, that part of the self that makes reference to itself and develops over the ﬁrst 2 years of life. The growth of the self in this view is a function of both brain maturational processes and socialization. Lombardo and Baron-Cohen focus on childhood autism and ‘‘mindblindness’’ as the core of the social-communication impairments in autism spectrum conditions (ASC). They discuss how the mindblindness in ASC affects processes impacting the autistic child’s sense of self and other and how a better understanding of these processes might facilitate social-communicative abilities in ASC. The ﬁnal section emphasizes some philosophical approaches to theories of the self. Zahavi and Roepstorff analyze two paradigms in the neuroscientiﬁc investigation of self – the study of facial self-recognition and the study of adjectival selfattribution. They examine the implicit assumptions regarding the self that are contained within these investigations and argue that ‘‘conceptual and theoretical reﬂections on the structure, function and nature of self have either disappeared altogether or receded into the background’’. They urge that future investigations employ a multidisciplinary approach that takes into account both philosophical as well as neuroscientiﬁc approaches. In their article, Gallagher and Cole consider whether verbal narratives can in and of themselves be considered ‘‘pathological’’. They consider different analytic approaches to narrative analysis that occur in select psychopathological conditions and propose that the interpretation of such narratives must take into account a variety of phenomenological and philosophical issues. They describe an empirical study of ‘‘narrative distance’’ and discuss how the results can be interpreted in two different ways with regard to the issue of dissociation. Finally, Hirstein and Sifferd try to narrow the gap between how the legal system deﬁnes and thinks about the self and what neuroscience has learned about it. They believe that neuroscience is a useful tool for understanding the connection between mental processes and the law, and chart a path for how these differing points of view may be reconciled. Speciﬁcally, they argue that ‘‘executive processes are the seat of a person‘s decision-making, intention-forming, planning, and behavior-inhibiting processes, all of which are absolutely crucial to his legal and ethical being. Hence we call the set of executive processes – the legal self’’. I trust the reader will agree that these articles taken together represent an excellent overview of current thinking and research on the self. Todd E. Feinberg Departments of Neurology and Psychiatry, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, USA Beth Israel Medical Center, New York, NY 10003, USA E-mail address: [email protected] Available online 8 January 2011

Consciousness and Cognition 20 (2011) 4–15

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The nested neural hierarchy and the self q Todd E. Feinberg Albert Einstein College of Medicine, Yarmon Neurobehavior and Alzheimer’s Disease Center, Beth Israel Medical Center, First Avenue at 16th Street, New York, NY 10003, United States

a r t i c l e

i n f o

Article history: Available online 15 October 2010 Keywords: Nested neural hierarchy Neuroanatomy of self Interoceptive Integrative self system Interoself system Exterosensorimotor system

a b s t r a c t In spite of enormous recent interest in the neurobiology of the self, we currently have no global models of the brain that explain how its anatomical structure, connectivity, and physiological functioning create a uniﬁed self. In this article I present a triadic neurohierarchical model of the self that proposes that the self can be understood as the product of three hierarchical anatomical systems: The interoself system, the integrative self system, and the exterosensorimotor system. An analysis of these three systems and their functional features indicates that the neural hierarchy possesses features of both non-nested and nested hierarchies that are necessary for the creation of a uniﬁed consciousness and self. These functional properties also make the central nervous system a biologically unique entity unlike anything else in nature. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction The question I address in this paper is: ‘‘How does the brain create a self?” For the purposes of the present inquiry, I deﬁne the self by three critical features – its relationship to consciousness, its unity, and its persistence in time. Thus, I propose the following deﬁnition: the self is a unity of consciousness in perception and action that persists in time (Feinberg, 2009). This definition of the self presupposes that there must be something conscious, that the consciousness in question must exist in some sense uniﬁed, and that this consciousness possesses a temporal aspect. In other words, whatever there is ‘‘something that it is like to be” any conscious organism (Nagel, 1974) must be uniﬁed and endure beyond the instantaneous moment of that organism’s perception and action. While there is an emerging ‘‘neuroscience of the self” that addresses important issues regarding which brain structures are critical for the creation and maintenance of the self, we currently lack unifying models that ‘‘bridge the gap” between the microstructure, macrostructure, connectivity and physiology of the brain and the uniﬁed sense of self as subjectively experienced. In this article I present a neurobiological model of the self but focus primarily on the issues of the self as a unity of consciousness in perception and action. For a discussion of the issue of its persistence in time, see Feinberg, 2009. 2. Three hierarchical systems create the self I have argued previously (Feinberg, 2009) that the brain creates the neural apparatus of the self through three dissociable hierarchical systems: the anatomically central/medial interoself system, the anatomically peripheral exterosensorimotor system, and the integrative self system that is interposed between the other two (Fig. 1). These three systems can be understood as the result of the two major neuroanatomical patterns of organization. The ﬁrst is the radial pattern (or medial–lateral trend; Feinberg, 2009) in which the neuroaxis is organized concentrically from its center or core outward to its periphery q

Forthcoming in T. E. Feinberg (Ed.) Brain and self: Bridging the gap. A special issue of Consciousness and Cognition. E-mail address: [email protected]

1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.016

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

5

Fig. 1. The neural self is constructed from three interlocking system(s): The interoself system, the integrative self system, and the exterosensorimotor system. Lower levels of each system are nested within higher (more abstract) levels. These levels can be visualized as anatomically concentric rings with the interoself system representing the anatomically innermost (core) ring that is synaptically closest to the internal milieu and serves the requirements of homeostasis, the exterosensorimotor system is anatomically peripheral, synaptically closet to the external environment and serves the needs of the organism’s interaction with the external environment, and the integrative self system is anatomically intercalated between the other two.

like the growth rings of a tree. The second is a caudal–rostral pattern that results in the hierarchical growth of later developing and more complex neurological structures superimposed and added upon those with simpler and more basic organization and function. These two patterns of organization are evident along the length of the neuroaxis and result in three major systems that contribute and deﬁne the neural substrate of the self. 2.1. The interoself system The anatomical structure of the interoself system or systems reﬂects the two aforementioned patterns of neural organization. The radial pattern of organization of the neuroaxis is evident at the hierarchically lowest and phylogenetically earliest evolved parts of the central nervous system and is evident caudally at least at the level of the brainstem. The progressive extension of the core or center of the brainstem outward from the centrally located cerebral aqueduct creates concentrically or radially arranged periaqueductal sectors. Nieuwenhuys (1996, 1998) and colleagues (Nieuwenhuys, Veening, & van Domburg, 1988–1989; Nieuwenhuys, Voogd, & van Huijzen, 2007) term the zones located in closest proximity to the aqueduct the core of the neuroaxis and the sectors lateral to these the medial and lateral paracore (Fig. 2). According to their analysis, these medial core–paracore zones are responsible for the homeostatic regulation of the animal’s bodily systems and serve as well as for instinctual and self-protective behaviors. The sensory systems of the core–paracore regions are chieﬂy concerned with interoceptive stimuli (such as pain, thirst, hunger, etc.) and the feelings generated from the core regions are self referential in that they are experienced as originating from within the body (‘‘I am hungry, afraid, in pain” etc.) as opposed to being referred to the outside environment (see below, projicience). The sensations generated from the core of the neuroaxis are intrinsically motivational, a feature that interoceptive feelings share with emotions in general (Craig, 2002, 2003a, 2003b, 2003c, 2009; see Fig. 4). Organized in a radial ring surrounding the core–paracore zones are the exterosensorimotor systems. These sectors form the radially outermost or anatomically peripheral zones of the brainstem (Fig. 3). In contrast to the anatomically central core and paracore structures, these regions mediate the interaction of the animal with the external environment and represent the classical sensorimotor systems (Nieuwenhuys, 1996, 1998; Nieuwenhuys et al., 1988–1989, 2007) a point I will return to later. The interoself system displays extension in a caudal–rostral direction to the hierarchically higher levels of the neuroaxis. Nieuwenhuys and co-workers describe a caudal–rostral extension of the core and paracore system into the diencephalon and forebrain in which are included a large number of medially located structures that display many of the essential features of the core and paracore regions that play a key role in the internal sense of self. In support of this model, Nauta (1958, 1973), Nauta and Haymaker (1969) demonstrated that there are robust connections between the medial and paramedian zones within the midbrain and the higher limbic regions within the forebrain that are consistent with the caudal–rostral trend outlined above. Nauta envisioned a limbic forebrain–limbic midbrain complex

6

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

Fig. 2. Transverse sections of the brainstem. C – core; LPC – lateral paracore; MPC – median paracore. (Based upon Nieuwenhuys, 1996, 1998;Nieuwenhuys et al., 2007) If one then views the brainstem in cross-section, its ‘‘center,” the most medial or innermost zones, are the sectors that are situated closest to the aqueduct of the ventricular system, the cerebral spinal ﬂuid ﬁlled cavities of the brain. Nieuwenhuys and co-workers describe three periaqueductal zones in the brainstem that share anatomical, neurochemical, and functional features These areas are designated the core, the medial paracore, and the lateral paracore of the neuroaxis. The extero sensorimotor systems are anatomically peripheral in the brainstem and can be seen on its external surface (Fig. 3).

Fig. 3. Diagram of the brainstem and diencephalon. The brainstem is the most caudal (lowest) part of the brain and is the region that is in closest proximity to the spinal cord and it represents its rostral (upward) extension. The motor pathways (cerebral peduncle and pyramid) and sensory relay pathways (e.g. medial and lateral geniculate bodies) are located anatomically on the periphery of the brainstem and can be readily observed on the surface of the brainstem in contrast to the central/medial locations of the interoself systems (Fig. 2).

consisting of interconnected structures that extends from the amygdala and hippocampal formation rostrally to the septal and preoptic regions, hypothalamus, and the mesencephalic raphé nuclei and central gray caudally. Expansion of this early work on the forebrain–limbic midbrain connections led to the description of the greater (Nieuwenhuys, 1996, 1998;

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

7

Fig. 4. The greater limbic and cortical self systems. Top. Rostral extension of medially located core and paracore regions into the forebrain creates the greater or extended limbic system. Within this system are included are a large number of medially located structures that display many of the essential features of the core and paracore (Based upon Nieuwenhuys, 1996, 1998;Morgane & Mokler, 2006; Morgane et al., 2005; Nieuwenhuys et al., 2007) The rostral extension of these systems include medially located paralimbic and heteromodal cortices critical for self-related functions. (The cortical midline structures (CMS); Northoff & Bermpohl, 2004; illustration adapted from Feinberg, 2009, with permission.) Bottom. These levels are visualized as anatomically concentric rings with the interoself system representing the innermost (core) ring synaptically closest to the internal milieu, the exterosensorimotor system is anatomically peripheral, synaptically closet to the external environment with the integrative self system roughly intercalated between the other two. The insula (insert) is considered part of the paralimbic ring.

Nieuwenhuys et al., 1988–1989, 2007) and the distributed (Morgane, Galler, & Mokler, 2005; Morgane & Mokler, 2006) limbic system comprised of medially located structures that play a critical role in the creation of the internal sense of self . The behaviors contingent upon these structures are varied and complex. The interoself system contributes to the organism’s relationship to the internal milieu, and serves homeostatic and self preservative functions. The interoself system is specialized therefore for the processing and awareness of aspects of the self that pertain to homeostatic internal processes, self-preservation, motivation, and emotion. With reference to the role of behaviors served by one of the critical centers within this system, the hypothalamus, Nieuwenhuys notes ‘‘All of these behavioral patterns are related to the maintenance of the internal milieu (homeostasis), to the maintenance of the integrity of the individual or to the preservation of the species. In fact, the entire hypothalamus can be deﬁned as a center which generates integrated somatomotor, visceromotor and endocrine responses directly aimed at the survival of the individual and of the species.” There are other medially located homeostatic and hence self-related systems that overlap with that system described by Nieuwenhuys. For example, Critchley, Wiens, Rothstein, Öhman, and Dolan (2004) propose that information concerning the internal state of the body and subjective feeling states is conveyed via a lamina 1 spinothalamocortical pathway that projects rostrally to ‘‘interoceptive centers” located in insular (especially right) and orbitofrontal cortices. Sewards and Sewards

8

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

(2002) proposed that the motivational aspects of pain originate in spinal laminae VII and VII and deep layers of the trigeminal complex and project to periaqueductal gray and posterior hypothalamic nucleus (structures that are within the core– paracore and greater limbic system) and the intralaminar thalamic nuclei that in turn project to anterior cingulate cortex. In one of the most recent and complete models, Craig (2002, 2003a, 2003b, 2003c, 2009) suggests that interoceptive information regarding the physiological condition of the entire body (including pain sensation) travels from lamina 1 spinal neurons (sympathetic pathway) and the nucleus of the solitary tract (NTS; parasympathetic pathway) to related portions of the thalamic ventromedial nucleus (VMpo and VMb, respectively) which in turn project to the dorsal posterior insular cortex. This information is re-represented in the mid- and anterior insular cortex (AIC) to create a model of the internal bodily self (Fig. 5, right side).These systems carry primarily interoceptive information and are reciprocally connected and act in tandem with the core–paracore homeostatic systems. In Fig. 5 I have summarized some of the major hierarchically organized struc-

Fig. 5. Two medially located overlapping homeostatic and interoceptive systems. On the left side of the ﬁgure is a pattern of organization based primarily on the core–paracore system of Nieuwenhuys and co-workers described above and on the right side is a schematic representation of the homeostatic interoceptive pathways as emphasized by Craig and others (see text for details).

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

9

tures and pathways that contribute to the interoself system. The left side of the ﬁgure outlines a pattern of organization based primarily on the core–paracore system of Nieuwenhuys and co-workers described above and on the right side is a schematic representation of the homeostatic interoceptive pathways as emphasized by Craig and others. The key to understanding how the neural self is organized at its highest levels lies in how the medial–lateral and caudal– rostral trends interact as they ascend the neural hierarchy. One of the most useful and best known ways of organizing the hierarchical aspects of the caudal–rostral trend has been proposed by Mesulam (2000). Mesulam suggests that the cerebral cortex can be roughly organized into ﬁve hierarchically arranged subtypes – limbic, paralimbic, heteromodal association, unimodal association, and primary sensory-motor regions. Mesulam’s neural hierarchy is based upon several organizing principles. For example, while each zone within the hierarchy has both extensive extramural connections with elements of other functional regions, as well as intramural connections within its own functional zone, the most robust connections of each region are within its own region and those immediately lower or higher on the hierarchy. In this way a clear hierarchical pattern of connectivity is maintained in spite of extensive cross-talk between different regions at various hierarchical levels. Another feature of Mesulam’s hierarchy is that it utilizes the features of brain anatomy that are felt to reﬂect more primitive versus later evolving structure. Levels that are lower on the hierarchy such as the limbic regions are characterized by primitive cortex with a relatively simple neural architecture, fewer clearly deﬁned cellular layers and less differentiated cell types when compared with zones higher on the neural hierarchy. The hierarchically highest zones show later evolving architectonic features such as an increase in the number of well differentiated cellular layers (a feature called lamination) and more highly specialized cell types. There is also an increase in cell density as ones moves from lower to higher zones (Mesulam, 2000). Mesulam’s limbic (corticoid and allocortical) and paralimbic areas represent the cortical extension of Nieuwenhuy’s greater limbic structures. The insula is a nodal cortical structure in both pathways illustrated in Fig. 5. As noted by Heimer and Van Hoesen (2006) it is often overlooked that although the insula is traditionally thought of as having a ‘‘lateral” location in the cerebral hemisphere, it actually is part of the limbic lobe, and thus it is readily integrated into the basic interoceptive–exteroceptive, medial–lateral, central–peripheral dichotomization I have proposed. As they observed: It is not always appreciated that a large part of the insula, which is located in the depth of the lateral sulcus on the lateral side of the hemisphere, is included in the limbic lobe. . . the anterior and ventral parts of the insular cortex is non-isocortical in nature (Mesulam & Mufson, 1985), and its anteroventral part is directly continuous with the posterior orbital cortex. In other words, a large anteroventral part of the insula is an integral part of a continuous circular structure formed by the limbic lobe (Yakovlev, 1972). The insula and neighboring parts of the limbic lobe are currently being promoted as especially important in the context of feelings. (Heimer & Van Hoesen, 2006, p.136) 2.2. The exterosensorimotor system It is an underappreciated fact that just as the medially located core–paracore systems maintain their anatomically central localization within the cortical zones of the greater limbic system, the great sensory and motor pathways that are involved with the transmission and action upon external sensory information maintain a peripheral location within the cortex anatomically removed and synaptically distant from the medially located systems (Figs. 1 and 3). As in the brainstem, the sensory aspects of these higher order systems are exteroceptive, that is, they respond to stimuli from the environment, and in the case of the special senses (vision, hearing, taste, and olfaction) the feelings or qualia generated by these regions display a particular phenomenal quality known as projicience, a term introduced by Sherrington (1947) to designate the mental projection of sensations into the external environment, a mechanism that is characteristic of the distance receptors. In contrast to the interoceptive stimuli that are generated by the medial core regions and are experienced as emanating from within the organism and relating to the somatic self, the qualia experienced from the external sensory receptors relate to some space or object outside the body (‘‘It looks, sounds, tastes, etc.”). This division of the regional neuroanatomy (medial versus lateral) parallels a similar division between the subjective features of the stimulus (inner versus outer) that may be the single most important factor in creating the margins of the phenomenal self. Like the interoself system, the extero systems make a major contribution to the self. First and perhaps foremost, the highly evolved extero systems – in contrast to the interoself system – create the aforementioned quality of projicience, the mental externalization of stimuli away from body that makes self-object discrimination possible, a quality of mind that is critical for the creation of consciousness and the self. 2.3. The integrative self system Finally, the integrative self system plays a third and possibly determining role in the creation of the self. This system serves to assimilate the interoself systems with the extero systems, and mediate of the organism’s internal needs with the external environment. Most important for the integrative self system are the heteromodal association cortices that are situated between the paralimbic zones that are most allied with the limbic functions of the maintenance of homeostasis and self-preservation and the conditions of the internal milieu, and the unimodal zones that represent the most abstract aspects of sensorimotor processing. These regions serve as a ‘‘convergence zone” (Damasio, 1999) for the integration of the other

10

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

two systems. According to the current model, the highest most advanced aspects of the self are organized within the integrative self system where the two trends converge (Fig. 6). Consistent with this model recent brain imaging studies have speciﬁcally implicated medially located cortical heteromodal zones to be involved in diverse self-related functions. Northoff and co-workers have suggested that a group of cortical midline structures (CMS) constitute the anatomical core of more evolved aspects of the self (Northoff & Bermpohl, 2004; Northoff et al., in press; Northoff et al., 2006, 2009). Many of the CMS structures that are activated during self-related tasks may actually be viewed anatomically as the rostral extension of the interoself systems, and therefore overlap to a large extent with the greater limbic system and emotional systems in general. With regard to one heteromodal region, the dorsolateral prefrontal cortex (DMPFC), Northoff et al. (2009) noted that many studies have provided evidence that self-related stimuli induce higher activity in the DMPFC. Schore (2003) notes another heteromodal region, the OMPFC, serves as a critical integration zone between internal (interoceptive) and external (exteroceptive) information. In summary, the rostral extension of the medial–lateral trend gives rise to three inter-related self related hierarchically organized systems. The interoself system is concerned with the internal milieu and homeostatic needs of the organism. The products of processes generated by the interoself system are experienced as feelings coming from within the organism itself. The exterosensorimotor system is devoted to the organism’s interactions with the environment, including sensations experienced as coming from the external world. In the evolution of our species, as well as in our own individual development during our lifespan, an enormous increase in brain growth and complexity results in a unique expansion of the integrative self system. It is this system, represented for the most part by heteromodal association cortex that is responsible for the most abstract and highest order aspects of the human self (Fig. 6). The result of this process is that the structural organization of the self can be visualized as constructed of a series of concentric rings (Fig. 1) that maintain their respective positions the length of the neuroaxis (Fig. 6).The innermost ring represents the core of the neuroaxis and it creates the interoself system that is most aligned with homeostasis and maintenance of the internal milieu. In contrast, the anatomically outermost (most peripheral) ring is synaptically closest to the external environment and responsible for the perception of the world. The middle zones interposed between the

Fig. 6. A Model of the Neural Self. In the highly schematic model presented here, the neural self is comprised of three self systems in which lower levels of the neural hierarchy are viewed as nested within higher levels, and all levels of the neural hierarchy make a contribution to the self. Levels and their anatomy are derived and adapted primarily from Nieuwenhuys (1996, 1998), Nieuwenhuys et al. (2007) and Mesulam (2000). For the sake of simplicity and emphasis, the numerous cross connections between ascending and descending sensory and motor pathways are not represented in the diagram. (Adapted from Feinberg (2009).)

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

11

two are responsible for the integration of the inner world of feeling and the perception of the external environment. These integrative self systems are responsible for the most advanced aspects of the self. 3. Nestedness and the neural hierarchy At the outset of my discussion I offered a deﬁnition of the self that included a ‘‘unity of consciousness in perception and action”. A fundamental enigma is how the multiple zones and hierarchical levels critical to the self allow the subjective sense of uniﬁcation. The key to understanding how the self and mind are uniﬁed lies in the analysis of the nature of biological hierarchies (Feinberg, 2000, 2001a, 2001b, 2008, 2009; Feinberg & Keenan, 2005). There are two basic types of hierarchies, non-nested or control hierarchies and nested or compositional hierarchies (Fig. 7). A non-nested hierarchy has a pyramidal structure in which higher levels within the hierarchy are not physically composed of the lower levels as, for example, in the metaphor of military command with a general at the top and successively lower levels of command below. While a general directs and controls the operations of the lower levels of the hierarchy he is not physically composed of his subordinates (Fig. 7A). The other type of hierarchy is known as a nested or compositional hierarchy (Fig. 7B). All living organisms when considered without their nervous systems are nested hierarchies. In a nested hierarchy the lower levels of the hierarchy are physically combined or nested within higher levels to create increasingly complex wholes (also known as ‘‘holons”). An organ of the body such as a liver is entirely comprised of its constituent cells, and the entire body, the highest level of the hierarchy of a living organism, is entirely composed of the constituent organs of which it is comprised (Ahl & Allen, 1996; Allen & Starr, 1982; Pattee, 1973; Salthe, 1985). Non-nested and nested hierarchies also differ in the manner in which the hierarchy is centralized. In a non-nested hierarchy such as an army, the number of constituent parts tends to decrease as one moves up the hierarchy and the top of the hierarchy is more physically centralized relative to lower levels. In a nested hierarchy, although parts are organized into greater wholes, the number of constituent parts tends to increase and there is no physically or functionally centralized control of the system. One result of these differences in centralization is that non-nested and nested hierarchies will differ in the degree of constraint (Ahl & Allen, 1996; Allen & Starr, 1982; Pattee, 1973; Salthe, 1985) that higher levels impose upon lower levels. In a non-nested hierarchy there is a centralized top of the hierarchy that controls all lower levels and provides strong constraint of higher upon lower levels. In a nested hierarchy the constraint of the system to a greater degree is embodied within the entire hierarchical system and this creates weak constraint of higher upon lower levels. In nested hierarchies like organisms, individual cells constrain the organelles to perform cellular metabolism, the organs of the body in turn constrain the cells, and at the highest level the entire organism constrains the individual organs to perform all the functions necessary its survival – but there is no ‘‘top command” controlling the entire system and the entire organism’s constraint upon individual cells is slow and incomplete. When we analyze the functions of the nervous system, there are some aspects that are characteristic of non-nested hierarchies. For example, due to the hierarchical arrangement of neuronal processes in sensory pathways (Fig. 6) neurons in heteromodal zones located in anatomically ‘‘downstream” positions in the neuroaxis possess more abstract and integrated response characteristics than neurons in primary sensory-motor and unimodal association zones that are positioned earlier (‘‘upstream”) in the perceptual pathways (Zeki, 1993, 2008). This leads to some cells – referred to as ‘‘pontiﬁcal” cells (also

Fig. 7. Different types of hierarchies. All hierarchies display emergent properties that are created through the interaction of lower and higher levels and constraint of higher upon lower levels. A. In a non-nested hierarchy the entities or holons at higher levels of the hierarchy are physically independent from the entities at lower levels and there is strong constraint of higher upon lower levels. B. In a biological nested hierarchy, higher levels are physically composed of lower levels, and there is no central control of the system resulting in weak constraint of higher upon lower levels. C. A neural nested hierarchy is a unique biological system that displays features of both non-nested and nested hierarchies. (See Feinberg, 2009; Northoff, Feinberg, and Panksepp, in press.)

12

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

called ‘‘grandmother,” ‘‘cardinal,” or ‘‘gnostic” cells) far along on the processing stream that respond to highly integrated visual stimuli such as a hand or a face, or even the face of Julia Roberts or speciﬁc buildings such as the Tower of Pisa (Barlow, 1995; Fried, MacDonald, & Wilson, 1997; Gross, 2002, 2008; Gross, Bender, & Rocha-Miranda, 1969; Gross, Rocha-Miranda, & Bender, 1972; Kreiman, Koch, & Fried, 2000; Quiroga, Kreiman, Koch, & Fried, 2008; Quiroga, Reddy, Kreiman, Koch, & Fried, 2005). Although these cells are ‘‘grandmother-ish” in actually they probably are not nearly as speciﬁc as the term might suggest if strictly applied (Quiroga et al. 2005, 2008). However, for a host of reasons this hierarchical process known as topical convergence or labeled lines coding cannot be the sole explanation for mental uniﬁcation or the ‘‘binding,” of perceptual elements into a uniﬁed awareness (see for example Llinas & Ribary, 2001; Singer, 1999). From the standpoint of the neural hierarchy outlined in Fig. 6, one immediately obvious problem is that as information moves from primary sensory to association cortices, topical convergence creates higher order cells such as a face cells with very large receptive ﬁelds that react to a face appearing almost anywhere in the visual ﬁeld. Thus, cells early in the visual stream with small receptive ﬁelds ‘‘know” where each line of the face is but do not ‘‘know” that a given line is part of a face; and downstream face cells ‘‘know” there is a face in the visual ﬁeld, but because of their large receptive ﬁelds they do not know where the face is located in space. Therefore, although cells in sensory processing pathways project in a hierarchical fashion to code for increasingly speciﬁc complex and abstract properties, the information coded by cells earlier in the pathway cannot be lost to awareness Thus, in order to explain how the neural hierarchy operates, we must explain how the brain could simultaneously be physically distributed across multiple connected but anatomically discrete levels yet allow for the creation of higher order integrated (whole) and abstract conscious awareness but still ensure that both lower and higher levels of the hierarchy make a contribution to the entire mental experience and still be uniﬁed in awareness. The nervous system accomplishes this feat by displaying – in addition to the non-nested features outlined above – functional properties that possess the characteristics of a nested hierarchy (Table 1). Consider the following simple example, the multi-modal experience of observing an apple falling from a tree. The overall shape of the apple is comprised of tens of thousands of individual line segments that are represented very early in the processing stream at the level of V1. As visual information is processed further downstream through visual association cortices, a feature such as the stem emerges in awareness as ‘‘part of” – or nested within – something else, in this case the overall shape of the apple. In this fashion, short line segments are bound to longer line segments to create the outline of the apple, just as a small patch of red of the apple is bound to a larger red patch that is part of the redness of the entire image. Further downstream, as the consciousness representation evolves, the overall shape of the apple is bound to the redness of the apple (via transmission from V1 to color areas in V4 and inferotemporal cortex; Zeki & Marini, 1998) which is bound to its movement (especially via input from area V5; Beckers & ZeKi, 1995) as it falls from the tree. This visual information is eventually bound to the sound the apple makes as it falls to the ground as auditory information initially processed within the temporal cortices is integrated with visual information in heteromodal association cortices, and in consciousness, all of these anatomically distributed and multimodal representations are bound together to create the uniﬁed experience (see for example Cleeremans, 2003; Crick, 1994; Crick & Koch, 1990; Engel, Fries, König, Brecht, & Singer, 1999; Revonsuo, 2006, 2010; Roskies, 1999; von der Malsburg, 1995). In the dynamic systems terms outlined in Table 1, with the unfolding of this process, lower order features combine in consciousness as ‘‘part of” – or nested within – higher order features and to say that an element is ‘‘bound” to another element is simply another way of saying that they are represented in awareness dependently and are nested together (Feinberg, 2001a, 2001b). Consider another classic problem of mental uniﬁcation ﬁrst outlined by Sherrington (1947) – the creation of the cyclopean eye in visual perception. When we use both eyes we seem to see from a ‘‘virtual single eye” located somewhere between our eyes and the bridge of the nose. But of course there is no such Cartesian Theater (Dennett, 1992) where the information from the eyes is ‘‘physically” merged for uniﬁed viewing by an ‘‘inner eye”. Indeed, Sherrington actually proposed that since there was no place in the brain where perception could be physically uniﬁed, consciousness itself was non-material (for discussion, Feinberg, 2001a, 2001b).

Table 1 Features of non-nested and nested hierarchies. Hierarchical organization is the essential to producing organisms with consciousness and selves, and nervous systems that produce consciousness and selves are unique among biological entities in their display of both non-nested and nested hierarchical organization (see Fig. 7).

Lower levels physically nested within higher levels Lower levels functionally nested within higher levels Rapid (millisecond) communication between levels Higher and lower levels contribute to the system Interaction at lower levels contribute to higher levels (emergent properties) Upper levels constrain lower levels (downward constraint) Centralization of levels Representation (modeling)

Non-nested hierarchies

Nested hierarchies

Brain with consciousness

Brain with consciousness & self

+/ + +

+ + +/ + +

+ + + +

+ + ++ ++

+ + +

++ ++ ++

+ + +/

+ +/

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

13

However, in terms of the hierarchical systems approach I outline in Table 1, we may say that the lower order elements – in this case the information from each eye – are constrained to combine into a single uniﬁed visual percept. In this instance, top down constraint allows us to see a single visual percept instead of two independent images. Furthermore, by integrating the information generated by each eye independently the brain creates an additional ‘‘emergent” feature of the visual scene – depth perception – a feature that is only possible by the integration of visual information emanating from both eyes. The essential point is that even though there are convergent (non-nested) pathways that create binocular neurons that are responsive to information from both eyes, the brain does not physically centralize the two visual images, one from each eye. Rather, the two images create a consciousness where their meanings are conjointly nested in awareness to produce a higher level centralization of meaning – a single image now seen in depth. This constraint or control of the whole upon the parts does not physically eliminate these individual parts from consciousness because the image from each eye continues to make a contribution to the uniﬁed cyclopean eye. Rather, the constraint leads to the elimination of the independence of each part from each other when operating within the framework of the nested hierarchy of consciousness. It is within the brain’s design to create this centralization of meaning that provides the constraint that ‘‘pulls” the mind together to form the ‘‘inner I” of the self. In a similar fashion, when the control of action is considered as a nested hierarchy, it is a centralization of purpose that provides the constraint and guiding force of the self, the ‘‘ghost in the machine” (Feinberg, 2001a, 2001b, 2009). This nested quality of uniﬁed awareness is also manifest when one considers the interactions between the intero- and extero aspects of the self as outlined in Figs. 1 and 6. If we are stuck by a pin on the ﬁnger, we feel the pain in the ﬁnger as ‘‘coming from” the pin. The quality of pain ascends the neural hierarchy within the interoself system up and beyond to the insular cortex (ﬁgure Craig, 2002, 2003a, 2003b, 2003c, 2009). But at the same time the extero system aspects of seeing and feeling the pin prick on the surface of the skin is integrated with the experience of pain. What ultimately enters into awareness is the entire nested experience of feeling a sharp pinprick on the tip of the ﬁnger. All aspects of the experience are nested within consciousness in spite of the widely distributed neural substrate of the experience. If consciousness were not organized as a nested hierarchy, the uniﬁed aspects of awareness would not be possible. One plausible way to explain this process of nested mental uniﬁcation is that the brain uses synchronized oscillations to bind perception (Buzsáki, 2006; Crick, 1994; Crick & Koch, 1990; Edelman, 1989; Engel, Fries, & Singer, 2001; Engel & Singer, 2001; Engel et al., 1999; Fries, Roelfsema, Engel, König, & Singer, 1997; Singer, 1999, 2001; Singer et al., 1997). There is increasing experimental evidence that spatially distributed neural networks integrate – in systems terminology entify – a stimulus by temporally synchronizing the ﬁring of neurons or groups of neurons representing the same stimulus. Thus, neurons that are responding to the same object are able to represent that object in consciousness as an integrated percept because they are ﬁring their response to that object in temporal synchrony both within and across hierarchical levels (Engel et al., 1999, 2001) Similar cross-modal synchrony could also be responsible for binding elements from different sensory modalities into single and uniﬁed percepts (Melloni et al., 2007). With the nested model I propose, temporal synchrony could provide the constraint necessary for perceptual binding and serve as an essential mechanism for structuring the hierarchical arrangements of consciousness and the self. Thus, a combination of convergent hierarchical pathways and higher level temporal oscillation could provide sufﬁcient integration of neural activity to enable the binding across and within hierarchical levels (Engel et al., 1999, 2001; Singer, 1999; Uhlhaas et al., 2009) and provide for a nested hierarchical consciousness that I have outlined. As expressed by Uhlhaas et al., these two processes in combination offer a potential solution to the problem of mental unity in the face of widely distributed cortical networks: First, devoted architectures of connections that assume coordinating and binding functions through convergence and divergence of labeled lines, and second, dynamics that allow for the self-organization of ever changing spatio-temporal activity patterns on the backbone of ﬁxed anatomical connections. These two strategies are not mutually exclusive but can coexist and compliment one another (Uhlhaas et al., 2009, p. 1). In conclusion, when we analyze the functions of the nervous system that involve a uniﬁed consciousness and a self, we observe some extraordinary and biologically unique features. On the one hand, there is centralization of neural activity in which the highest levels of the hierarchy display more abstract and integrated responses patterns when compared to lower levels – a feature of non-nested hierarchies. However, at the same time lower order elements contributing to conscious experience are still represented in the totality of awareness in the fashion of a nested hierarchy most likely via synchronized oscillations that bridge higher order neurons across and within hierarchical levels. I wish to emphasize that as far as I have been able to determine, only the nervous system simultaneously operates in both a nested and non-nested fashion and in this regard appears to be unique among biological and non-biological systems. Furthermore, it appears that all conscious things have this structure and function, all things with this structure/function possess consciousness, and no things without this structure/function are conscious. Thus this feature may be necessary and perhaps sufﬁcient for the creation of consciousness and the self. Finally, an essential feature of the self is the role that maps, representations or modeling play in the construction of consciousness and the self. Neural maps that represent distorted but complete representations of the body’s surface are referred to as somatotopic maps. The spatial relations of visual ﬁeld are preserved in multiple visual regions in the brain that are referred to as retinotopic and the auditory system is organized in a tonotopic map such that neurons responsive to speciﬁc tones are in close neural proximity. These maps or representations could only be constructed via hierarchical organization since, for example in a somatotopic map, these stimuli contact the soma at widely separated physical locations on the body surface

14

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

and must converge upon neurons that are centrally integrated in order to preserve their original spatial relationship to the body. As outlined in Table 1, when all these aforementioned features are considered collectively we can see how nervous systems construct an integrated consciousness and ultimately what we call ‘‘self.” When viewed within the perspective of biological hierarchical organization, as the nervous system increases the number of its constituent and differentiated parts and at the same time achieves greater degrees of nested centralization and top down constraint, there appears to be a natural progression in complexity that leads from consciousness to self and ultimately to self awareness (Table 1). 4. Conclusions I have presented here a theory of the neurobiology of the self and consciousness that are embodied by systems that are necessarily hierarchical. This raises the question whether in fact the self and consciousness are ontologically necessarily both biological and hierarchical. I propose that for a system to possess consciousness and a self it must be engineered as an evolving hierarchical system (Salthe, 1985). I believe that given that all known instances of entities which possess a self conform to these constraints that this is indeed the case. Acknowledgments This work was supported in part from grants from The Gerald J. and Dorothy R. Friedman New York Foundation for Medical Research. The author thanks Jill K Gregory, CMI, FAMI Medical Illustrator, Multimedia Department, Continuum Health Partners for the illustrations in this article. References Ahl, V., & Allen, T. F. H. (1996). Hierarchy theory. New York: Columbia University Press. Allen, T. F. H., & Starr, T. B. (1982). Hierarchy. Perspectives for ecological complexity. Chicago: University of Chicago Press. Barlow, H. (1995). The neuron doctrine in perception. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 415–435). Cambridge: MIT Press. Beckers, G., & Zeki, S. (1995). The consequences of inactivating areas V1 and V5 on visual motion perception. Brain, 118, 49–60. Buzsáki, G. (2006). Rhythms of the brain. New York: Oxford. Cleeremans, A. (2003). The unity of consciousness: Binding integration, and dissociation. New York: Oxford. Craig, A. D. (2002). How do you feel? Interoception: The sense of the physiological condition of the human body. Nature Reviews Neuroscience, 3, 655–666. Craig, A. D. (2003a). Interoception: The sense of the physiological condition of the body. Current Opinion in Neurobiology, 13, 500–505. Craig, A. D. (2003b). Pain mechanisms: Labeled lines versus convergence in central processing. Annual Review of Neuroscience, 26, 1–30. Craig, A. D. (2003c). A new view of pain as a homeostatic emotion. Trends in Neurosciences, 26, 303–307. Craig, A. D. (2009). How do you feel-now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10, 59–69. Crick, F. (1994). The astonishing hypothesis. New York: Basic Books. Crick, F. H. C., & Koch, C. (1990). Towards a neurobiological theory of consciousness. Seminars in Neurosciences, 2, 263–275. Critchley, H. D., Wiens, S., Rothstein, P., Öhman, A., & Dolan, R. J. (2004). Neural systems supporting interoceptive awareness. Nature Neuroscience, 7, 189–195. Damasio, A. R. (1999). The feeling of what happens. Body and emotion in the making of consciousness. New York: Harcourt Brace & Company. Dennett, D. (1992). Consciousness explained. Boston: Little, Brown, and Company. Edelman, G. M. (1989). The remembered present: A biological theory of consciousness. New York: Basic Books. Engel, A. K., Fries, P., König, P., Brecht, M., & Singer, W. (1999). Temporal binding, binocular rivalry, and consciousness. Consciousness and Cognition, 8, 128–151. Engel, A. K., Fries, P., & Singer, W. (2001). Dynamic predictions: Oscillations and synchrony in top-down processing. Nature Reviews Neuroscience, 2, 704–716. Engel, A. K., & Singer, W. (2001). Temporal binding and the neural correlates of sensory awareness. Trends in Cognitive Sciences, 5, 16–25. Feinberg, T. E. (2000). The nested hierarchy of consciousness: A neurobiological solution to the problem of mental unity. Neurocase, 6, 75–81. Feinberg, T. E. (2001a). Altered egos: How the brain creates the self. New York: Oxford University Press. Feinberg, T. E. (2001b). Why the mind is not a radically emergent feature of the brain. Journal of Consciousness Studies, 8, 123–145. Feinberg, T.E.(2008). The nested hierarchy theory of consciousness (NHTC).Toward a Science of Consciousness, April 8-12, 2008. Consciousness Research Abstracts, 158. Feinberg, T. E. (2009). From axons to identity: Neurological explorations of the nature of the self. New York: W.W. Norton. Feinberg, T. E., & Keenan, J. P. (2005). Where in the brain is the self? Consciousness and Cognition, 14, 661–678. Fried, I., MacDonald, K. A., & Wilson, C. L. (1997). Single neuron activity in human hippocampus and amygdala during recognition of faces and objects. Neuron, 18, 753–765. Fries, P., Roelfsema, P. R., Engel, A. K., König, P., & Singer, W. (1997). Synchronization of oscillatory responses in visual cortex correlates with perception in interocular rivalry. Proceedings of the National Academy of Science, 94, 12699–12704. Gross, C. G. (2002). Genealogy of the ‘‘Grandmother Cell”. The Neuroscientist, 8, 512–518. Gross, C. G. (2008). Single neuron studies of inferior temporal cortex. Neuropsychologia, 46, 841–852. Gross, C. G., Bender, D. B., & Rocha-Miranda, C. E. (1969). Visual receptive ﬁelds of neurons in inferotemporal cortex of the monkey. Science, 166, 1303–1306. Gross, C. G., Rocha-Miranda, C. E., & Bender, D. B. (1972). Visual properties of neurons in inferotemporal cortex of the Macaque. Journal of Neurophysiology, 35, 96–111. Heimer, L., & Van Hoesen, G. W. (2006). The limbic lobe and its output channels: Implications for emotional functions and adaptive behavior. Neuroscience and Biobehavioral Reviews, 30, 126–147. Kreiman, G., Koch, C., & Fried, I. (2000). Category-speciﬁc visual responses of single neurons in the human medial temporal lobe. Nature Neuroscience, 3, 946–953. Llinas, R., & Ribary, U. (2001). Consciousness and the brain. The thalamocortical dialogue in health and disease. Annals of the New York Academy of Sciences, 929, 166–175. Melloni, L., Molina, C., Pena, M., Torres, D., Singer, W., & Rodriguez, E. (2007). Synchronization of neural activity across cortical areas correlates with conscious perception. The Journal of Neuroscience, 27, 2858–2865. Mesulam, M.-M. (2000). Principles of behavioral and cognitive neurology (2nd ed.). New York: Oxford University Press.

T.E. Feinberg / Consciousness and Cognition 20 (2011) 4–15

15

Mesulam, M. M., & Mufson, E. J. (1985). The insula of Reil in man and monkey. In A. Peters & E. G. Jones (Eds.), Cerebral cortex (pp. 179–226). NewYork: Plenum Press. Morgane, P. J., Galler, J. R., & Mokler, D. J. (2005). A review of systems and networks of the limbic forebrain/limbic midbrain. Progress in Neurobiology, 75, 143–160. Morgane, P. J., & Mokler, D. J. (2006). The limbic brain: Continuing resolution. Neuroscience and Biobehavioral Reviews, 30, 119–125. Nagel, T. (1974). What is it like to be a bat? Philosophical Review, 4, 435–450. Nauta, W. J. H. (1958). Hippocampal projections and related neural pathways to the midbrain in the cat. Brain, 81, 319–340. Nauta, W. J. H. (1973). Connections of the frontal lobe with the limbic system. In L. V. Laitinen & K. E. Livingston (Eds.), Surgical approaches in psychiatry (pp. 303–314). Lancaster: Medical and Technical Publishing. Nauta, W. J. H., & Haymaker, W. (1969). Hypothalamic nuclei and ﬁber connections. In W. Haymaker, E. Anderson, & W. J. H. Nauta (Eds.), The hypothalamus (pp. 136–209). Springﬁeld: Thomas. Nieuwenhuys, R. (1996). The greater limbic system, the emotional motor system and the brain. Progress in Brain Research, 107, 551–580. Nieuwenhuys, R. (1998). Overall functional subdivision of the mammalian brain. In R. Nieuwenhuys, H. J. ten Donkelaar, & C. Nicholson (Eds.). The central nervous system of vertebrates (Vol. 3, pp. 2023–2097). Berlin: Springer-Verlag. Nieuwenhuys, R., Veening, J. G., & van Domburg, P. (1988–1989). Core and paracores; some new chemoarchitectural entities in the mammalian neuraxis. Acta Morphologica Neerlando – Scandinavica, 26, 131–163. Nieuwenhuys, R., Voogd, J., & van Huijzen, C. (2007). The human central nervous system (4th ed.). New York: Springer. Northoff, G., & Bermpohl, F. (2004). Cortical midline structures and the self. Trends in Cognitive Sciences, 8, 102–107. Northoff, G., Feinberg, T.E., Panksepp, J. (in press). The ‘resting state hypothesis’ of major depressive disorder-translational subcortical–cortical integration. Neuroscience and BioBehavioral Reviews. 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, 440–457. Northoff, G., Schneider, F., Rotte, M., Matthiae, C., Tempelmann, C., Wiebking, C., et al (2009). Differential parametric modulation of self-relatedness and emotions in different brain regions. Human Brain Mapping, 30, 369–382. Pattee, H. H. (1973). The physical basis and origin of hierarchical control. In H. H. Pattee (Ed.), Hierarchy theory. The challenge of complex systems (pp. 73–108). New York: George Braziller, Inc. Quiroga, R. Q., Kreiman, G., Koch, C., & Fried, I. (2008). Sparse but not ‘grandmother-cell’ coding in the medial temporal lobe. Trends in Cognitive Sciences, 12, 87–91. Quiroga, R. Q., Reddy, L., Kreiman, G., Koch, C., & Fried, I. (2005). Invariant visual representation by single neurons in the human brain. Nature, 435, 1102–1107. Revonsuo, A. (2006). Inner presence: Consciousness as a biological phenomenon. Cambridge: MIT Press. Revonsuo, A. (2010). Consciousness: The science of subjectivity. Hove: Psychology Press. Roskies, A. L. (1999). The binding problem. Neuron, 24, 7–9. Salthe, S. N. (1985). Evolving hierarchical systems: Their structure and representation. New York: Columbia University Press. Schore, A. N. (2003). Affect regulation and the repair of the self. New York: W.W. Norton. Sewards, T. V., & Sewards, M. A. (2002). The medial pain system: Neural representations of the motivational aspect of pain. Brain Research Bulletin, 59, 163–180. Sherrington, C. (1947). The integrative action of the nervous system. New Haven: Yale University Press. Singer, W. (1999). Neuronal synchrony: A versatile code for the deﬁnition of relations? Neuron, 24, 49–65. Singer, W. (2001). Consciousness and the binding problemP. C. Marijuán (Ed.). Cajal and consciousness: Scientiﬁc approaches to consciousness on the centennial of Ramón y Cajal’s textura. Annals of New York Academy of Sciences, 929, 123–146. Singer, W., Engel, A. K., Kreiter, A. K., Munk, M. H. J., Neuenschwander, S., & Roelfsema, P. R. (1997). Neuronal assemblies: Necessity, signature and detectability. Trends in Cognitive Sciences, 1, 252–261. Uhlhaas, P. J., Pipa, G., Lima, B., Melloni, L., Neuenschwander, S., Nikolic, D., et al (2009). Neural synchrony in cortical networks: History, concept and current status. Frontiers in Integrative Neuroscience, 3, 1–19. von der Malsburg, C. (1995). Binding in models of perception and brain function. Current Opinion in Neurobiology, 5, 520–526. Yakovlev, P. I. (1972). A proposed deﬁnition of the limbic system. In C. H. Hockman (Ed.), Limbic system mechanisms and automatic function (pp. 241–283). Springﬁeld: Charles C. Thomas. Zeki, S. (1993). A vision of the brain. Oxford: Blackwell Scientiﬁc Publications. Zeki, S. (2008). The disunity of consciousness. Progress in Brain Research, 168, 267–268. Zeki, S., & Marini, L. (1998). Three cortical stages of colour processing in the human brain. Brain, 121, 1669–1685.

Consciousness and Cognition 20 (2011) 16–39

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Memory, autonoetic consciousness, and the self q Hans J. Markowitsch a,b,⇑, Angelica Staniloiu a a b

Physiological Psychology, University of Bielefeld, Bielefeld, Germany Alfried Krupp Institute for Advanced Study, Greifswald, Germany

a r t i c l e

i n f o

Article history: Available online 14 October 2010 Keywords: Dissociative amnesia Self-consciousness Emotion Episodic-autobiographical memory (EAM) Perspective taking Time

a b s t r a c t Memory is a general attribute of living species, whose diversiﬁcation reﬂects both evolutionary and developmental processes. Episodic-autobiographical memory (EAM) is regarded as the highest human ontogenetic achievement and as probably being uniquely human. EAM, autonoetic consciousness and the self are intimately linked, grounding, supporting and enriching each other’s development and cohesiveness. Their development is inﬂuenced by the socio-cultural–linguistic environment in which an individual grows up or lives. On the other hand, through language, textualization and social exchange, all three elements leak into the world and participate to the dynamic shaping and re-shaping of the cultural scaffolding of the self, mental time traveling and EAM formation. Deﬁcits in selfrelated processing, autonetic consciousness, emotional processing and mental time traveling can all lead to or co-occur with EAM disturbances, as we illustrate by ﬁndings from EAM impairments associated with neurological or psychiatric disorders. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Memory is a multi-facetted attribute of all animals and may even be found in rudimentary forms in the ﬂora and in so-called intelligent machines. This speaks for a million year-long process of memory evolution. More recent theories in psychology and the neurosciences have acknowledged this process by proposing different memory systems, especially in phylogenetically advanced species. The endowment with a developed body that has enabled the exploration of the environment over wide distances (embodiment; Pfeifer & Bongard, 2007) has most likely led to the diversiﬁcation of memories as well as the need to store information long term (cf. Campbell & Garcia, 2009). This is evident in species such as certain birds (Miyata, Gajdon, Huber, & Fujita, 2010; Weir, Chappell, & Kacelnik, 2002), whales and dolphins (Reiss & Marino, 2001), elephants (Plotnik, de Waal, & Reiss, 2006), the great apes (Bard, Todd, Bernier, Love, & Leavens, 2006; Call & Tomasello, 2008; Kitchen, Denton, & Brent, 1996), and New World capuchin monkeys (de Waal, Dindo, Freeman, & Hall, 2005). Furthermore, being a social animal and engaging in cooperative behavior (Brosnan & Bshary, 2010; de Waal & Suchak, 2010; Melis & Semmann, 2010) required and enabled a more ﬂexible application of mental capacities (Blakemore, 2010), though it is still debated whether and to what degree animals developed at least rudimentary abilities of foresight, prospection, and theory of mind (e.g., Gilbert & Wilson, 2007; Hare & Tomasello, 2005; Miyata et al., in press; Osvath, 2010; Roberts & Feeney, 2009; Suddendorf, Addis, & Corballis, 2009a; Suddendorf, Corballis, & Collier-Baker, 2009b). Similar to memory, basic self–non-self distinctions, such as the ones linked to physiological processes of immunity or digestion, are features of all viable species. The main unanswered question is the extent to which different species are capable of higher levels of conscious self-representations and self-awareness. Tulving (2005) has a clear position when stating ‘‘I argue that only human beings possess q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap.

⇑ Corresponding author. Address: Physiological Psychology, University of Bielefeld, P.O.B. 100131, D-33501 Bielefeld, Germany. Fax: +49 5211066049. E-mail address: [email protected] (H.J. Markowitsch). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.005

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

17

‘autonoetic’ episodic memory and the ability to mentally travel into the past and into the future, and that in that sense they are unique.” (p. 4). Though debated, his position has been supported by a number of ﬁndings also from human memory research (Botzung, Denkova, & Manning, 2008). Firstly, not all human beings possess the ability for mental time traveling: Patients with severe mental retardation or dementia may lack this capacity and patients with other disorders may show deﬁcits in integrating autobiographic memories with autonoetic consciousness and their selfhood. Examples are individuals with Asperger syndrome or autism. Tanweer, Rathbone, and Souchav (2010) found that adult Asperger individuals recalled in comparison to matched controls fewer events from their personal past and rated them much less speciﬁcally (responding more frequently to them as being just ‘known’, but not ‘remembered’ – a ﬁnding typical for patients with amnesia; see, e.g., Bengner & Malina, 2008; Hirano, Noguchi, Hosokowa, & Takayama, 2002; Noulhiane et al., 2008; Yonelinas, Kroll, Dobbins, Lazzara, & Knight, 1998). They also made in comparison to the controls fewer social identity statements and provided more abstract, trait-linked identities. Individuals with autism were found to have an atypical neural response pattern to judgments about their self, when their brains were studied with functional neuroimaging methods (Lombardo et al., 2010). The authors also detected ‘‘that the magnitude of neural self-other distinctions in the ventro medial prefrontal cortex was strongly related to the magnitude of early childhood social impairments” (p. 611). Secondly, developmental studies have emphasized socio-cultural–linguistic mechanisms that may be unique to the development of EAM. Small children – similar to animals – live initially in the here and now (Nelson, 2005a, 2005b). Their episodic-autobiographical memory (EAM) develops together with their self, theory of mind capacities, emotional conceptual knowledge and capacity for mental time traveling (Ghetti, DeMaster, Yonelinas, & Bunge, 2010; Rochat, 2010). Social context plays a critical role in the development of all the above neuro-cognitive functions. It was, for example, shown that simply listening to the voice from a tape recorder is not sufﬁcient for learning early aspects of language, but instead infants require the social presence of a person (Adolphs, 2010; Kuhl, Tsao, & Liu, 2003). Theory of mind functions furthermore ‘‘appear only after children have become experienced verbal communicators” (Surian, Caldi, & Sperber, 2007, p. 580). The onset of EAM and the ToM capacities in the offspring (Nelson & Fivush, 2004) depend on the degree of elaboration of the reminiscing style of their mothers. The importance of the social component for the emergence of ToM capacities is also reﬂected by ﬁndings that in institutionalized children ToM capacity correlated with the adult–child ratio (Bedny, Pascual-Leone, & Saxe, 2009). Developmental changes of EAM (e.g. pertaining to autonetic consciousness) extend however beyond childhood into early adolescent years (Picard, Reffuveille, Eustache, & Piolino, 2009) and may include in late childhood (ages 8–12 years) the ability to suppress memories (Paz-Alonso, Ghetti, Matlen, Anderson, & Bunge, 2009). On the brain level these EAM developmental changes in humans (from infancy to early adulthood) are reﬂected in the extensive structural and functional reorganization of different components of the neural networks supporting EAM, autonoetic consciousness and self-referential processing, ToM capacities and ability for emotional regulation (Shing et al., 2010). From a comparative cognitive-neuroscience perspective, frontopolar cortex (BA10) shows the biggest relative increase between great apes and human beings. BA10 activation recently has been found to be correlated with working memory capacity and general intelligence (Colom, Jung, & Haier, 2007). The basolateral nuclear group of amygdala shows a ‘‘progressive enlargement from insectivores to prosimians and ﬁnally simians” (Sarter & Markowitsch, 1985b, p. 348), while vice versa the centromedial nuclear group ‘‘shows a clear regression” along this phylogenetic scale. This is in line with ideas that in more phylogenetically evolved species the basolateral nuclear group expands to encompass higher cognitive–emotional functions such as EAM in the case of human beings (Cahill, Babinsky, Markowitsch, & McGaugh, 1995). One question which remains is concerned with the role of EAM in humans. Does indeed the EAM through its intrinsic feature of mental time traveling play a main function in the survival, as it has lately been emphasized repeatedly? And if the appearance of EAM is indeed adaptive, why did it not occur in other species? Is it in fact possible that certain claims of cognitive differences between humans and other species are the product of an underdeveloped experimental methodology rather than species differences per se? Or may it be the case that the main function of EAM is in fact social – a suggestion that was put forth by several authors, though it has not been an explicit focus of extensive experimental investigations yet (Markowitsch & Welzer, 2009; Welzer & Markowitsch, 2005). A hint in favor of this hypothesis comes from the work with patients with Alzheimer’s dementia: Fargeau et al. (in press) found that the social self was impaired earliest in this patient group. In the current paper, after a presentation of memory systems and their neural correlates, we will provide a review of the relationship between EAM, autonoetic consciousness and self, by preponderantly drawing on the socio-cultural–linguistic developmental model advanced by Nelson and Fivush (2004). We will then argue that EAM disturbances can result from deficits in the accurate re-collection of the encoding context, mental time traveling, emotional disturbances or self-related processing. By describing several EAM disturbances associated with both neurological and psychiatric diseases, we will demonstrate that many (especially severe) EAM impairments arise or exist in combination with dysfunctions in the realms of emotion, self, mental time traveling and social functioning. 2. Memory systems Memory is not unitary, but can be deconstructed along a time and content axis, respectively. Along the time axis, memory was traditionally divided into short-term and long-term memory. The short-term memory has a limited capacity of a few bits (4–7) (Cowan, 2000; Miller, 1956) and encompasses a time range of seconds to minutes. Any information that is not lost and exceeds the limited capacity of short-term memory is assigned to long-term memory stores. The above time-related

18

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

dichotomy of memory was later expanded by the addition of ‘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. Another time-related categorization of memory consists of the distinction between old and new and anterograde and retrograde memories, respectively. The compromised ability to access information that happened before the memory-impairing incident corresponds to retrograde memory impairment, while anterograde memory impairment refers to the compromised capacity to long-term acquire new information after the incident. A different facet of the relation between memory and time is concerned with the phenomenological experience of time in memory disturbances (inner time perception, mental time traveling). The classiﬁcation of long-term memory systems along the content dimension has undergone several changes, especially since Tulving had proposed a distinction between episodic and semantic memory in 1972. Presently two overlapping classiﬁcations dominate the memory research literature – one was initiated by Larry Squire and the other one was advanced by Endel Tulving. In Squire’s classiﬁcation, a main distinction is made between declarative and non-declarative memory. Under declarative memory, episodic and semantic memories – that is (biographical) events and general facts – are subsumed. Nondeclarative memory contains several other forms of memory, which are considered to be automatically processed. Tulving’s content-based classiﬁcation contains ﬁve long-term memory systems, which are considered to build-up on each other phylo- and ontogenetically. According to the SPI-model (SPI = serial, parallel, independent) that was proposed by Tulving (1995), it is assumed that information is encoded serially into these systems, may be stored in parallel in different systems and can be retrieved independently of the system in which encoding occurred. These ﬁve memory systems distinguish themselves by different levels of consciousness (such as autonoetic, noetic or anoetic) and distinct or partly distinct neural correlates. Wheeler, Stuss, and Tulving (1997, p. 335) deﬁned autonoetic consciousness as the capacity ‘‘that allows adult humans to mentally represent and to become aware of their protracted existence across subjective time”. They differentiated autonoetic from noetic consciousness (knowing) – which refers to the awareness of symbolic representations of the world, and from anoetic consciousness – that describes the simple awareness of external stimuli. The latter form of consciousness approximates the one evoked by the following citation of Mesulam (2000): ‘‘the existence of consciousness might be inferred when a living organism responds to environmental events in an adaptive way that is not entirely automatic (p. 93) (cf. Markowitsch, 2003). The hierarchy of the memory systems proposed by Tulving is depicted in Fig. 1. While excluding very basic forms of memories such as habituation, sensitization, classical conditioning, this hierarchy starts with procedural and priming memory systems – two simple memory systems that are still devoid of the need for conscious reﬂection upon the environment (‘‘anoetic”). Procedural memory is mainly motor-based, but includes also sensory and cognitive skills (‘‘routines”). Examples are riding a bike, skiing, playing piano, or reading words presented in a mirror-image. While procedural memory is largely an action-based memory system, the reverse is true for the priming system: Priming refers to a higher probability to identify stimuli, which were previously perceived in the same (perceptual priming) or a related way (conceptual priming). The ‘perceptual memory system’ acts ‘consciously’ (noetically), 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

Fig. 1. The ﬁve long-term memory systems and their assumed brain bases (for further description see the text).

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

19

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. Semantic memory, which was also termed ‘knowledge system’ or – by Tulving and Markowitsch (1998) – ‘declarative memory’, is context-free and refers to general facts. It is accompanied by noetic awareness. The deﬁnition of episodic memory has been submitted to several revisions over the years. While several decades ago the term episodic memory could be applied to laboratory stimuli with a speciﬁc embedding in time and place (Tulving, 1972), the eliciting of what and where and when information is no longer regarded as a sufﬁcient condition for fulﬁlling the requirement for being an episode memory (‘‘episodicity”). Nowadays episodic memory is deﬁned as the conjunction of subjective time, autonoetic consciousness and the experiencing self (Tulving, 2005) and subsequently the episodic memory system is currently viewed as being equivalent to the episodic-autobiographical memory system (EAM). Though the term autobiographical is still at times used interchangeably with the term episodic, not all the components of autobiographical memory, however, have an episodic quality. Therefore a distinction is emphasized between autobiographical-episodic memory and autobiographical-semantic memory. The latter refers to knowledge of name, date of birth or self-traits and may be preserved or updated in spite of blocked access to episodic memory for personal events. In comparison to autobiographical-semantic memory that requires noetic awareness only, the EAM presupposes a higher level of consciousness (‘‘autonoetic”). Autonoetic consciousness entails a ‘‘sense of self in time and the ability to relive subjective experiences from the encoding context by mentally travelling back in time” (Lemogne et al., 2006, p. 260). Given that most EAM reminiscences are affectively-laden, the reliving of the subjective experiences from the encoding context is usually intimately linked to an emotional evaluation of the signiﬁcance of these past experiences for oneself and with respect to one’s own position in his social and biological environment. This emotional evaluation may in turn shape someone’s motivation for planning for the future and engaging in future acts. This emphasizes that EAM has not only a ‘‘retrospective function”, such as ‘‘the conservation of certain conditions, their reproduction, and their localization in the past” (Ribot, 1882 p. 10), but also a prospective one. The classiﬁcations of Squire and Tulving, which were presented above, partly stem from opposing theoretical assumptions on the brain’s processing of episodic and semantic memory information. Squire’s theory emphasizes the commonalities between episodic and semantic memory processing, both from a behavioral and an anatomical perspective, while Tulving’s theory stresses the dissimilarities. Apart from these two categorizations, we currently witness a search for new models for memory systems, such as a model based on processing modes (Henke, 2010) rather than consciousness. This search may partly be prompted by the ‘‘grand challenge of consciousness” (Seth, 2010), in particular the challenge posed by designing a testing instrument (methodology) that provides an objective estimate (measure) of the subjective phenomenon of autonoetic consciousness or mental time traveling, which can be used in young children and non-human beings (Perner, Kloo, & Rohwer, 2010; Roberts & Feeney, 2009). 3. EAM and the brain The formation of stable long-term EAMs requires several information processing stages (encoding and consolidation). Debate about the process of consolidation still exists, with some authors arguing that the process may extend to years (Haist, Bowden Gore, & Mao, 2001). Once the information is consolidated, it is stored and then usually available for retrieval. Each retrieval is followed by re-encoding of the EAM in the newly present context. According to adherents to the reconsolidation theory, consolidated memories that are recalled by a reminder enter after retrieval a vulnerability (labilization) phase, during which they might become susceptible to disruption or strengthening or incorporation of new information (Forcato et al., 2010); this is followed by a process of stabilization (reconsolidation).The reconsolidation theory constitutes the basis for pharmacological studies in humans that aim to weaken the vivid and disturbing traumatic memories associated with certain forms of post-traumatic stress disorder by interfering with their assumed post-retrieval reconsolidation (Brunet et al., 2008). The incorporation of new information may have as the result a ‘‘re-contextualization” of the retrieved material (Modell, 2006) or the introduction of falsiﬁed details that may lead to false memories or confabulations (Loftus, 2000; Loftus & Hoffman, 1989; Loftus & Pickrell, 1995; Borsutzky, Fujiwara, Brand, & Markowitsch, 2008, 2010). Freud wrote about a re-transcription of memories ‘‘in accordance to fresh circumstances” (Masson, 1985, p. 207). And Bartlett (1932) and later Tulving (1995, 2002, 2005) and Schacter (2001) noted that human beings construct and reconstruct their personal memories, perhaps in an attempt 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). Edelman (1998) remarked that every act of memory is to some degree an act of imagination – a remark that later on was substantiated by ﬁndings that the hippocampal formation may also be involved in mental construction of complex scenes. Furthermore, Edelman wrote that ‘‘memory has the properties that allow perception to alter recall and recall to alter perception” (1998, cf. Modell, 2006, p. 37). This sentence hints to an important feature of EAM –namely its state-dependency. This feature implies that memories are optimally retrieved if the environmental and mood and physical state conditions match those during encoding. Inspired by Semon’s description (1904), Tulving (1983, 1985, 1995) coined this state-dependency of memory retrieval ‘‘ecphorizing”. Tulving (1983) employed the term ‘ecphory’ to describe the process by which retrieval cues interact with stored information so that an image or a representation of the information in question appears. A mismatch between encoding and retrieval conditions may lead to a spectrum of memory retrieval disturbances, ranging from common tip-of-the-tongue phenomena to complete pathological retrieval blockades (such as in dissociative amnesic conditions; see below).

20

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

The processes associated with the EAM‘s formation and retrieval engage a widespread network that contains several brain structures and long-distance ﬁber connections (Table 1). Some of these structures show a preferential specialization for mnemonic processing, while others (such as the amygdala) act as hubs that integrate emotive and cognitive functions. This supports the current theoretical approach to memory processing, which blends elements from both the previous localizationist and Gestalt-like theories. Encoding and consolidation of EAM depend on the integrity of brain networks that comprise several structures and their ﬁber connections, including medial temporal lobe, diencephalon and parts of prefrontal cortex. The role of hippocampal formation in the encoding and consolidation of EAM (and semantic memory) has been substantiated by an overwhelming number of ﬁndings from patients with brain damage as well as healthy people who underwent functional brain imaging investigations during test paradigms that tapped on EAM functions. As emotion is according to several authors intrinsic to EAM (see Bluck, Alea, Habermas, & Rubin, 2005), it is not surprising that both the Papez circuit and the basolateral limbic loop (mediodorsal nucleus of the thalamus, subcallosal area, amygdala and interconnecting ﬁbers) are engaged in the formation of EAM. The Papez and the basolateral limbic circuits constitute a group of bottleneck structures (Brand & Markowitsch, 2003) that are interconnected and of high relevance for the extraction of the emotional, biological and social signiﬁcance of newly incoming information. While prefrontal regions have also been implicated in the formation of EAM’s, they are nevertheless more strongly associated with the retrieval of personal events. Recruitment of limbic structures was also evidenced during the retrieval of EAM. Some limbic structures (such as the hippocampus) have been ascribed functions in binding speciﬁc details of past events, while others (such as the amygdala) have been opined to charge sensory information with appropriate emotional cues, in order to guide successful memory re-collection of emotionally signiﬁcant events. Several studies found that the combined activation of right-hemispheric fronto-temporal regions serves as trigger stations for retrieving stored EAM events (Brand & Markowitsch, 2008; Fink et al., 1996; Kroll, Markowitsch, Knight, & von Cramon, 1997; LaBar & Cabeza, 2006). The corre-

Table 1 Structures involved in EAM processes. Structure Telencephalon, cortical (principally cortical midline structures) Hippocampal formation Entorhinal, perirhinal, parahippocampal cortex Anterior cingulate cortex Posterior cingulate/retrosplenial cortex, supracommissural hippocampus Insula Telencephalon, subcortical Amygdaloid body Basal forebrain (including the septal nuclei, diagonal band of Broca, basal nucleus of Meynert, i.e., cholinergic nuclei) Claustrum Diencephalon (subcortical midline structures) Mediodorsal thalamic nucleus Anterior thalamic nucleus Nonspeciﬁc thalamic midline nuclei (e.g., paratenial nucleus) Mammillary nuclei

Functional implications (Episodic-autobiographical) memory, spatiotemporal integration, prospection, part of DMN (Semantic) memory Attention, ToM, awareness (VEN in its anterior area), emotion, reward, selfreference EAM, ToM, self-referential processing, part of DMN, imagination, familiarity Sensory-motivational integration, consciousness, reward, empathy, self processing, VEN Emotional-tagging of EAM, self-referential, imagining the future?, emotional awareness, reward and punishment EAM, inhibition, emotional evaluation, inner time perception, reward, social memory Support for conscious processing, sensory integration Encoding/consolidating EAM, consciousness, sleep, emotion, self-referential processing Emotional ﬂavoring, attention, support in encoding and consolididating EAM, consciousness? Consciousness, EAM? Encoding/consolidating of EAM, emotion?

Associated regions, especially of the expanded limbic system (paralimbic cortex) Orbitofrontal, medial prefrontal cortex EAM, self, ToM, prospection, part of DMN, time, monitoring veracity or feeling of rightness of a memory, episodic prospection of future rewards, emotional evaluation, social memory, accuracy of self-evaluation, temporal context, cortical midline structure Prefrontal cortex (esp. inferolateral and ventrolateral portions) Support in encoding and retrieving of EAM and semantic memory, ToM, part of mirror neuron system, consciousness, self, part of DMN, EAM retrieval, strategic search of memory, prospection Temporal pole Initiation of recall, memory-related sensory integration, ToM, self Temporo-parietal junction area Involved in EAM retrieval (recollective quality, i.e., vividness, conﬁdence), ToM, prospection, part of DMN Precuneus Imagination, ToM, self, EAM Lateral temporal cortex Memory storage (semantic and EAM), part of DMN Inferior parietal lobule Part of DMN, EAM, ToM, self?, mirror neuron system Abbreviations: DMN, default mode network; EAM, episodic-autobiographical memory; ToM, Theory of Mind; VEN, von Economo neurons. For explanations see text.

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

21

sponding regional complex in the left hemisphere seems to trigger the retrieval of semantic old memories (Markowitsch, Calabrese, Neufeld, Gehlen, & Durwen, 1999). The preponderant engagement of the right hemisphere in the retrieval of EAM has been linked to several factors, including the ascribed role of the right hemisphere in the processing of emotions and self-awareness (Calabrese et al., 1996; Feinberg & Keenan, 2005; Kaplan, Aziz-Zadeh, Uddin, & Iacobini, 2008; Keenan, Rubin, Racioppi, Johnson, & Barnacz, 2005; Keenan, Wheeler, Gallup, & Pascual-Leone, 2000; LaBar & Cabeza, 2006; Markowitsch, 1999; Schore, 2002). The right hemisphere was hypothesized to be preferentially involved in sympathetic arousal, such as the one that would result from negative feedback, while the left was preferentially linked to ‘‘parasympathetic quietude” (Allman et al., 2010; Wittling, 1997). Several authors have provided evidence for the involvement of the left hemisphere in parasympathetic inﬂuences on cardiovascular functioning, while the right hemisphere was shown to mediate the sympathetic regulation of cardiovascular activity (Foster, Drago, Ferguson, & Harrison, 2008; Wittling, 1997; Wittling, Block, Genzel, & Schweiger, 1998). Anomalies in resting heart rate have been studied in relationship to aggressive behavior and emotional processing in psychopathy and antisocial personality disorder and postulated to arise from a right hemispheric dysfunction or an altered inter-hemispheric connectivity (Raine, 2003; Raine et al., 2003). Alexithymic traits – consisting of a difﬁculty with identifying (especially negatively valenced) emotions and feelings, and distinguishing between feelings and the bodily sensations of emotional arousal – have also been linked to a dysfunction of the right hemisphere (Moriguchi et al., 2006; Schore, 2002) or an impaired interhemispheric transfer (Romei et al., 2008). They have been connected to a history of early trauma and a heightened susceptibility for dissociative disorders (Staniloiu, Markowitsch, & Brand, 2010). Interestingly, the right uncinate fascicle that links portions of the frontal and temporal lobe was found by a histopathological study to contain 33% more ﬁbers and be 27% larger in the right hemisphere than in the contralateral one (Highley, Walker, Esiri, Crow, & Harrison, 2002). The ventral portion of right uncinate fascicle is involved in ecphorizing affect-laden personal events. According to some authors, the uncinate fascicle may also contribute to the formation of memories (Sepulcre et al., 2008). The ventral branch of the uncinate fascicle is additionally a component of an emotional processing circuitry that connects the amygdala with the orbitofrontal cortex and the anterior cingulate cortex. As opposed to other brain ﬁber connections, the uncinate fascicle matures later and more slowly and may continue its development beyond the age of 30 years (Lebel, Walker, Leemans, Phillips, & Beaulieu, 2008). This may enable a higher structural plasticity in relationship to a variety of environmental inﬂuences, including physical or psychological stress-related insults. Microstructural abnormalities of the uncinate fascicle were for example reported in children, who were raised in a neglectful environment (Govindan, Behen, Helder, Makki, & Chugani, 2010). Levine et al. (1998, 2009) described a case of isolated dense retrograde EAM 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 uncinate fascicle. Despite normal performance on standard anterograde memory tests, the above-mentioned patient reported a feeling of disconnection from the post-accident autobiographical events. Subsequent reﬁned testing of his anterograde EAM revealed that he assigned signiﬁcantly less ‘‘remember”-ratings to his post-injury autobiographical events in comparison to normal subjects, suggesting an impaired ﬁrst-person autonoetic connection with his post-accident explicit memories of personal events. Fujie et al. (2008) reported both memory and emotional recognition impairments in amnestic MCI together with abnormalities of the uncinate fascicle. In agreement with the idea that white matter adjacent to cortical associations areas that matures later also undergoes an earlier age-related decline in myelination, Davis et al. (2009) recently found evidence via diffusion tensor tractography of a stronger age-related microstructural white matter change (demyelination) of the uncinate fascicle (especially the right one) in comparison to other ﬁber tracts (such as inferior longitudinal fascicle, cingulum, splenium). These results point to the importance of taking into consideration the integrity of fronto-temporal connections, when performing or analyzing studies that investigate the effects of aging on emotional processing or EAM. We demonstrated in several studies that the retrieval of emotionally-laden autobiographical events relies more on the right temporo-frontal region than on the left (Calabrese et al., 1996; Fink et al., 1996; Kroll et al., 1997; Markowitsch et al., 1993; Driessen et al., 2004). Botzung, Rubin, Miles, Cabeza, and LaBar (2010) also found a right-hemispheric amygdalar activation during the retrieval of highly emotional memories. In patients with dissociative amnesia – a condition that is accompanied by a failure of integration of emotive with memory processing functions – we found evidence for right hemispheric dysfunction, such as a hypometabolism of the right inferior lateral prefrontal cortex (Brand et al., 2009). Similarly, Tramoni et al. (2009) found subtle structural right-hemispheric prefrontal white matter abnormalities in a case of functional amnesia that occurred on a background of signiﬁcant psychological stressors. In a study that investigated the neural correlates of visual retrieval perspective in EAM, Eich et al. (2009) visualized an increase in the right posterior amygdala activity during the recall of ﬁeld (ﬁrst person perspective) memories, which was interpreted as a higher degree of subjective emotionality, associated with these memories in comparison to those that are retrieved from a third person (‘‘theatrical” or ‘‘disembodied”) perspective (Northoff et al., 2006). This argument echoes previous suggestions that ‘‘the boundary between self and not-self is one ‘s emotional attitude about an object or thought” (Barresi, 2002, quoted by Northoff et al., 2006, p. 453) and that self-referential processing in subcortical structures may in particular be intimately linked to emotional processing. Along with the emphasized role of the right hemisphere in several self-related tasks (such as self-face processing and selfregulatory control) (Keenan et al., 2000; Schore, 2002; Marsh et al., 2006), a study that compared the retrieval of autobiographical episodes with the retrieval of ﬁctitious ‘‘episodes” visualized a right-hemispheric activation of the amygdala only during the recall of personal – authentic – events, while the recall of ﬁctitious material activated the retrosplenial/precuneus

22

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

area only (Markowitsch, Thiel, et al., 2000). Another study that investigated the neural correlates of deception showed right prefrontal activation associated with rehearsed lies that were part of a coherent story (Ganis, Kosslyn, Stose, Thompson, & Yurgelun-Todd, 2003). Although the latter results could be subjected to different interpretations, one speculation is that through repeated rehearsal, triggered by motivated agendas, self-related deceptive material may become familiar to our ‘‘narrative” self to the point that the deceptive material is not only part of a story we tell others, but also part of a story we tell to ourselves and in some cases a story we may live by. The idea that the right prefrontal cortex might disproportionately contribute to processing the familiarity of information has been advanced by some authors (Dobbins, Foley, Schacter, & Wagner, 2002; Dobbins, Simons, & Schacter, 2004). Recently, Benoit, Gilbert, Volle, and Burgess (2010) showed that there is a relation to closeness and similarity of others to oneself, which leads to more or less activation of the medial rostral prefrontal cortex. In the same study self judgments were also associated with greater activation of right lateral rostral prefrontal cortex and stronger activation of the right insula. In addition to medial rostral prefrontal cortex a cluster in right lateral prefrontal cortex also showed less activity for contrasting self with other judgments to the extent that the other was perceived as more similar. Another pursued explanatory avenue for a preferential right hemispheric lateralization during the retrieval of explicit memories of personal events concerns the relationship between EAM and autonoetic consciousness. Keenan et al. wrote in 2005: ‘‘The evidence that there is a right-hemsipheric bias in terms of self-awareness is overwhelming.” (p. 700) and further, ‘‘that the right hemisphere is dominant for higher-order consciousness” (p. 702). In a recent article, Allman et al. (2010) reported that the right hemisphere of the human postnatal insular-frontal cortex contains signiﬁcantly more von Economo neurons (VENs) than the left. The VENs, which have been described in the insula and anterior cingular cortex (ACC), have been ascribed functions in consciousness, intuition, social cognition and regulation of appetite (Allman et al., 2010). Their degeneration was for example found to be associated with loss of emotional awareness, loss of empathy, altered self-consciousness and aberrant eating behaviors in patients with a behavioral variant of fronto-temporal dementia (Seeley, 2008). Especially the right fronto-insular cortex atrophy in these patients was found to correlate with the presence of core clinical symptoms. It was speculated that these symptoms arise from ‘‘failures to integrate visceral guidance cues with behaviour” (Seeley, 2008, p. 704), a speculation that is congruent with other authors’ idea that self-consciousness is in many ways intertwined to bodily consciousness and that unknowing and knowing consciousness are in fact two poles of a continuum (Vandekerckhove & Panksepp, 2009). A relationship between impairments of conscious access and integrity of large white matter bundles, particularly involving prefrontal cortex, was described by a new study (Reuter et al., 2008). Damage preferentially located to right parieto-occipital or right temporo-occipital cortex was also identiﬁed as a neurosubstrate of autoscopic phenomena (Blanke & Metzinger, 2009). While most of the above mentioned results hint to a preferential lateralization for the retrieval of EAM, self-awareness and other self-related tasks, other ﬁndings emphasize the modulation of this lateralization by a wide range of variables. In the case of EAM, these variables comprise gender, valence, nature of stimuli, and remoteness of memories, testing language (in case of bilinguals), visual retrieval perspective and genetic polymorphisms (Buchanan, Tranel, & Adolphs, 2005; Eustache et al., 2004; Markowitsch, 1998/1999; Markowitsch, Vandekerckhove, Lanfermann, & Russ, 2003; Piefke, Weiss, Zilles, Markowitsch, & Fink, 2003). Furthermore, it is increasingly acknowledged nowadays that the networks underpinning mnemonic processing, self-referential processing and consciousness, include structures that are located in both hemispheres. Especially recent studies that aimed to capture the neural correlates of EAM retrieval in settings that approximate the real-world ones (Botzung, LaBar, Kragel, Miles, & Rubin, 2010) have supported the view that EAM retrieval is basically organized bilaterally (Vandekerckhove et al., 2005). Powell, Macrae, Cloutier, Metcalfe, and Mitchell (2009) pointed to the self as a construct of collection of distinct mental operations distributed throughout the brain. They made the important distinction between the self as containing conceptual knowledge of one’s own personality traits and the self as an agent (e.g., freely choosing objects or watching passively as one is chosen). This second view of the self is also followed in the work of Vogeley (David et al., 2006; Vogeley et al., 2004). Northoff and Panksepp (2008) proposed that subcortical-cortical midline structures are engaged in self-related processing that supports a primal form of self-representation (a core self), which may exist in other mammalian species as well. Craig (2009) hypothesized that the insula supports a so-called sentient (feeling) self. The fronto-parietal human mirror system areas have been ascribed functions in the experiential understanding of others and were opined to be able ‘‘to effectively function as bridges between self and other, by co-opting a system for recognizing the actions of others to support self representation functions” (Uddin, Iacoboni, Lange, & Keenan, 2007, p. 154). Furthermore it has been pointed out that the human mirror neuron system shares connections with the default mode network (DMN). The DMN comprises several areas of the brain areas (ventral-medial prefrontal cortex, dorso-medial prefrontal cortex, posterior cingulate cortex, retrosplenial cortex, inferior parietal lobule, lateral temporal cortex and hippocampal formation) that display high baseline metabolic activity at rest and are considered to be important for mind-wandering, introspection, prospection and EAM processing. Though not recognized as a core structure of the DMN network, areas of the lateral prefrontal cortex have been reported to be recruited during EAM as well as self-referential processing (Spreng, Mar, & Kim, 2009). As relatively recently Northoff et al. (2006) remarked, ‘‘distinct concepts of self differ in the class of stimuli and their speciﬁc material or content reﬂecting what is called different domains”; ‘‘what remains unclear, however, is what unites these distinct concepts of self allowing us to speak of a self in all cases” (p. 440). Feinberg (2005) proposed that ‘‘we experience a uniﬁed and single self, as opposed to multiple selves from multiple hierarchical levels, because lower-order as well as higher-order elements are part of the nested hierarchy of the self” (p. 47). Pinker talked about ‘‘I” as being the ‘‘unity of selfness over time” (1997, p. 564). We argue that the experience of the unity of selfness or the personal sameness

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

23

over time (Locke, 1689/1964) is closely linked to EAM and autonetic consciousness, whose appearance and maturation enable us to override the awareness of discontinuity or multiplicity and lead us to attain and conserve a feeling of a cohesive personal identity over time and across contexts (Bromberg, 1994). 4. Memory development and the development of a self The development of memory systems starts with systems that involve simple, unconscious (anoetic) processing of information (procedural memory, priming), continues with conscious (noetic) systems and apparently culminates with the EAM system – which requires autonoetic consciousness (cf. Piolino, Desgranges, & Eustache, 2009). The EAM system is considered the last human ontogenetic acquisition, but also the highest achievement on the phylogenetic ladder (Tulving, 2005). The latter view is also reﬂected in the cautious employment of the term episodic-like memory to refer to the mnemonic abilities of species (e.g. scrub-jays) that show that they know – some scientists even use the expression recall – the what and where and how long ago of a salient event (Clayton & Dickinson, 1998; Roberts & Feeney, 2009). Despite increasingly sophisticated studies that have been carried out to examine animals’ ability for mental time traveling (Roberts & Feeney, 2009), there is at this point in time no conclusive evidential material to counteract the assertion of Tulving (2005, p. 9) that the EAM is ‘‘probably unique to humans”. ‘‘The present point is that mental time travel always occurs not only in subjective time but also in mental space, and that the mental space of the remembered past and imagined future may be different from the present space. Individuals with (autonoetic) episodic memory can, if the situation calls for it, think here and now about personal happenings in other places and other times. Therefore, if we ask whether other animals have episodic memory, we ask, among other things, whether they can also do so.” (Tulving, 2005, p. 7). Tulving (2000, 2002, 2005) assumed the existence of a hierarchy of memory systems which principally follows that of the memory systems listed in Fig. 1 from left to right. It is perhaps on the ﬁrst glance counterintuitive that memory for facts (semantic memory) is an earlier ontogenetic acquisition than EAM, given the common observation that the repetition of single similar episodes leads to their generalization and thereby ‘‘semantization” (Fig. 2). However, ﬁndings from human developmental studies emphasize that children ﬁrst acquire the semantic memory system and only thereafter their EAM neuro-cognitive capacities emerge (Nelson, 2003, 2005a, 2005b; Nelson & Fivush, 2004; Markowitsch & Welzer, 2009). When infants are asked to talk about what happened yesterday (‘‘What happened during lunch yesterday?”), they usually make very general – semantic – statements and do not provide vivid episodes. The emergence of EAM is dependent on or takes place in concert with the development of language and conceptual knowledge, higher levels of self and self-understanding, the ability to understand the feelings and intentions of others, executive functions, working memory, maternal reminiscing style, the capacity for mental time traveling and the maturation of the nervous system (Markowitsch & Welzer, 2009; Morrison & Conway, 2010; Nelson & Fivush, 2004; Picard et al., 2009). The relationship between self and episodic-autobiographical memory is powerful and dynamic. It can be discussed from different angles, such as the phenomenological momentanous experience of self accompanying the act of remembering of the ‘‘rememberer” or the relationship between EAMs and the creation and articulation of narratives that support outlasting self constructs or

Fig. 2. Relations between EAMs and semantic memories. It is assumed that most semantic memories are strengthened by repetition, while the reverse is true when one is repeatedly exposed to similar events.

24

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

self-concepts (moral, social, etc.). Similarly to the proposed hierarchical development of the memory systems, several authors put forth models for a hierarchy of self development. This may start with a very basic ‘‘proto-self’’ that is grounded in the sensory and motor domains (Panksepp, 1998), continue with a pre-linguistic, affective ‘‘core self” (Damasio, 1999; Northoff & Panksepp, 2008), be followed by a cognitive self (Howe & Courage, 1993) and thereafter by an ‘‘autobiographical” (Damasio, 1999) or ‘‘narrative self” (Gallagher, 2000) (Fig. 3). The appearance of a cognitive self late in the second year of life (Howe & Courage, 1993) seems to be a necessary, but not sufﬁcient condition for the EAM development. The EAM establishment entails in addition an enduring sense of self across time. As detected through the delayed self recognition paradigm of Povinelli and his colleagues (Povinelli, Landau, & Perilloux, 1996), an enduring sense of self is typically described in most 4 and 5 years old ones. Around age 4 children also typically succeed on a standard false belief task (Baron-Cohen, Leslie, & Frith, 1985) – a task that may require metarepresentational abilities (Perner, 2000). The autonoetic ‘‘experiencing self” or the ‘‘rememberer” features the capacity to ﬂexibly travel in mental time and space and a superior awareness of oneself as a person in a social (and biological) environment, with a past and a future. Several studies have concluded that EAM, autonoetic consciousness and mental time traveling appear between the age of 4 to 5 years approximately (Perner, Kloo, & Gornik, 2007; Perner & Ruffmann, 1995; Piolino et al., 2007). Morrison and Conway (2010) suggest that the formation of conceptual knowledge that is abstracted from details in early personal memories needs ﬁrst to be established, before speciﬁc EAMs can be retrieved. Childhood amnesia would therefore be related to the time until this ability is formed. (This corresponds to Tulving’s ideas of state-dependency and encoding speciﬁcity; Tulving, 2005; Tulving & Thompson, 1973). The development of EAM and mental time traveling extends however beyond the above mentioned age bracket. Piolino and co-workers investigated the establishment of EAMs and autonoetic consciousness in children up to the age of 11 years (Picard et al., 2009; Piolino et al., 2007). The authors proposed that mental time traveling may be one of the last features of the EAM to become fully operational. Picard et al. (2009) found that the EAM still improves in quality and quantity considerably over the ﬁrst ten years of life. Children provide more factlike information when requested to provide old episodes; only the recent episodes of older children are more detailed and authentic, while the remote ones remain mainly semantic. Probably, on the brain level, the establishment of inhibitory processes, acting from the late developing prefrontal cortex (Gibson, 1991; Huttenlocher, 1994) on the posterior (temporo-parietal) association cortex partly accounts for the developmental changes of self-related processing, EAM, and autonoetic consciousness. Findings from lying in a coherent way, where the right prefrontal cortex seems to play a pivotal role, add to this idea (Ganis et al., 2003). The reverse process seems to occur in older age, where autobiographical memories become more semantized (Piolino et al., 2006) (cf. Fig. 2). As children grow up they acquire greater gist memory that may however increase their susceptibility for false recall or recognition of information (Kühnel, Woermann, Mertens, & Markowitsch, 2008). The development of EAM may also enhance the ability of children to fantasize (to construct complex mental scenes) as well as to deceive. The capacity for deceiving has not only been linked to EAM (Ganis et al., 2003), but also to theory of mind and empathizing abilities as well as the capacity for emotional regulation (Adolphs, 2010). Profound changes of the sense of self take that take place from adolescence (Sebastian, Burnett, & Blakemore, 2008) to early adulthood are also reﬂected in the quality of EAM’s (Oddo et al., 2010). Beginning with early adolescence the individuals start to habitually incorporate the perspectives of others (e.g. peers) during the process of self-evaluation (J. H. Pfeifer,

Fig. 3. Sketch on possible relationships between states of mind and complexity and diversity of representation.

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

25

Dapretto, & Lieberman, 2010). Along this line it was found that the retrieval of recent memories in young women engages the ventro-medial prefrontal cortex, a region known to be recruited not only by self-evaluating tasks, but also by the evaluation of perceived similar others. The complex relationship between the development of self and EAM is captured by the model advanced by Nelson and Fivush (2004), which depicts the stages of EAM development from birth to age ﬁve (Fig. 4). The fact that the model had been conceived before Tulving’s last revision of EAM (2005) and therefore reﬂected Tulving’s deﬁnition of episodic memory from the last century (e.g., Tulving, 1995), in which the embeddedness of episodic memory in time and place was emphasized, does not diminish its powerful argumentation for the important contribution of the socio-cultural milieu to the emergence of EAM (Fig. 4). As Courage and Cowan (2009) pointed out, the concept of context implied by the ‘‘encoding speciﬁcity” principle (Tulving & Thompson, 1973) is viewed in the model of Nelson and Fivush (2004) more broadly to encompass family, society and culture (Fivush & Nelson, 2004; Nelson, 2007; Nelson & Fivush, 2004; Pope, Poliakoff, Parker, Boynes, & Hudson, 2007; Zhu, Zhang, Fan, & Han, 2007). The way in which the mothers structure their conversations when they engage in reminiscing with their offspring has a major impact on the onset of earlier EAM and the manner in which the offspring narrate their personal past. The maternal reminiscing style is, however, partly mediated by the nature of the mother–child attachment, with the consequence that insecure attachments may negatively impact on the quality of EAM’s. As Valentino, Toth, and Cicchetti (2009) showed, children with a history of abuse or neglect manifest a reduced EAM speciﬁcity (overgeneral memory effect). The relationship between the maternal reminiscing style and the development of the EAM of the offspring however extends beyond the familial context to encompass the cultural context (Harbus, 2010). Cultural differences were reported in EAM and were linked to childrearing practices, views of selfhood and past, preferences for high arousal versus low arousal emotional states or degree of attention to social context (Gutchess & Indeck, 2009; Nelson & Fivush, 2004). For example Asians’ EAM’s, in contrast to the ones of Caucasian Americans, were found to emphasize more social interactions and contain more people (Wang & Conway, 2004). Peterson, Wang, and Hou (2009) observed that Canadian children, who talked about re-collections of their EAMs showed a more autonomous self-construal style, while Chinese children showed a more relational one. As Adolphs (2010) pointed out, identifying the source of cultural difference requires the consideration of at least three potential variables: the ethnicity, the culture of origin and the cultural milieu where the testing is performed. Part of the ethnic differences may in fact reﬂect a regional distribution of a certain genetic polymorphism, which may bias the brain connectivity between structures involved in cognitive or emotive functions. The short (S) allele variant of the promoter region of the serotonin transporter gene (5HTTLPR) was for example found to be more prevalent in Japanese populations than in European ones. The S allele has been linked to alterations in microstructure of the uncinate fascicle, decreased gray matter volume in the amygdala and medial prefrontal cortex and altered functional connectivity between the amygdala and the prefrontal cortex (Caspi, Hariri, Holmes, Uher, & Mofﬁtt, 2010). In the EAM domains, the 5HTTLPR polymorphism has been hypothesized to modulate the retrieval perspective (Lemogne et al., 2009). The claim that one’s individual’s representations of self (independent versus interdependent self-construal) is inﬂuenced by the culture of origin has received support from functional neuroimaging studies. Zhu and colleagues (2007) compared subjects from an individualistic (Western) culture and subjects from a collectivistic (Chinese) culture during a task where they had to judge personal trait adjectives with respect to self, mother, or a public person. They found that the medial prefrontal and anterior cingulate areas were activated

Fig. 4. Hypothetical relations in developments from 1 to 5 years of age leading to the emergence of autobiographical memory. Larger arrows indicate more direct inﬂuences; double-headed arrows indicate reciprocal inﬂuences. Years (yr.) in the bottom scale indicate approximate ages when inﬂuences come into play on average in normal development. Areas above the center are presumed to be more endogenous and those below more exogenous as sources of development. Figure and ﬁgure legend copied from Fig. 1 of Nelson and Fivush (2004).

26

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

strongest during self-related judgments in subjects of both cultures. In contrast to subjects from Western cultures, in Chinese subjects the medial prefrontal cortex was also activated when making judgments about the mother. The authors concluded that Westerners use the medial prefrontal cortex exclusively for representing their own self. However, Benoit et al. (2010) recently showed that the degree of activation of the medial rostral prefrontal cortex depends on the proximity and similarity of others to oneself. Given the increasing economic globalization and the accompanying frequent migration or relocation, it has been emphasized that one should broaden the concept of culture-sensitivity in human cognition and consider the dynamic shaping and re-shaping, which is not only imposed by culture, but in feedback also by the individual (Han & Northoff, 2008; Vogeley & Roepstorff, 2009). Studies of bilingual or bicultural individuals and migrants suggest that the language and the degree acculturation (that is partly measured by language proﬁciency) can impact on memory performance, cognitive strategies, selfconcept and self-other distinctions. Bilingual Chinese children were found to recruit more elaborate and self-focused EAMs when speaking in English compared to when speaking Chinese (Wang, Shao, & Li, 2010). Kobayashi, Glover, and Temple (2008) assessed the response towards false-belief tasks (in the form of ‘x thinks that y thinks that . . .’) in early (children) and late (adult) bilingual Japanese. While they found medial prefrontal cortex activation in the children who had acquired the second language early, adults had more diverse brain activations, suggesting a more automatic processing in the adults and differences in activation dependent on the language used. Perner and Aichhorn (2008) conclude from the results of Kobayashi et al. ‘‘that the brain physiology . . . might be more liable to cultural and linguistic variation with intriguing links to similar developmental variations” (p. 125). Discussions about the role of language in shaping memory (and cognition in general) have a long history. Tulving (2005) asserted that language is not a necessary condition for EAM, but acknowledged that it supports and enriches its development, while others proposed a stronger connection between language and EAM (Nelson & Fivush, 2004; Suddendorf, Addis, et al., 2009). In the last few years it was found that a tribe of Amazon Indians – the Pirahã – lacks many of those attributes which are commonsense for our understanding of the selfhood and existence across time. They do not think of past and future and consequently cannot imagine the life of persons from history or past (Everett, 2005, 2008). The language spoken by the Pirahã reportedly only comprises two rudimentary temporal markers (Everett, 2005; Suddendorf, Addis, et al., 2009). These ﬁndings are relevant in the light of the proposal of Suddendorf, Addis, et al. (2009) that language and mental time traveling might have co-evolved. Language might have been preceded by the mimetic gesture. And mental time traveling might have been rooted in a form of embodied cognition, grounded on species-speciﬁc propensities for moving in a particular sense. In keeping with the general idea that episodic memory ‘‘grows out” of semantic memory (Tulving, 2005), the ability to ‘‘mentally travel in time” has been viewed as an extension of ‘‘mental travel in space.” The foundation of mental time traveling indeed may include a spatial component. This has been suggested from a comparative biological perspective, from an experimental perspective, and from the results of patients with vestibular nerve damage accompanied by selective hippocampal degeneration. We have argued for a functional shift in hippocampal function from a predominant processing of spatial cues (as in rodents) to a predominant temporal processing in higher primates (Markowitsch & Welzer, 2009; Tulving & Markowitsch, 1997, 1998). Miles, Karpinska, Lumsden, and Macrae (2010) asked subjects to daydream while viewing a display that elicited an illusion of self-motion (vection). Backward vection evoked thinking about the past, while forward vection resulted in preponderantly future-oriented thoughts. And for patients with hippocampal degeneration deﬁcits in virtual maze performance and spatial memory were reported, indicating that in addition to the more recent functional attributes of the hippocampal formation, some more ‘‘phylogenetically” ancient ones are still present in human beings (Brandt et al., 2005; Hüfner et al., 2007, in press). Another phylogenetically old brain structure, the cerebellum, has been found to play a signiﬁcant role in coordinating space–time relations (Oliveri et al., 2009). The capacity for mental time traveling was opined to offer a survival advantage and perhaps be the main answer for the existence in humans of an EAM system that is fragile and only offers an imperfect repository of the past. Laboratory studies, which investigated the neural correlates of constructive episodic future thinking, conﬁrmed the prospective function of EAM by ﬁnding common, but also distinct neural correlates involved in EAM retrieval and episodic future thinking (Addis, Pan, Vu, Laiser, & Schacter, 2009). However some authors questioned the main survival function of EAM, by emphasizing that autonetic self-awareness might bring with it the awareness of one’s own ﬁnitude – ‘‘a possibly fatal piece of knowledge, full understanding of which would preclude any motivation to survive” (Adolphs, 2010, p. 762). So how could these two views be reconciled? Roberts and Feeney depicted mental time traveling like a bicone and emphasized that typically our plans in the forthcoming weeks are more detailed and those for the years to come are more vague and overgeneralized. Consistent with the idea that the healthy individuals generally tend to expect more positive events than is rational to expect, a recent imaging study that related medial prefrontal cortex activity to self-evaluation advanced the hypothesis that the default network activity might be biased towards a positive self-evaluation, instead of an accurate one (Beer, 2007). A positivity memory bias was reported in elderly people, though the neural underpinnings of it are still under debate (Addis, Leclerc, Muscatell, & Kensinger, 2010). Furthermore it has been hypothesized that self-awareness could only have emerged in parallel with mechanisms such as faith and religion (Varki, 2009). Gilbert and Wilson (2007) pointed out to another aspect of future thinking. They remarked that people tend in fact to make several errors during simulation of future events, such as reliance on the most recent memories or the most salient ones. They also offered as a solution for the optimization of future simulation, namely getting advice. Suddendorf, Corballis, et al. (2009) commented on the authors’ suggestion, stating that it ‘‘resonates with our proposal that one adaptive function of language may be to allow us to improve our mental time travel by drawing on the descriptions others can offer of what the future may hold” (p. 1322). They subsequently proposed that the

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

27

main function of EAM may in fact be a social one. The possible contribution of EAM to social regulation was also implied by a recent study that showed that severe EAM deﬁcits interfere with the updating of moral character judgments of others and consequently may inﬂuence the way we perceive and behave towards others (Croft et al., 2010). 5. Disturbances of (autonoetic) consciousness and EAM in neurological diseases Similar to memory and self, consciousness is not regarded as a unitary phenomenon. Seth (2010) distinguishes between conscious level and conscious content. Conscious level is measured by tools such as Glasgow Coma Scale and ranges from brain death over coma to full alertness and awareness of one’s environment. Conscious contents differentiate themselves into ‘‘multimodal sensory contents related to the world; experiences of selfhood, volition and agency, and affective and somatic perceptions” (Seth, 2010, p. 1). Damasio (1999) proposed a core consciousness and an extended consciousness, respectively and, as mentioned above, Tulving repeatedly described three forms of consciousness: anoetic, noetic and autonoetic. Block (1995) further made the distinction between phenomenological consciousness and access consciousness. The quest for a neural correlate of level and content consciousness prompted studies of the phenomenal aspects of perception, sleep and wakefulness and vegetative and minimal conscious states as well as investigations of self- awareness (Newen & Vogeley, 2003) in healthy humans or clinical populations with different degrees of cognitive impairments or neurological or psychiatric diseases (Alkire, Hudetz, & Tononi, 2008; Coleman et al., 2009; Langnickel & Markowitsch, 2006; Markowitsch, 1995, 2000, 2003, 2005; Reinhold & Markowitsch, 2007; Staniloiu & Markowitsch, in press). Functional imaging and electrophysiological studies have attempted to disentangle the neural underpinnings of experiences of ownership and agency, selfface recognition or remembering. Theoretical and empirical work has also focused on understanding the interactions between unconscious forms of information processing and the conscious ones. Vandekerckhove and Panksepp proposed that higher forms of consciousness ‘‘are embedded in the ancient affective soil of anoetic consciousness” (2009, p. 1026). Under anesthetic conditions, unconsciousness occurs when the brain’s ability to integrate information is blocked (Alkire et al., 2008; Mhuircheartaigh et al., 2010); loss or reduction of consciousness was also observed with other various brain changes (e.g., damage, degeneration, hormonal changes). One question in consciousness research is concerned with the necessary and sufﬁcient processes that underpin healthy human consciousness, with some authors favoring the view that the thalamocortical system is the ‘‘seat of the relevant neural machinery” (Seth, 2010, p. 1; Tononi & Edelman, 1998). Autonoetic consciousness is generally regarded to depend on a well-functioning cerebral cortex (Markowitsch, 1995, 2005; Melloni et al., 2007; Uddin, Iacoboni, Lange, & Keenan, 2007). The functions in the spheres of EAM, autonoetic consciousness and self are, to a considerable degree, under the control of portions of the prefrontal cortex, in particular its ventromedial region (Beer, Lombardo, & Bhanji, 2010; Lee et al., 2010; Northoff et al., 2006; Passingham, Bengtsson, & Lau, 2010; Sebastian et al., 2008). However, functional imaging methods revealed that, in fact, the network is broader. There is also a more lateral prefrontal, right-hemispheric engagement (Keenan et al., 2000) and the anterior and posterior cingulate cortex, the precuneus, and the temporo-parietal junction area are engaged in addition to medial prefrontal regions (Buckner & Carroll, 2007; Plateck, Keenan, Gallup, & Mohamed, 2003; Vanhaudenhuyse et al., 2010). A reduction in consciousness together with an inability to form new autobiographical memories long-term may however accompany certain forms of diencephalic brain damage, in particular those affecting thalamic midline structures (Bogen, 1995; Bressler, 1995; Hart, 1995; Markowitsch, Cramon, & Schuri, 1993). We reported the case of a patient with above average intelligence, but severe amnesia due to an infarct that affected the thalamic mediodorsal nucleus bilaterally. After diencephalic damage this patient became totally anterogradely amnesic while his retrograde memory was partially preserved. The patient had a façade of normality when talking to him; he still displayed his premorbid jovial attitude in spite of his memory for new events being in the range of seconds only. On the other hand, the patient was able to acquire the skill of reading words, written in a mirror-image manner and he also demonstrated priming learning, as tested with an incomplete pictures test. He however showed a signiﬁcant impairment of his ability to reﬂect on his condition, which was attributed to the severity of his anterograde memory impairment. When he was asked open questions about his memory, he responded that it was ‘‘normal”. And when he was questioned in more focused manner (e.g. ‘‘Aren’t there areas where you have problems?”), he responded: ‘‘Yes, I quickly forget my dreams and I quickly forget jokes.” When he was asked about politics, he initially replied that he was not interested in politics. When the interviewer persisted with questioning, the patient made general remarks, such as ‘‘Well, you know, our continuing quarrels with France, you no longer want to hear that.” In addition to amnesia and disturbances of insight and autonoetic consciousness (Markowitsch, 2003) the patient showed a tendency to engage in confabulations. Confabulations – the production of ﬁctitious narratives – might arise from a combination of amnesia, impairments of autonetic consciousness and executive dysfunctions and perhaps impaired inner time perceptions (Borsutzky et al., 2008). They have been described in other diencephalic amnesic conditions (e.g. Korsakoff’s syndrome). Damage to the basal forebrain (such as the one associated with aneurysms of anterior communicating arteries) may also lead to an enhanced vulnerability to certain types of false memories (such as provoked confabulations and false recognition in procedural memory) (Borsutzky et al., 2010). The involvement of the thalamus in autonoetic consciousness and the self is also reﬂected by ﬁndings in other patients with bilateral thalamic infarcts, who manifested persisting childish behavior and symptoms suggestive of Ganser syndrome (Fukatsu, Fujii, Yamadori, Nagasawa, & Sakurai, 1997) and a disorientation with respect to time (e.g., time of the day or season of the year) (Kumral, Gulluoglu, & Dramali, 2007; Spiegel, 1982; Spiegel, Wycis, Orchinik, & Freed, 1956).

28

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

The relationship between time and memory and awareness has been emphasized since long and from very different perspectives. Time awareness (Brown, 1990), time orienting (Trivino, Correa, Arnedo, & Lupianez, 2010) and a proper handling of time epochs (Pribram & Tubbs, 1967) and temporal order (St. Jacques, Rubin, LaBar, & Cabeza, 2008) are related especially to the prefrontal cortex, while the proper perception of time and time epochs seems to engage the parietal cortex (Hoff & Pötzl, 1938; Pötzl, 1939, 1942, 1951, 1958) as well as the diencephalic structures as mentioned before (but see also Häfner, 1954). Damage to these areas may disturb the sense of time (including the the ability to successively link events in time) considerably with the consequence that the affected patients also become unable to encode or store new EAMs successfully. On the other hand, the capacity for episodic future thinking (self projection in the future) is an essential feature of EAM and autonoesis (Azry, Collette, Ionta, Fornari, & Blanke, 2009: Addis et al., 2009; Buckner & Carroll, 2007; Levine, Svoboda, Turner, Mandic, & Mackey, 2009; Perner et al., 2010; Schacter, Addis, & Buckner, 2007; Vogeley & Kupke, 2007; Zöllig et al., 2007). The ability to re-experience the past events and pre-experience personal future events might also be severely impaired after bilateral medial temporal lobe damage (Levine et al., 2009; Noulhiane et al., 2007; Piolino et al., 2004). The most wellknown and well-studied amnesic (‘‘hippocampal”) patient, H.M. reportedly made the following statement: ‘‘Every day is alone, whatever enjoyment I’ve had, and whatever sorrow I’ve had” (Milner, Corkin, & Teuber, 1968, p. 217). This statement suggests that H.M. knew the concept of time and that the life was a continuum. He showed at least partial insight into his memory problems and the ability to reﬂect on them. He furthermore possessed emotional knowledge and likely some capacity to experience feelings (in spite of having bilateral amygdalar damage!). One could speculate that he might have even still felt the lingering feelings associated with the personal experiences (events) from the previous hours, though he was unable to explicitly remember any of those experiences (Feinstein, Duff, & Tranel, 2010). Many other patients with bilateral medial temporal lobe pathology have been found to share similar impairments – an inability to encode new EAMs and new facts long-term. The ability of mental time traveling may be impaired both with respect to future and past in patients with major medial temporal lobe damage (e.g., due to encephalitis) and even in some patients with developmental amnesia (Andelman, Hooﬁen, Goldberg, Aizenstein, & Neufeld, 2010; Damasio, Eslinger, Damasio, Van Hoesen, & Cornell, 1985; Kwan, Carson, Addis, & Rosenbaum, 2010). While the above described patients suffered from focal, circumscribed brain damage in so-called bottleneck structures (Brand & Markowitsch, 2003) – another group of patients is characterized by pathologies accompanied by more widespread cortical damage, which might result in more global impairments of self-referential processing, autonoetic consciousness and mnemonic abilities. This group includes patients with degenerative forms of dementia (Banks & Weintraub, 2008; Greene, Hodges, & Baddeley, 1995; Gregory, Lough, Stone, Erzinclioglu, & Martin, 2002; O’Keeffe et al., 2007; Piolino, Belliard, Desgranges, Perron, & Eustache, 2003). 6. Disturbances of EAM, autonoetic consciousness and self in psychiatric disorders Complete loss of the EAM is rare in psychiatric disorders as well as neurological ones. It has been reported in psychogenic or dissociative amnesic conditions (see below). Less extensive or severe EAM disturbances have, however, been described in several other psychiatric conditions, such as schizophrenia, major depressive disorder, bipolar disorder, post-traumatic stress disorder or personality disorders (e.g. borderline personality disorder). In patients with schizophrenia studies reported impairments of EAM, theory of mind functions, self-referential processing, autonoetic consciousness and emotional processing (Corcoran & Frith, 2003; Satterthwaite et al., 2010). Certain positive symptoms of psychosis, such as paranoid delusions and hallucinations were linked to defects in mnemonic and emotional processing (Satterthwaite et al., 2010). One recent study showed, for example, that auditory verbal hallucinations were preceded by deactivation of the parahippocampus, suggesting that they may arise from spontaneous re-experiencing of memories (Diederen et al., 2010). Patients with active symptoms of major depressive disorder demonstrated abnormalities of self-referential processing, such as excessive self-focus (Grimm et al., 2009) and qualitative changes of the EAM’s (reduced speciﬁcity of details, increased retrieval from third person-perspective, changes in autonetic consciousness) in comparison to healthy people (Lemogne et al., 2006; Williams & Scott, 1988), which concerned preponderantly the memories of positively-valenced personal events (Lemogne et al., 2006). A third person retrieval perspective for positive EAM’s was found to persist in patients with major depressive disorder, even after clinical remission (Bergouignan et al., 2008). Patients with active symptoms of depression often display hopelessness and suicidal thoughts, which were hypothesized to be connected with their difﬁculties with imagining (pre-experiencing) positive personal future events (Sharot, Riccardi, Raio, & Phelps, 2007). These observations are consistent with studies that indicate that memory encoding and retrieval of negative versus positive material engage different brain networks (Markowitsch et al., 2003). In patients with post-traumatic stress disorder, the relationship between a history of trauma and impairments of EAMs in the form of reduced speciﬁcity of recalled past events was found to be moderated by the nature of trauma, the age at the onset of trauma (Stokes, Dritschel, & Bekerian, 2004; Valentino et al., 2009) and the cognitive style (Lemogne et al., 2008). In healthy young adults a relationship was observed between a diminution of EAM’s and a cognitive style characterized by avoidance of emotional material (Lemogne et al., 2008). Traumatic experiences have also been identiﬁed to trigger more global loss of EAM, such as in dissociative or psychogenic amnesic conditions. Psychogenic amnesic conditions have traditionally been viewed as episodes of amnesia, which were

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

29

preponderantly etiologically linked to psychological factors and occurred in the absence of signiﬁcant brain damage (as detected by conventional structural imaging techniques). Most cases of psychogenic amnesia were reported to have a retrograde nature. In a few instances anterograde (Markowitsch, Kessler, Kalbe, & Herholz, 1999; Smith et al., 2010; Staniloiu, Markowitsch, et al., 2010) instead of retrograde amnesia (or a combination of the two) has been documented (Markowitsch, Kessler, Van der Ven, Weber-Luxenburger, & Heiss, 1998). The interest in psychogenic amnesia has ﬂuctuated over the years, dating back to around 140 years ago (Markowitsch, 1992). The etiological underpinnings of this disorder have from the beginning been a source of debate. This is also reﬂected in the various terms, which have over the years been used to capture this condition, such as ‘epileptic somnolence‘(Burgl, 1900), ‘hysteria’ (Breuer & Freud, 1895; Janet, 1907), ‘hysterical state of somnolence’, ‘multiple personality (disorder)’ (Sidis & Goodhart, 1905), dissociative, functional or medically unexplained amnesia or mnestic block syndrome (Markowitsch, 2002). The subsuming of this condition under the concept of ‘‘hysteria” was reinforced by the strong tradition of French psychiatry (Pierre Janet, Jean-Marie Charcot), which was followed by Sigmund Freud, a pupil of Charcot. In the current main ofﬁcial classiﬁcations of diseases (DSM-IV-TR, 2000 and ICD-10, 1992) earlier diagnostic designations of hysterical or psychogenic amnesia are now preponderantly incorporated by the diagnostic categories of dissociative disorders in DSM-IV-TR and dissociative (conversion) disorders in ICD-10, but also by other diagnostic subcategories such as somatization disorder (in DSM-IV-TR only), post-traumatic stress disorder (PTSD) and acute stress disorder (DSMIV-TR and ICD-10) or certain personality disorders (such as borderline personality disorder). However, the term ‘hysteria’ is still in use (Stone, Smyth, Carson, Warlow, & Sharpe, 2006; Thomas-Anterion, Guedj, Decousus, & Laurent, 2010). There are authors who still debate the pathogenetic mechanisms or the legitimacy of the diagnostic entity of dissociative amnesia; Pope et al. (2007) consider it to be culture-bound and as having arisen in the 19th century. Though culture deﬁnitely plays a role in molding perceptions of trauma, selfhood, memory, future and past, there is nowadays signiﬁcant evidence for the contribution of psychological stress or trauma to the genesis of dissociative disorders across a variety of cultures (Staniloiu, Borsutzky, & Markowitsch, 2010). A large body of data has linked psychological stress (in the form of acute massive stress or chronic stress) to a variety of psychiatric and non-psychiatric conditions (Staniloiu, Markowitsch, et al., 2010). Although psychological stress or trauma is necessary for the emergence of dissociative amnesic disorders, it is not a sufﬁcient condition, which in fact has been suggested since Janet’s time. Genetic dispositions, epigenetic mechanisms, gender, personality traits and the nature, severity, recurrence and the age of onset of trauma all modulate the impact of psychological stress on the structural and functional integrity of the brain network subserving mnemonic processes (see below). In line with Janet’s view of the mechanism of dissociation as ‘‘an inability of the personal self to bind together the various mental components in an integrated whole under its control” (Janet, 1907, p. 23), dissociative disorders are deﬁned in DSMIV-TR (2000) as disturbances of the integrated organization of memory, perception, consciousness and identity. Among dissociative disorders, amnesia is a prominent symptom of dissociative amnesia, dissociative fugue, dissociative identity disorder (multiple personality disorder) and dissociative trance disorder (possession trance). Apart from dissociation, other psychological mechanisms, such as hyper-suppression or cognitive avoidance (Fujiwara et al., 2008; Lemogne et al., 2008; Tramoni et al., 2009) have been identiﬁed to underlie at times concurrently amnesic conditions, which are etiologically linked to psychological factors. These ﬁndings partly explain why some authors still favor the use of the term psychogenic amnesia over the term dissociative (McKay & Kopelman, 2009). Dissociative amnesia has as primary symptom the ‘‘loss” of memory for autobiographical events, especially those of a traumatic nature. This symptom should not be better explained by other somatic diseases (such as closed head injury), normative forgetfulness or other psychiatric disturbances. Psychologically motivated feigning of memory disturbances (factitious disorder) and malingering have to be ruled out (Jenkins, Kapur, & Kopelman, 2009). This is however, not always an easy task as some cases of dissociative amnesia can occur on a forensic background (Markowitsch, 1992, 2010; Markowitsch, Calabrese, et al., 1997). On the other hand the most commonly malingered cognitive impairment by individuals who seek ﬁnancial compensation after an accident is memory impairment (Serra, Fadda, Buccione, Caltagirone, & Carlesimo, 2007). A variant of dissociative amnesia is the dissociative fugue. In this condition retrograde amnesia for personal events is accompanied by suddenly leaving the customary environment – home, city and workplace – and compromised knowledge about personal identity. A failure of integration of self-referential, emotive and cognitive processes can however be observed not only in patients with dissociative amnesic conditions, but also in patients with related symptomatologies, such conversion paralysis, where body parts (most frequently the extremities) become paralyzed in the context of major psychological stress (Burgmer et al., 2006) or body identity integrity disorders (where individuals experience a strong wish for amputation of one of their limbs or for becoming paralyzed) (Oddo et al., 2009). A relevant example of a case of dissociative fugue is the one of a 37-year-old family father, who developed an episode of retrograde amnesia and loss of personal identity, seemingly out the blue (Markowitsch, Fink, Thöne, Kessler, & Heiss, 1997). One morning he took his bike, with the plan to go to the nearby bakery to buy roles for breakfast. However, instead of returning for breakfast, he apparently continued to ride his bike for the next couple of days from north to south Germany until he reached the suburbs of a major city. There he met a few people who suggested to him to continue to ride to the central railway station. After doing so, he met a female ofﬁcer from the Salvation Army, who advised him to go the nearby university psychiatric clinic. He followed her advice, went to the hospital and got admitted to the psychiatric ward, where he was diagnosed with psychogenic (dissociative) fugue. In the hospital the patient made friends and appeared not to care about having no access to his past – a presentation which has been noted in other patients with dissociative amnesia and was termed la belle indifférence by Janet (1907). The description of this case points so far to several similarities with other case-reports from literature. It is not uncommon for patients with dissociative fugue to present themselves to the emergency services

30

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

of hospitals or the police stations or wander in railway stations (Kluft, 1988; Kritchevsky, Chang, & Squire, 2004; Loewenstein, 1991) and to be susceptible to inﬂuences (advice) from others (Reinhold & Markowitsch, 2009; Stone et al., 2006). In fact many patients with dissociative disorders are highly hypnotizable on formal testing (Maldonado & Spiegel, 2008). The patients’ lack of concern about their condition may bear a close link to the impairment of their capacity for mental time traveling, which presupposes an emotional engagement with both past and future (Azry et al., 2009). Later on, the patient described having had a strong need (compulsion) to continue to ride during his fugue episode, which in fact he did for a couple of days. His state of altered (autonetic) consciousness apparently did not interfere with his automatic performance (‘‘ambulatory automatism”) (Loewenstein, 1991). In fact there are reports that some dissociative states, which are characterized by suspension of critical thought, may even enhance performance of coordinated and complex motor acts, such as in athletes (Maldonado & Spiegel, 2008). The patient also described that during his ride, he at times looked at himself in the windows of shops in the cities, but his face did not look familiar to him. This again suggests an impairment of self-referential processing (self-face processing) and self-awareness, perhaps rooted in a dysfunction of the right hemisphere (see below). We initially tested the patient neuropsychologically. He demonstrated normal anterograde memory; however, he showed no evidence of any retrograde EAM. He later relearned his past, including his autobiographical-semantic knowledge and then could retrieve portions of his past in a neutral, affectless manner. Other cognitive functions were largely spared or regained within weeks. We investigated the patient with functional brain imaging, using a paradigm in which he was confronted with events from his autobiographical past. While the normal subjects activated in the water-PET study predominantly the fronto-temporal region of the right hemisphere (Fink et al., 1996; Kroll, Markowitsch, Knight, & von Cramon, 1997; Markowitsch et al., 1993), the patient activated largely the left hemisphere, which is considered to be relevant for the retrieval of neutral facts or knowledge (Grossi, Trojano, Grasso, & Orsini, 1988; Markowitsch, Calabrese, et al., 1999). As mentioned above, the right temporal-frontal connections are important for ecphorizing the affect-laden personal events. Therefore those imaging ﬁndings suggested that the patient’s access to his past occurred via his semantic memory system. As it was also transparent from his behavior, the patient lacked that ﬁrst-person autonoetic connection and feeling of warmth that is characteristic to EAM retrieval. Interestingly, the patient‘s personality underwent changes as well. While he had been an avid car driver prior to his ‘‘escape” (fugue), afterwards he avoided car driving or riding a bike and said that cars were too fast for human beings. His relinquishing of a previously rewarding behavior (car driving) occurred in the absence of evidence of impairment in his procedural skills. Furthermore, the patient changed his food preferences and gained considerable weight. Interestingly, he ceased having asthma attacks or allergic reactions, conditions which may be viewed as partly having a psychosomatic basis. Alterations in previously rewarding activities (such as eating, smoking, cooking, drinking alcohol) and personality after the onset of psychogenic (dissociative) amnesia have been reported by several other authors (Fujiwara et al., 2008; Tramoni et al., 2009). They may reﬂect the fact that several structures that are involved in EAM processing play also a role in reward and are sensitive to stress (Ulrich-Lai & Herman, 2009). Furthermore, they also converge with a recent view of self as having connections to reward (Enzi, de Greck, Prösch, Tempelmann, & Northoff, 2009). Contrary to old teachers’ lore, confrontation with his wife did not result in dissolving the patient’s amnesia. Instead he even thought that the hospital personal wanted to couple him with an unknown woman. He, however, moved back to the family house. Upon arrival, he complained about the existing furniture and tapestry. Nevertheless he pretty much followed all suggestions or directions of his wife instantaneously, as he had done before the onset of the amnesia. A review of the patient’s developmental history revealed that he had been raised by parents who abused alcohol and fought a lot against each other. The patient was dressed as a girl until starting elementary school because his mother reportedly wanted to have a daughter instead of a son. Later his mother frequently criticized him for being weak and behaving in a girlish manner. She predicted that he would never become successful in life and instead would ruin their business. The patient grew up with very low self-esteem. In his early 20ies, he married a woman who apparently behaved similarly to his mother and possessed the self-conﬁdence he was missing. Being not very successful with their business, the couple always suffered from a lack of money. Nevertheless, the patient’s wife reportedly decided one day that they should leave by car for a longer vacation. Though the patient knew that they would not have enough ﬁnancial resources for the suggested trip, he did not object. Three days prior to the date of the intended holidays, he, however started riding his bike along the river Rhine. This case paradigmatically reﬂects the disruption (disintegration) of the cohesiveness between autonoetic consciousness, EAM and the self in the dissociative amnesic condition: The patient had lost ﬁrst-person autonoetic access to the personal past of his entire life, while being able to retrieve impersonal semantic facts. He could, however still read, write and calculate. He displayed a blunted affective disposition and did not seem to care about his future life (la belle indifference; Reinhold & Markowitsch, 2009; Stone et al., 2006). Though his knowledge about how to behave in social situations seemed largely preserved, there were several indications that his amnesia might have impacted on his ability to judge the feelings of close others, regulate his social behavior and ﬂexibly adapt it to the new circumstances. He for example became closer and friendlier with comrades from the psychiatric ward than with his family, which resulted in signiﬁcant interpersonal difﬁculties with his close family members. The patient’s presentation is similar to the clinical descriptions of other patients with dissociative amnesic conditions, though there are a number of variants of these conditions. One particular variant is the Ganser syndrome, which at times had been described in association with dissociative fugue conditions. The syndrome has been submitted to several diagnostic revisions and debates over the years. In comparison to previous DSM editions, where Ganser syndrome was presented as a Factitious Disorder, Ganser syndrome is currently included under the category of Dissociative Disorders Not Otherwise

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

31

Speciﬁed in DSM-IV-TR and it is simply deﬁned by giving approximate answers to questions (vorbeireden). Ganser’s (Ganser, 1898, 1904, 1965; cf. Feinstein & Hattersley, 1988) original description of the syndrome was, however, much broader than the current DSM-IV-TR one. It included a hysterical semitrance or twilight state, characterized by a tendency to give approximate answers, impairments of consciousness, amnesia and hallucinations, being more consistent with the later views of this disorder as a brief reactive psychosis to stress. In 1904 Ganser wrote on page 34 (our translation): ‘‘The most prominent phenomenon was that the patients were unable to respond properly to questions of the most simple kind, they were given, though they indicated through their mode of answering that they principally had captured the sense of the questions, and that they revealed in their answers an astonishing lack of knowledge for facts they deﬁnitely had possessed or were still possessing. ... After days, and in some cases after repeated remissions and intermissions, a total clarity with recurrence of normal knowledge appeared, whereby for the period of the somnolence a gap of reminiscence remained.” Though initially linked to forensic background, Ganser syndrome was also reported in non- forensic contexts. The syndrome was found to affect preponderantly young men with a mean age of 35 years, although there were some case-reports in women and children as well. Both a transient, limited course and a chronic course have been recorded. We investigated two male patients with this syndrome, one with a clear forensic background and one without (Staniloiu et al., 2009). Both patients showed a global deterioration of their intellectual capacities suggestive of a pseudo- dementic picture. The patients became lethargic, lost interest and only reacted when being directly addressed and otherwise relied on the guidance provided by their wives. They lost joy, did not attend to their children and other stimuli of the environment and appeared unable to understand and to react appropriately to even simple commands. The patients appeared to be very insecure. Even very simple questions, for instance, pertaining to their date of birth were not directly answered (‘‘I have to look in my passport” was, for example, a response to questioning about the date of birth.) Their clinical presentation followed a chronic course, despite treatment. This is congruent with the reports of other authors who evidenced a prolonged course of memory deﬁcits in a substantial number of dissociative amnesic conditions (Kritchevsky et al., 2004) and is in contrast with ﬁndings from older studies, that described that memory recovered in most patients within a month from the onset of psychogenic memory loss (Kanzer, 1939). Although in both cases of Ganser syndrome the structural brain imaging investigations were not indicative of organic impairment, in the patient where additional functional imaging was performed, a global signiﬁcant reduction in the brain metabolism was visualized. The mentioned functional imaging result and the chronic course of the above mentioned two patients are relevant, in the light of ﬁndings from previous studies. Some of those pointed to a possible organic basis to the psychiatric presentation of the Ganser syndrome that can especially become apparent over time. For example, Ladowsky-Brooks and Fischer (2003) described a patient, who presented with features of Ganser syndrome and severe cognitive deﬁcits in other domains. However, the individual’s cognitive decline over a period of a year, in combination with ﬁndings from functional imaging, resulted in a diagnosis of fronto-temporal dementia. The connection between stress-related cognitive decline, the potential contribution of glucocorticoids to accelerated aging, fronto-temporal dysfunction, and the later development of dementia was already pointed out by Porter and Landﬁeld (1998). Recent studies have further conﬁrmed a relationship between different forms of stress-related psychopathologies (such as major depressive disorder or post-traumatic stress disorder) and later risk for the development of dementia (e.g. Yaffe et al., 2010). We have studied several dozens of patients with a condition of long-lasting dissociative amnesia. All of them encountered major negative life events that had occurred mostly in early childhood, but in part continued or recurred in their later life (e.g., Markowitsch, Fink, et al., 1997; Markowitsch et al., 1998, 2000; Markowitsch, Kessler, Russ, et al., 1999; Fujiwara et al., 2008; Reinhold & Markowitsch, 2007). In some of these cases there was an additional background of immigration (Fujiwara et al., 2008; Staniloiu, Borsutzky, et al., 2010; Staniloiu, Markowitsch, et al., 2010). Similar cases with a (likely) background of immigration were reported by others (Thomas-Anterion et al., 2010). In our pathogenetic model of dissociative amnesic conditions, we propose that stressful events lead to the release of a variety of stress hormones and initiate a neurotoxicity cascade (Joels & Baram, 2010; Markowitsch, 2000; O’Brien, 1997). We assume that the release of stress hormones – in particular glucocorticoids – is principally responsible for the block in the function of brain structures that are essential for the encoding or retrieval of information. In fact, it is known that several key brain areas involved in memory processing show especially during their periods of development or regression (Lupien, McEwen, Gunnar, & Heim, 2009) an increased susceptibility to the hormonal effects of stress and that enduring alterations in stress hormone responses in response to environment (e.g., early life experiences) might occur via epigenetic mechanisms, which act during windows of heightened vulnerability (McGowan et al., 2009; Szyf, 2007). The effects of stress and environment (Brand & Markowitsch, 2009) on mental health might be clinically observable after a lag period. This may be accounted for by the fact that the early stress effects on synaptic organization and brain structure and function only may become obvious after the completion of synaptic organization (Lupien et al., 2009) We (Markowitsch et al., 1998, 2000; Reinhold, Kühnel, Brand, & Markowitsch, 2006) and others (Thomas-Anterion et al., 2010; Tramoni et al., 2009) have found evidence for stress-induced changes of the brain network engaged in mnemonic processes in dissociative amnesic conditions by employing different functional brain imaging methods and structural imaging techniques specialized for estimating white matter integrity. Via these methods we provided evidence for a desynchronization between regions of the brain that are involved in the processing of factual information and regions of the brain that are involved in the processing of emotion, self-awareness and EAM (limbic system, in particular the amygdala and the septal region/basal forebrain). This is consistent with knowledge that receptors for stress hormones exist in the regions of the hippocampal formation and the amygdala – structures which synchronize facts and emotions and are therefore crucial for EAM

32

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

(de Kloet, Oiztl, & Joels, 1999; Ellen, Kaut, & Lord, 2009; Joels & Baram, 2010). As both brain imaging results from patients with dissociative amnesic conditions (summarized in Reinhold et al. (2006)) and theoretical accounts point to, the expanded limbic system (Nauta, 1979; Nieuwenhuys, 1996), including the frequently mentioned medial prefrontal/orbitofrontal cortex and the insula (e.g., Ganis, Morris, & Kosslyn, 2009; Modinos, Ormel, & Aleman, 2009) constitutes a structural network for autonoetic consciousness, EAM, and the self. Clinical and imaging ﬁndings from patients with dissociative amnesic conditions re-emphasize the importance of a proper emotional charging for EAMs (Markowitsch & Staniloiu, in preparation-a). There is evidence from many sites that EAM and emotions are strongly interdependent (Sarter & Markowitsch, 1985a). Especially patients with bilateral amygdala damage demonstrate that they are impaired in properly encoding EAM (Cahill et al., 1995; Markowitsch et al., 1994; Siebert, Markowitsch, & Bartel, 2003; see Murty, Ritchey, Adcock, and LaBar (in press) and Markowitsch and Staniloiu (in preparation-b), for reviews). But recent work also shows that there are several interdependencies between different forms of emotions, activation of brain regions and awareness and interpretation of the environmental signals (Amting, Greening, & Mitchell, 2010). 7. Conclusions The self, autonoetic consciousness and EAM are intimately interlocked. The relationship between them evokes the reference to the threefold cord that is not easily broken (Ecclesiastes 4:12). The three ‘‘cords” appear and evolve both ontogenetically and phylogenetically in strong connection with each other, supporting and enriching each other’s development. Their appearance and development take place in concert with other cognitive and emotive functions – the ability of perspective taking, executive functions, language, the feeling of empathy, the ability to reﬂect on oneself, solving abilities with divert thinking, etc. (e.g., Sutin & Robins, 2008). Given the multi-functionality of EAM and the multiple facets of self (Klein & Gangi, 2010; Neisser, 1988) and consciousness, the relationship between EAM, autonoetic consciousness and self can be approached from multiple perspectives. Neisser (1988) distinguished the ecological, the interpersonal, the conceptual, the remembered, and the private self. Klein and Gangi (2010) wrote about the ‘‘multiplicity of the self” and pointed to the composition of the self of EAMs, of semantic representations of one’s personality traits, of semantic knowledge of facts about one’s life, an experience of continuity through time, a sense of personal agency and ownership, the ability to self-reﬂect, and the physical self. Along the same vein Kaplan et al. (2008) had pointed to the multimodality of the self. Similar to Vandekerckhove and Panksepp (2009), Davis (1996) had described a ﬂow between multiple levels of consciousness: ‘‘Not one unconscious, not the unconscious, but multiple levels of consciousness and unconsciousness, in an ongoing state of interactive articulation as past experience infuses the present and present experience evokes state-dependent memories of formative interactive representations. Not an onion, which must be carefully peeled, or an archeological site to be meticulously unearthed and reconstructed in its original form, but a child’s kaleidoscope in which each glance through the pinhole of a moment in time provides a unique view; a complex organization in which a ﬁxed set of colored, shaped, and textured components rearrange themselves in unique crystalline structures determined by way of inﬁnite pathways of interconnectedness” (p. 197). The overlap between EAM, the self, and autonoetic consciousness is depicted as a Venn-diagram below (Fig. 5).

Fig. 5. EAM, the self, and autonoetic consciousness overlap considerably. They are connected by their embeddedness in time.

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

33

As it is evident from the results of work with patients with neurological or psychiatric diseases, as well as from results on the developing human individual, different components of memory contribute to the formation of the self and make it autonoetic. The dimensions of time as reﬂected in mental time traveling and autonetic consciousness are central to the trans-temporal cohesiveness of self across contexts. As Wheeler and colleagues (1997) put forth (see above), autonoetic consciousness allows the individual ‘‘to become aware of their protracted existence across subjective time” (p. 335) (Fig. 5). Furthermore it gives memory that binding power, which prevents the one-ness from getting lost in the multiplicity (CooperWhite, 2008). This binding (connecting) power of memory that is also metaphorically conveyed by the etymology of the word re-collection (‘‘re-collection” derives from the word ‘‘collection” and the latter has similar roots as ‘‘colligation”, which means binding) (Casey, 2000) was beautifully captured by Hering. He wrote in 1870 (p. 12): ‘‘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.” 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 imagining and remembering. Neuropsychologia, 47, 2222–2238. Adolphs, R. (2010). Conceptual challenges and directions for social neuroscience. Neuron, 65, 752–767. Alkire, M. T., Hudetz, A. G., & Tononi, G. (2008). Consciousness and anesthesia. Science, 322, 876–880. 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. Amting, J. M., Greening, S. G., & Mitchell, D. G. V. (2010). Multiple meachnisms of consciousness: The neural correlates of emotional awareness. The Journal of Neuroscience, 30, 10039–10047. Andelman, F., Hooﬁen, D., Goldberg, H., Aizenstein, O., & Neufeld, M. Y. (2010). Bilateral hippocampal lesion and a selective impairment of the ability for mental time travel. Neurocase. doi:10.1080/13554791003623318. Azry, 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. 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. Banks, S., & Weintraub, S. (2008). Self-awareness and self-monitoring of cognitive and behavioral deﬁcits in behavioral variant frontotemporal dementia, primary progressive aphasia and probable Alzheimer’s disease. Brain and Cognition, 67, 58–68. Bard, K. A., Todd, B. K., Bernier, C., Love, J., & Leavens, D. A. (2006). Self-awareness in human and chimpanzee infants: What is measured and what is meant by the mark and mirror test? Infancy, 9, 191–219. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘‘theory of mind”? Cognition, 21, 37–46. Barresi, J. (2002). From ’the Thought is the Thinker’ to ‘the Voice is the Speaker’: William James and the dialogical self. Theory & Psychology, 21, 237–250. Bartlett, F. C. (1932). Remembering. Cambridge, UK: Cambridge University Press. 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. Beer, J. S. (2007). The default mode: Feeling good or being right? Trends in Cognitive Sciences, 11, 187–189. Beer, J. S., Lombardo, M. V., & Bhanji, J. P. (2010). Roles of medial prefrontal cortex and orbitofrontal cortex in self-evaluation. Journal of Cognitive Neuroscience, 22, 2108–2119. Bengner, T., & Malina, T. (2008). Remembering versus knowing during face recognition in unilateral temporal lobe epilepsy patients with or without hippocampal sclerosis. Brain and Cognition, 68, 148–156. Benoit, R. G., Gilbert, S. J., Volle, E., & Burgess, P. W. (2010). When I think about me and simulate you: Medial rostral prefrontal cortex and self-referential processes. NeuroImage, 50, 1340–1349. Bergouignan, L., Lemogne, C., Foucher, A., Longin, E., Vistoli, D., Allilaire, J.-F., et al. (2008). Field perspective deﬁcit for positive memories characterizes autobiographical memory in euthymic depressed patients. Behaviour Research and Therapy, 46, 322–333. Blakemore, S.-J. (2010). The developing social brain: Implications for education. Neuron, 65, 744–747. Blanke, O., & Metzinger, T. (2009). Full-body illusions and minimal phenomena selfhood. Trends in Cognitive Sciences, 13, 7–13. Block, N. (1995). On a confusion about a function of consciousness. Behavioral and Brain Sciences, 18, 227–247. 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. Bogen, J. E. (1995). On the neurophysiology of consciousness: I. An overview. Consciousness and Cognition, 4, 52–62. Borsutzky, S., Fujiwara, E., Brand, M., & Markowitsch, H. J. (2008). Confabulations in alcoholic Korsakoff patients. Neuropsychologia, 46, 3133–3143. Borsutzky, S., Fujiwara, E., Brand, M., & Markowitsch, H. J. (2010). Susceptibility to false memories in patients with ACoA aneurysm. Neuropsychologia, 48, 2811–2823. Botzung, A., Denkova, E., & Manning, L. (2008). Experiencing past and future personal events: Functional neuroimaging evidence on the neural bases of mental time travel. Brain and Cognition, 66, 202–212. Botzung, A., LaBar, K. S., Kragel, P., Miles, A., & Rubin, D. C. (2010). Component neural systems for the creation of emotional memories during free viewing of a complex, real-world event. Frontiers in Human Neuroscience, 4(Article 34), 1–10. Botzung, A., Rubin, D. C., Miles, A., Cabeza, R., & LaBar, K. S. (2010). Mental hoop diaries: Emotional memories of a college basketball game in rival fans. Journal of Neuroscience, 30, 2130–2137. 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, & 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, T., Schautzer, F., Hamilton, D., Brüning, R., Markowitsch, H. J., Kalla, R., et al. (2005). Vestibular loss causes hippocampal atrophy and impaired spatial memory in humans. Brain, 128, 2732–2741. Bressler, S. J. (1995). Large-scale cortical networks and cognition. Brain Research Reviews, 20, 288–304. Breuer, J., & Freud, S. (1895). Studien über Hysterie [Studies on hysteria]. Wien: Deuticke.

34

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

Bromberg, P. (1994). ‘Speak! that I may see you’: Some reﬂections on dissociation, reality, and psychoanalytic listening. Psychoanalytic Dialogues, 4, 517–547. Brosnan, S. F., & Bshary, R. (2010). Cooperation and deception: Rom evolution to mechanisms. Philosophical Transactions of the Royal Society B, 365, 2593–2598. Brown, J. W. (1990). Psychology of time awareness. Brain and Cognition, 14, 144–164. Brunet, A., Orr, S. P., Tremblay, J., Robertson, K., Nader, K., & Pitman, R. K. (2008). Effect of post-retrieval propranolol on psychophysiologic responding during subsequent script-driven traumatic imagery in post-traumatic stress disorder. Journal of Psychiatric Research, 42, 503–506. Buchanan, T. W., Tranel, D., & Adolphs, R. (2005). Emotional autobiographical memories in amnesic patients with medial temporal lobe damage. Journal of Neuroscience, 25, 3151–3160. Buckner, R. L., & Carroll, D. C. (2007). Self-projection and the brain. Trends in Cognitive Sciences, 11, 49–57. 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 consciousnes before the criminal judge]. Münchener medizinische Wochenschrift, 37, 1270–1273. Burgmer, M., Konrad, C., Jansen, A., Kugel, H., Sommer, J., Heindel, W., et al. (2006). Abnormal brain activation during movement observation in patients with conversion paralysis. Neuroimage, 29, 1336–1343. Cahill, L., Babinsky, R., Markowitsch, H. J., & McGaugh, J. L. (1995). Involvement of the amygdaloid complex in emotional memory. Nature, 377, 295–296. Calabrese, P., Markowitsch, H. J., Durwen, H. F., Widlitzek, B., Haupts, M., Holinka, B., et al. (1996). Right temporofrontal cortex as critical locus for the ecphory of old episodic memories. Journal of Neurology, Neurosurgery, and Psychiatry, 61, 304–310. Call, J., & Tomasello, M. (2008). Does the chimpanzee have a theory of mind? 30 years later. Trends in Cognitive Sciences, 12, 187–192. Campbell, B. C., & Garcia, J. R. (2009). Neuroanthropology: Evolution and emotional embodiment. Frontiers in Evolutionary Neuroscience, 1, 1–6. Casey, E. S. (2000). Remembering. A phenomenological study (2nd ed.). Bloomington, IN: Indiana University Press. Caspi, A., Hariri, A. R., Holmes, A., Uher, R., & Mofﬁtt, T. E. (2010). Genetic sensitivity to the environment: The case of the serotonin transporter gene and its implications for studying complex diseases and traits. American Journal of Psychiatry, 167, 509–527. Clayton, N. S., & Dickinson, A. (1998). Episodic-like memory during cache recovery in scrub-jays. Nature, 395, 272–274. Coleman, M. R., Davis, M. H., Rodd, J. M., Robson, T., Ali, A., Owen, A. M., et al. (2009). Towards the routine use of brain imaging to aid the clinical diagnosis of disorders of consciousness. Brain, 132, 2541–2552. Colom, R., Jung, R. E., & Haier, R. J. (2007). General intelligence and memory span: Evidence for a common neuroanatomic framework. Cognitive Neuropsychology, 24, 867–878. Conway, M. A. (2009). Episodic memories. Neuropsychologia, 47, 2305–2313. Cooper-White, P. (2008). Interrogating integration, dissenting dis-integration: Multiplicity as a positive metaphor in therapy and theology. Pastoral Psychology, 57, 3–15. Corcoran, R., & Frith, C. D. (2003). Autobiographical memory and theory of mind: Evidence of a relationship in schizophrenia. Psychological Medicine, 33, 897–905. Courage, M. L., & Cowan, N. (2009). Introduction: What’s new in research on the development of memory in infants and children? In M. L. Courage & N. Cowan (Eds.), The development of memory in infancy and childhood (pp. 1–10). Hove, East Sussex: Psychology Press. Cowan, N. (2000). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24, 87–185. Craig (Bud), A. D. (2009). How do you feel – now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10, 59–70. Croft, K. E., Duff, M. C., Kovach, C. K., Anderson, S. W., Adolphs, R., & Tranel, D. (2010). Detectable or marvellous? Neuroanatomical correlates of character judgment. Neuropsychologia, 48, 1789–1801. Damasio, A. (1999). The feeling of what happens. Body and emotion in the making of consciousness. New York, NY: Harcourt Brace. Damasio, A. R., Eslinger, P. J., Damasio, H., Van Hoesen, G. W., & Cornell, S. (1985). Multimodal amnesic syndrome following bilateral temporal and basal forebrain damage. Archives of Neurology, 42, 252–259. 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, 898–910. Davies, J. M. (1996). Dissociation, repression, and reality testing in the countertransference: False memory in the psychoanalytic treatment of adult survivors of childhood sexual abuse. Psychoanalytic Dialogues, 6, 189–218. Davis, S. W., Dennis, N. A., Buchler, N. G., White, L. E., Madden, D. J., & Cabeza, R. (2009). Assessing the effects of age on long white matter tracts using diffusion tensor tractography. Neuroimage, 46, 530–541. de Kloet, E. R., Oiztl, M. S., & Joels, M. (1999). Stress and cognition: Are corticosteroids good or bad guys? Trends in Neurosciences, 22, 422–426. de Waal, F. B., Dindo, M., Freeman, C. A., & Hall, M. J. (2005). The monkey in the mirror: Hardly a stranger. Proceedings of the National Academy of Sciences of the USA, 102, 11140–11147. de Waal, F. B. M., & Suchak, M. (2010). Prosocial primates: Selﬁsh and unselﬁsh motivations. Philosophical Transactions of the Royal Society B, 365, 2711–2722. Diederen, K. M., Neggers, S. F., Daalman, K., Blom, J. D., Goekoop, R., Kahn, R. S., et al. (2010). Deactivation of the parahippocampal gyrus preceding auditory hallucinations in schizophrenia. American Journal of Psychiatry, 167, 427–436. Dobbins, I. G., Foley, H., Schacter, D. L., & Wagner, A. D. (2002). Executive control during episodic retrieval: Multiple prefrontal processes subserve source memory. Neuron, 35, 989–996. Dobbins, I. G., Simons, J. S., & Schacter, D. L. (2004). FMRI evidence for separable and lateralized prefrontal memory monitoring processes. Journal of Cognitive Neuroscience, 16, 908–920. DSM-IV-TR (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. Edelman, G. M. (1998). Building a picture of the brain. Daedalus, 127, 37–70. Ellen, P. A., Kaut, K. P., & Lord, R. R. (2009). Emotion and episodic memory. In E. Dere, A. Easton, & J. P. Huston (Eds.). Handbook of behavioral neuroscience: Episodic memory research (Vol. 18, pp. 115–132). Amsterdam: Elsevier. Enzi, B., de Greck, M., Prösch, U., Tempelmann, C., & Northoff, G. (2009). Is or self nothing but reward? Neuronal overlap and distinction between reward and personal relevance and its relation to human personality. PLoS ONE, 4, e8429. Eustache, F., Piolino, P., Giffard, B., Viader, F., De La Sayette, V., Baron, J.-C., et al. (2004). ‘In the course of time: A PET study of the cerebral substrates of autobiographical amnesia in Alzheimer’s disease. Brain, 127, 1549–1560. Everett, D. L. (2005). Cultural constrains on grammar and cognition in Pirahã. Current Anthropology, 46, 621–646. Everett, D. L. (2008). Don’t sleep, there are snakes. New York, NY: Vintage. Fargeau, M. N., Jaafari, N., Ragot, S., Houeto, J. L., Pluchon, C., & Gil, R. (in press). Alzheimer’s disease and impairment of the self. Consciousness and Cognition. Feinberg, T. E. (2005). Neural hierarchies and the self. In T. E. Feinberg & J. P. Keenan (Eds.), The lost self: Pathologies of the brain and identity (pp. 33–49). New York: Oxford University Press. Feinberg, T. E., & Keenan, J. P. (2005). Where in the brain is the self? Consciousness and Cognition, 14, 661–678. Feinstein, J. S., Duff, M. C., & Tranel, D. (2010). Sustained experience of emotion after loss of memory in patients without amnesia. Proceedings of the National Academy of Sciences of the USA, 107, 7674–7679. Feinstein, A., & Hattersley, A. (1988). Ganser symptoms, dissociation, and dysprosody. Journal of Nervous and Mental Disease, 176, 692–693. 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. Fivush, R., & Nelson, K. (2004). Culture and language in the emergence of autobiographical memory. Psychological Science, 15, 573–577.

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

35

Forcato, C., Rodrigurez, M. L., Pedreira, M. E., & Maldano, H. (2010). Reconsolidation in humans opens up declarative memory to the entrance of new information. Neurobiology of Learning and Memory, 93, 77–84. Foster, P. S., Drago, V., Ferguson, B. J., & Harrison, D. W. (2008). Cerebral moderation of cardiovascular functioning: A functional cerebral systems perspective. Clinical Neurophysiology, 119, 2846–2854. Fujie, S., Namiki, C., Nishi, H., Yamada, M., Miyata, J., Sakata, D., et al. (2008). The role of the uncinate fasciculus in memory and emotional recognition in amnestic mild cognitive impairment. Dementia and Geriatric Cognitive Disorders, 26, 432–439. 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. Fukatsu, R., Fujii, T., Yamadori, A., Nagasawa, H., & Sakurai, Y. (1997). Persisting childish behavior after bilateral thalamic infarcts. European Neurology, 37, 230–235. Gallagher, I. I. (2000). Philosophical conceptions of the self: Implications for cognitive science. Trends in Cognitive Sciences, 4, 14–21. Ganis, G., Kosslyn, S. M., Stose, S., Thompson, W. I., & Yurgelun-Todd, D. A. (2003). Neural correlates of different types of deception: An fMRI investigation. Cerebral Cortex, 13, 830–836. Ganis, G., Morris, M., & Kosslyn, S. M. (2009). Neural processes underlying self- and other-related lies: An individual difference approach using fMRI. Social Neuroscience, 4, 539–553. 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. Ganser, S. J. (1965). A peculiar hysterical state (transl. by C.E. Schorer). British Journal of Criminology, Delinguency, and Deviant Social Behaviour, 5, 120–126. Ghetti, S., DeMaster, D. M., Yonelinas, A. P., & Bunge, S. A. (2010). Developmental differences in medial temporal lobe function during memory encoding. Journal of Neuroscience, 30, 9548–9556. Gibson, K. R. (1991). Myelination and behavioral development: A comparative perspective of neoteny, altriciality and intelligence. In K. R. Gibson & A. C. Peterson (Eds.), Brain maturation and cognitive development: Comparative and cross-cultural perspectives (pp. 29–63). Hawthorne, NY: Aldine de Gruyter. Gilbert, D. T., & Wilson, T. D. (2007). Prospection: Experiencing the future. Science, 317, 1351–1354. 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. Greene, J. D. W., Hodges, J. R., & Baddeley, A. D. (1995). Autobiographical memory and executive function in early dementia of Alzheimer type. Neuropsychologia, 33, 1647–1670. Gregory, C., Lough, S., Stone, V., Erzinclioglu, S., & Martin, L. (2002). Theory of mind in patients with frontal variant frontotemporal dementia and Alzheimer’s disease: Theoretical and practical implications. Brain, 125, 752–764. 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. Grossi, D., Trojano, L., Grasso, A., & Orsini, A. (1988). Selective ‘‘semantic amnesia” after closed-head injury. Cortex, 24, 457–464. Gutchess, A. H., & Indeck, A. (2009). Cultural inﬂuences on memory. Progress in Brain Research, 178, 137–150. Häfner, H. (1954). Über Zeitdehnungs- und Zeitbeschleunigungsphänomen im Rahmen von Zwischenhirnstörungen [On phenomena of slowed-motion and speeded-up motion oft he time-sense in disturbances oft he diencephalon]. Monatsschrift für Psychiatrie und Neurologie, 127, 336–348. Haist, F., Bowden Gore, J., & Mao, H. (2001). Consolidation of human memory over decades revealed by functional magnetic resonance imaging. Nature 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. Harbus, A. (2010). Written autobiography as a source of inﬂuence on autobiographical memory. In W. Christensen, E. Schier, & J. Sutton (Eds.), ASCS09: Proceedings of the 9th conference of the Australasian Society for Cognitive Science (pp. 126–130). Sydney: Macquarie Centre for Cognitive Science. Hare, B., & Tomasello, M. (2005). Human-like social skills in dogs? Trends in Cognitive Sciences, 9, 439–444. Hart, E. (1995). The sketchpad model. A theory of consciousness, perception, and imagery. Consciousness and Cognition, 4, 346–368. Henke, K. (2010). A model for memory systems based on processing modes rather than consciousness. Nature Reviews Neuroscience, 11, 523–532. Hering, E. (1870). 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.: 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. Hirano, M., Noguchi, K., Hosokowa, T., & Takayama, T. (2002). I cannot remember, but I know my past events: Remembering and knowing in a patient with amnesic syndrome. Journal of Clinical and Experimental Neuropsychology, 24, 548–888. Hoff, H., & Pötzl, O. (1938). Anatomischer Befund eines Falles mit Zeitrafferphänomen [Anatomical proof of a case with a speeded-up motion phenomenon]. Deutsche Zeitschrift für Nervenheilkunde, 145, 150–178. Howe, M. L., & Courage, M. L. (1993). On resolving the enigma of infantile amnesia. Psychological Bulletin, 113, 305–326. Hüfner, K., Binetti, C., Hamilton, D.A., Stephan, T., Flanagin, V.L., Linn, J., et al. (in press). Structural and functional plasticity of the hippocampal formation in professional dancers and slackliners. Hippocampus. Hüfner, K., Hamilton, D. A., Kalla, R., Stephan, T., Glasauer, S., Ma, J., et al. (2007). Spatial memory and hippocampal volume in humans with unilateral vestibular deafferentiation. Hippocampus, 17, 471–485. Huttenlocher, P. R. (1994). Synaptogenesis in human cerebral cortex. In G. Dawson & K. W. Fischer (Eds.), Human behavior and the developing brain (pp. 137–152). New York: Guilford Press. ICD-10 (1992). Classiﬁcation of mental and behavioral disorders: Clinical descriptions and diagnostic guidelines. Geneva: World Health Organization. Janet, P. (1907). The major symptoms of hysteria. New York, NY: Macmillan. Jenkins, K. G., Kapur, N., & Kopelman, M. (2009). Retrograde amnesia and malingering. Current Opinion in Neurology, 22, 601–605. Joels, M., & Baram, T. Z. (2010). 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. Kaplan, J. T., Aziz-Zadeh, L., Uddin, L. Q., & Iacobini, M. (2008). The self across the senses: An fMRI study of self-face and self-voice recognition. Social, Cognitive and Affective Neuroscience, 3, 218–223. Keenan, J. P., Rubin, J., Racioppi, C., Johnson, A., & Barnacz, A. (2005). The right hemisphere and the dark side of consciousness. Cortex, 41, 695–704. Keenan, J. P., Wheeler, M. A., Gallup, G. G., Jr., & Pascual-Leone, A. (2000). Self-recognition and the right prefrontal cortex. Trends in Cognitive Sciences, 4, 338–344. Kitchen, A., Denton, D., & Brent, L. (1996). Self-recognition and abstraction abilities in the common chimpanzee studied with distorting mirrors. Proceedings of the National Academy of Sciences of the USA, 93, 7405–7408. 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. Kluft, R. P. (1988). The dissociative disorders. In J. A. Talbott, R. E. Hales, & S. C. Yudofsky (Eds.), Textbook of psychiatry (pp. 557–585). Washington, DC: American Psychiatric Press. Kobayashi, C., Glover, G. H., & Temple, E. (2008). Switching language switches mind: Linguistic effects on developmental neural bases of ‘Theory of Mind’. Social, Cognitive, and Affective Neuroscience, 3, 62–70.

36

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

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. Kuhl, P. K., Tsao, F.-M., & Liu, H.-M. (2003). Foreign-language experience in infancy: Effects of short-term exposure and social interaction on phonetic learning. Proceedings of the National Academy of Sciences of the USA, 100, 9096–10001. 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. Kumral, E., Gulluoglu, H., & Dramali, B. (2007). Thalamic chronotaraxis: Isolated time disorentation. Journal of Neurology, Neurosurgery and Psychiatry, 78, 880–882. Kwan, D., Carson, N., Addis, D. R., & Rosenbaum, R. S. (2010). Deﬁcits in past remembrances extend to future imaging in a case of developmental amnesia. Neuropsychologia, 48, 3179–3186. LaBar, K. S., & Cabeza, R. (2006). Cognitive neuroscience of emotional memory. Nature Reviews Neuroscience, 7, 54–64. Ladowsky-Brooks, R. L., & Fischer, C. E. (2003). Ganser symptoms in a case of frontal–temporal lobe dementia: Is there a common neural substrate? Journal of Clinical and Experimental Neuropsychology, 25, 761–768. Langnickel, R., & Markowitsch, H. J. (2006). Repression and the unconsciousness. Behavioral and Brain Sciences, 29, 524–525. Lee, T. M. C., Ip, A. K. Y., Wang, K., Xi, C.-H., Hu, P.-P., Mak, H. K. F., et al. (2010). Faux pas deﬁcits in people with medial frontal lesions as related to impaired understanding of a speaker’s mental state. Neuropsychologia, 48, 1670–1676. Lemogne, C., Bergouignan, L., Piolino, P., Jouvent, R., Allilaire, J.-F., & Fossati, P. (2008). Cognitive avoidance of intrusive memories and autobiographical memory: Speciﬁcity, autonoetic consciousness, and self-perspective. Memory, 17, 1–7. Lemogne, C., Bergouignan, L., Boni, C., Gorwood, P., Pelissolo, A., & Fossati, P. (2009). Genetics and personality affect visual perspective in autobiographical memory. Consciousness and Cognition, 18, 823–830. 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. 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. Locke, J. (1689/1964). An essay concerning human understanding. Cleveland: Meridian Books. Loewenstein, R. J. (1991). Psychogenic amnesia and psychogenic fugue: A comprehensive review. American Psychiatric Press Review of Psychiatry, 10, 189–222. Loftus, E. F. (2000). Remembering what never happened. In E. Tulving (Ed.), Memory, consciousness, and the brain: The Tallinn conference (pp. 319–330). Philadelphia, PA: Psychology Press. Loftus, E. F., & Hoffman, H. G. (1989). Misinformation and memory: The creation of new memories. Journal of Experimental Psychology: General, 118, 100–104. Loftus, E. F., & Pickrell, J. E. (1995). The formation of false memories. Psychiatric Annals, 25, 720–725. Lombardo, M. V., Chakrabarti, B., Bullore, E. T., Sadek, S. A., Pasco, G., Wheelwright, S. J., et al. (2010). Atypical neural self-representation in autism. Brain, 133, 611–624. 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. 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.). Arlington, VA: American Psychiatric Publ. Markowitsch, H. J. (1992). Intellectual functions and the brain. An historical perspective. Toronto: Hogrefe & Huber Publs. Markowitsch, H. J. (1995). Cerebral bases of consciousness: A historical view. Neuropsychologia, 33, 1181–1192. Markowitsch, H. J. (1998/1999). Differential contribution of the right and left amygdala to affective information processing. Behavioural Neurology, 11, 233–244. Markowitsch, H. J. (1999). Functional neuroimaging correlates of functional amnesia. Memory, 7, 561–583. Markowitsch, H. J. (2000). Repressed memories. In E. Tulving (Ed.), Memory, consciousness, and the brain: The Tallinn conference (pp. 319–330). Philadelphia, PA: Psychology Press. Markowitsch, H. J. (2002). Functional retrograde amnesia – mnestic block syndrome. Cortex, 38, 651–654. Markowitsch, H. J. (2003). Autonoëtic consciousness. In A. S. David & T. Kircher (Eds.), The self in neuroscience and psychiatry (pp. 180–196). Cambridge, UK: 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). Singapur: World Scientiﬁc Publ. Markowitsch, H. J., & Staniloiu, A. (in preparation-a). Amygdala in action: Relaying biological and social signiﬁcance to autobiographic memory. Neuropsychologia. Markowitsch, H. J., & Staniloiu, A. (in preparation-b). Amnesia. Lancet. Markowitsch, H. J., Calabrese, P., Liess, J., Haupts, M., Durwen, H. F., & Gehlen, W. (1993). Retrograde amnesia after traumatic injury of the temporo-frontal cortex. Journal of Neurology, Neurosurgery and Psychiatry, 56, 988–992. Markowitsch, H. J., Calabrese, P., Neufeld, H., Gehlen, W., & Durwen, H. F. (1999). Retrograde amnesia for famous events and faces after left fronto-temporal brain damage. 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., Cramon, D. Y. V., & 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., 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., Reinkemeier, M., Kessler, J., Koyuncu, A., & Heiss, W.-D. (2000). Right amygdalar and temporofrontal activation during autobiographic, but not during ﬁctitious memory retrieval. Behavioural Neurology, 12, 181–190. 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. 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. Markowitsch, H. J., & Welzer, H. (2009). The development of autobiographical memory. Hove, UK: Psychology Press.

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

37

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. Melis, A. P., & Semmann, D. (2010). How is human cooperation different? Philosophical Transactions of the Royal Society B, 365, 2663–2674. Melloni, L., Molina, C., Pena, M., Torres, D., Singer, W., & Rodriguez, E. (2007). Synchronization of neural activity across cortical areas correlates with conscious perception. Journal of Neuroscience, 27, 2858–2865. Mesulam, M.-M. (2000). Behavioral neuroanatomy: Large-scale networks, association cortex, frontal syndromes, the limbic system, and hemispheric specialization. In M.-M. Mesulam (Ed.), Principals of behavioral neurology (2nd Ed., pp. 1–120). New York: Oxford University Press. Mhuircheartaigh, R. N., Rosenorn-Lanng, D., Wise, R., Jbabdi, S., Rogers, R., & Tracey, I. (2010). Cortical and subcortical connectivity changes during decreasing levels of consciousness in humans: A functional magnetic resonance imaging study using Propofol. The Journal of Neuroscience, 30, 9095–9102. Miles, L. K., Karpinska, K., Lumsden, J., & Macrae, C. N. (2010). The meandering mind: Vection and mental time travel. PLoS ONE, 5(5), e10825. doi:10.1371/ journal.pone.0010825. Miller, G. A. (1956). The magical number seven plus minus two. Some limits on our capacity for processing information. Psychological Review, 63, 244–257. Milner, B., Corkin, S., & Teuber, H. L. (1968). Further analysis of the hippocampal amnesic syndrome: Fourteen year follow-up study of H.M. Neuropsychologia, 6, 215–234. Miyata, H., Gajdon, G. K., Huber, L., & Fujita, K. (in press). How do keas (Nestor notabilis) solve artiﬁcial-fruit problems with multiple locks? Animal Cognition. Modell, A. H. (2006). Imagination and the meaningful brain. Cambridge, MA: Bradford. Modinos, G., Ormel, J., & Aleman, A. (2009). Activation of anterior insula during self-reﬂection. PLoS ONE, 4(e4618), 1–8. 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. Morrison, C. M., & Conway, M. (2010). First words and ﬁrst memories. Cognition. doi:10.1016/j.cognition.2010.03.011. Murty, V. P., Ritchey, M., Adcock, R. A., & LaBar, K. S. (in press). fMRI studies of successful emotional memory encoding: A quantitative meta-analysis. Neuropsychologia. doi: 10.1016/j.neuropsychologia.2010.07.030. Nauta, W. J. H. (1979). Expanding borders of the limbic system concept. In T. Rasmussen & R. Marino (Eds.), Functional neurosurgery (pp. 7–23). New York: Raven Press. Neisser, U. (1988). Five kinds of self-knowledge. Philosophical Psychology, 1, 35–59. 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. (2005a). 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. (2005b). Emerging levels of consciousness in early human development. In H. Terrace & J. Metcalfe (Eds.), The missing link in cognition (pp. 116–141). New York: Oxford University Press. Nelson, K. (2007). The extended self. Journal of Physiology (Paris), 101, 223–229. Nelson, K., & Fivush, R. (2004). The emergence of autobiographical memory: A social cultural developmental theory. Psychological Review, 111, 486–511. Newen, A., & Vogeley, K. (2003). Self-representation: Searching for a neural signature of self-consciousness. Consciousness and Cognition, 12, 529–543. Nieuwenhuys, R. (1996). The greater limbic system, the emotional motor system and the brainG. Holstege, R. Bandler, & C. B. Saper (Eds.). The emotional motor system. Progress in Brain Research, Vol. 107, 551–580. 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, 440–457. Northoff, G., & Panksepp, J. (2008). The trans-species concept of self and the subcortical–cortical midline system. Trends in Cognitive Sciences, 12, 259–264. Noulhiane, M., Piolino, P., Hasboun, D., Clemenceau, S., Baulac, M., & Samson, S. (2007). Autobiographical memory after temporal lobe resection: Neuropsychological and MRI volumetric ﬁndings. Brain, 130, 3184–3199. Noulhiane, M., Piolino, P., Hasboun, D., Clemenceau, S., Baulac, M., & Samson, S. (2008). Autonoetic consciousness in autobiographic memory after medial temporal lobe resection. Behavioural Neurology, 19, 19–22. O’Brien, J. T. (1997). The ‘glucocorticoid cascade’ hypothesis in man. British Journal of Psychiatry, 170, 199–201. 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. Oddo, S., Thiel, A., Skoruppa, S., Klinger, D., Steis, N., Markowitsch, H. J., et al. (2009). Neurobiological and psychological aspects of BIID – an integrative approach. In A. Stirn, A. Thiel, & S. Oddo (Eds.), Body integrity identity disorder: Psychological, neurobiological, ethical and legal aspects (pp. 238–246). Lengerich: Pabst. O’Keeffe, F. M., Murray, B., Coen, R. F., Dockree, P. M., Bellgrove, M. A., Garavan, H., et al. (2007). Loss of insight in frontotemporal dementia, corticobasal degeneration and progressive supranuclear palsy. Brain, 130(3), 753–764. Oliveri, M., Bonni, S., Turriziani, P., Koch, G., Lo Gerfo, E., & Torriero, S. (2009). Motor and linguistic linking of space and time in the cerebrellum. PLoS ONE, 4(11), e7993. doi:10.1371/journal.pone.0007933. Osvath, M. (2010). Great ape foresight is looking great. Animal Cognition, 13, 777–781. Panksepp, J. (1998). Affective neuroscience: The foundations of human and animal emotions. New York: Oxford Univ. Press. Passingham, R. E., Bengtsson, S. L., & Lau, H. C. (2010). Medial frontal cortex: From self-generated action to reﬂection on one’s own performance. Trends in Cognitive Sciences, 14, 16–21. Paz-Alonso, P. M., Ghetti, S., Matlen, B. J., Anderson, M. C., & Bunge, S. A. (2009). Memory suppression is an active process that improves over childhood. Frontiers in Human Neuroscience, 3(Article 24), 1–6. 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., & Aichhorn, M. (2008). Theory of mind, language and the temporoparietal junction mystery. Trends in Cognitive Sciences, 12, 123–126. Perner, J., Kloo, D., & Rohwer, M. (2010). Retro- and prospection for mental time travel: Emergence of episodic remembering and mental rotation in 5- to 8year old children. Consciousness and Cognition, 19, 802–815. 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., & Ruffmann, T. (1995). Episodic memory and autonoetic consciousness: Developmental evidence and a theory of childhood amnesia. Journal of Experimental Child Psychology, 59, 516–548. Peterson, C., Wang, Q., & Hou, Y. (2009). ‘‘When I was little”: Childhood recollection in Chinese and European Canadian grade school children. Child Development, 80, 506–518. Pfeifer, R., & Bongard, J. (2007). How the body shapes the way we think. Cambridge, MA: MIT Press. 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. Picard, L., Reffuveille, I., Eustache, F., & Piolino, P. (2009). Development of autonoetic autobiographical memory in school-age children: Genuine age effect or development of basic cognitive abilities? Consciousness and Cognition, 18, 864–876.

38

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

Piefke, M., Weiss, P. H., Zilles, K., Markowitsch, H. J., & Fink, G. R. (2003). Differential remoteness and emotional tone modulate the neural correlates of autobiographical memory. Brain, 126, 650–668. Pinker, S. (1997). How the mind works. New York: W.W. Norton. Piolino, P., Belliard, S., Desgranges, B., Perron, M., & Eustache, F. (2003). Autobiographical memory and autonoetic consciousness in a case of semantic dementia. Cognitive Neuropsychology, 20, 619–639. Piolino, P., Desgranges, B., Clarys, D., Guillery-Girard, B., Taconnat, L., Isingrini, M., et al. (2006). Autobiographical memory, autonoetic consciousness, and self-perspective in aging. Psychology and Aging, 21, 510–525. 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., Giffard-Quillon, G., Desgranges, B., Chételat, G., Baron, J.-C., & Eustache, F. (2004). Re-experiencing old memories via hippocampus: A PET study of autobiographical memory. NeuroImage, 22, 1371–1383. 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. Plateck, S. M., Keenan, J. P., Gallup, G. G., Jr., & Mohamed, F. B. (2003). Where am I? The neurological correlates of self and other. Cognitive Brain Research, 19, 114–122. Plotnik, J. M., de Waal, F. B., & Reiss, D. (2006). Self-recognition in an Asian elephant. Proceedings of the National Academy of Sciences of the USA, 103, 17053–17057. Pope, H. G., Jr., Poliakoff, M. B., Parker, M. P., Boynes, M., & Hudson, J. I. (2007). Is dissociative amnesia a culture-bound syndrome? Findings from a survey oh historical literature. Psychological Medicine, 37, 225–233. Porter, N., & Landﬁeld, P. W. (1998). Stress hormones and brain aging: Adding injury to insult? Nature Neuroscience, 1, 3–4. Pötzl, O. (1939). Physiologisches und Pathologisches über das persönliche Tempo [Physiologic and pathologic ﬁndings on the personal tempo]. Wiener klinische Wochenschrift, 52, 569–573. Pötzl, O. (1942). Alterserkrankungen im Gehirn [Age-related diseases in the brain]. Wiener klinische Wochenschrift, 55, 3–8. Pötzl, O. (1951). Weiteres über das Zeitraffer-Erlebnis [More on the experience of speeded-up motion]. Wiener Zeitschrift für Nervenheilkunde, 4, 9–39. Pötzl, O. (1958). Zur Hirnphysiologie des Zeiterlebens [On the brain physiology of the experience of time]. Wiener Zeitschrift für Nervenheilkunde, 15, 370–393. 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. Powell, L. J., Macrae, C. N., Cloutier, J., Metcalfe, J., & Mitchell, J. P. (2009). Dissociable neural substrates for agentic versus conceptual representations of self. Journal of Cognitive Neuroscience, 22, 2186–2197. Pribram, K. H., & Tubbs, W. E. (1967). Short-term memory, parsing and the primate frontal cortex. Science, 156, 1765–1767. Raine, A. (2003). Annotation: The role of prefrontal deﬁcits, low autonomic arousal, and 790 early health factors in the development of antisocial and aggressive behavior in 791 children. Journal of Child Psychology and Psychiatry, 43, 417–434. Raine, A., Lencz, T., Taylor, K., Hellige, J. B., Bihrle, S., Lacasse, L., et al. (2003). Corpus callosum abnormalities in psychopathic antisocial individuals. Archives of General Psychiatry, 60, 1134–1142. 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 Neuropsychologia, 1, 53–64. Reinhold, N., & Markowitsch, H. J. (2009). Retrograde episodic memory and emotion: A perspective from patients with dissociative amnesia. Neuropsychologia, 47, 2197–2206. Reiss, D., & Marino, L. (2001). Mirror self-recognition in the bottlenose dolphin: A case of cognitive convergence. Proceedings of the National Academy of Sciences of the USA, 98, 5937–5942. Reuter, F., Del Cul, A., Malikova, I., Naccache, L., Confort-Gouny, S., Cohen, L., et al. (2008). White matter damage impairs access to consciousness in multiple sclerosis. Neuroimage, 44, 590–599. Ribot, T. (1882). Diseases of memory. New York: D. Appleton &Co. Roberts, W. A., & Feeney, M. C. (2009). The comparative study of mental time travel. Trends in Cognitive Sciences, 13, 271–277. Rochat, P. (2010). The innate sense of the body develops to become a public affair by 2–3 years. Neuropsychologia, 48, 738–742. 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. Sarter, M., & Markowitsch, H. J. (1985a). The amygdala’s role in human mnemonic processing. Cortex, 21, 7–24. Sarter, M., & Markowitsch, H. J. (1985b). The involvement of the amygdala in learning and memory: A critical review with emphasis on anatomical relations. Behavioral Neuroscience, 99, 342–380. Satterthwaite, T. D., Wolf, D. H., Loughead, J., Ruparel, K., Valdez, J. N., Siegel, S. J., et al. (2010). Association of enhanced limbic response to thread with decreased cortical facial recognition memory in schizophrenia. American Journal of Psychiatry, 167, 418–426. Schacter, D. L. (2001). Seven sins of memory: How the mind forgets and remembers. New York: Houghton Mifﬂin. Schacter, D. L., Addis, D. R., & Buckner, R. L. (2007). Remembering the past to imagine the future: The prospective brain. Nature Reviews Neuroscience, 8, 657–661. 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. Sebastian, C., Burnett, S., & Blakemore, S.-J. (2008). Development of the self-concept during adolescence. Trends in Cognitive Sciences, 12, 441–446. Seeley, W. W. (2008). Selective functional, regional, and neuronal vulnerability in frontotemporal dementia. Current Opinion in Neurology, 21, 701–707. Semon, R. (1904). Die Mneme als erhaltendes Prinzip im Wechsel des organischen Geschehens. Leipzig: Wilhelm Engelmann. 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. Seth, A. K. (2010). The grand challenge of consciousness. Frontiers in Psychology, 1(Article 5), 1–2. Sharot, T., Riccardi, A. M., Raio, C. M., & Phelps, E. A. (2007). Neural mechanisms mediating optimism bias. Nature, 450, 102–105. Shing, Y. L., Werkle-Bergner, M., Brehmer, Y., Müller, V., Li, S.-C., & Lindenberger, U. (2010). Episodic memory across the lifespan: The contributions of associative and strategic components. Neuroscience and Biobehavioral Reviews, 34, 1080–1091. Sidis, B., & Goodhart, S. P. (1905). Multiple personality: An experimental investigation into the nature of human individuality. New York, NY: Appleton. Siebert, M., Markowitsch, H. J., & Bartel, P. (2003). Amygdala, affect, and cognition: Evidence from ten patients with Urbach–Wiethe disease. Brain, 126, 2627–2637. 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, 48, 2833–2840. Spiegel, E. A. (1982). Guided brain operations. Methodological and clinical developments in stereotactic surgery. Contributions to the physiology of subcortical structures. Basel: Karger. Spiegel, E. A., Wycis, H. T., Orchinik, C., & Freed, H. (1956). Thalamic chronotaraxis. American Journal of Psychiatry, 113, 97–105. Spreng, R. N., Mar, R. A., & Kim, A. S. N. (2009). 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, 249–510.

H.J. Markowitsch, A. Staniloiu / Consciousness and Cognition 20 (2011) 16–39

39

St. Jacques, P. L., Rubin, D. C., LaBar, K. S., & Cabeza, R. (2008). The short and long of it: Neural correlates of temporal-order memory for autobiographical events. Journal of Cognitive Neuroscience, 20, 1327–1341. Staniloiu, A., & Markowitsch, H. J. (in press). Searching for the anatomical basis 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. Christiansen, E. Schier, & J. Sutton (Eds.), ASCS09: Proceedings of the 9th conference of the Australasian Society for Cognitive Science (pp. 316–324). Sydney: Centre for Cognitive Science. Staniloiu, A., Markowitsch, H. J., & Brand, M. (2010). Psychogenic amnesia – A malady of the constricted self. Consciousness and Cognition, 19, 778–801. Stokes, D. J., Dritschel, B. H., & Bekerian, D. A. (2004). The effect of burn injury on adolescents’ autobiographical memory. Behaviour Research and Therapy, 42, 1357–1365. Stone, J., Smyth, R., Carson, A., Warlow, C., & Sharpe, M. (2006). La belle indifference in conversion symptoms and hysteria: Systematic review. British Journal of Psychiatry, 188, 204–209. Suddendorf, T., Addis, D. R., & Corballis, M. C. (2009a). Mental time travel and the shaping of the human mind. Philosophical Transactions of the Royal Society London B, 364, 1317–1324. Suddendorf, T., Corballis, M. C., & Collier-Baker, E. (2009b). How great is great ape foresight? Animal Cognition, 12, 751–754. Surian, L., Caldi, S., & Sperber, D. (2007). Attribution of beliefs y 13-month-old infants. Psychological Science, 18, 580–586. Sutin, A. R., & Robins, R. W. (2008). When the ‘‘I” looks at the ‘‘Me”: Autobiographical memory, visual perspective, and the self. Consciousness and Cognition, 17, 1386–1397. Szyf, M. (2007). The dynamic epigenome and its implications in toxicology. Toxicological Sciences, 100, 7–23. Tanweer, T., Rathbone, C. J., & Souchav, C. (2010). Autobiographical memory, autonoetic consciousness, and identity in Asperger syndrome. Neuropsychologia, 48, 900–908. 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. Tononi, G., & Edelman, G. M. (1998). Consciousness and complexity. Science, 282, 1846–1851. 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. Trivino, M., Correa, A., Arnedo, M., & Lupianez, J. (2010). Temporal orienting deﬁcit after prefrontal damage. Brain, 133, 1173–1185. 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: Oxford University 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., & Markowitsch, H. J. (1997). Memory beyond the hippocampus. Current Opinion in Neurobiology, 7, 209–216. Tulving, E., & Markowitsch, H. J. (1998). Episodic and declarative memory: Role of the hippocampus. Hippocampus, 8, 198–204. Tulving, E., & Thompson, D. M. (1973). Encoding speciﬁcity and retrieval process in episodic memory. Psychological Review, 80, 352–373. 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 Sciences, 11, 153–157. 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. Vandekerckhove, M. P., Markowitsch, H. J., Mertens, M., & Woermann, F. (2005). Bi-hemispheric engagement in the retrieval of autobiographical episodes. Behavioural Neurology, 16, 203–210. Vandekerckhove, M., & Panksepp, J. (2009). The ﬂow of anoetic to noetic and autonoetic consciousness: A vision of unknowing (anoetic) and knowing (noetic) consciousness in the remembrance of things past and imagined futures. Consciousness and Cognition, 18, 1018–1028. Vanhaudenhuyse, A., Noirhomme, Q., Tshibanda, L. J.-F., Bruno, M.-A., Boveroux, P., Schnakers, C., et al. (2010). Default network connectivity reﬂects the level of consciousness in non-communicative brain-damaged patients. Brain, 133, 161–171. Varki, A. (2009). Human uniqueness and the denial of death. Nature, 460, 684. Vogeley, K., & Kupke, C. (2007). Disturbances of time consciousness from a phenomenological and a neuroscientiﬁc perspective. Schizophrenia Bulletin, 33, 157–165. 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 Neuroscience, 5, 817–827. Vogeley, K., & Roepstorff, A. (2009). Contextualising culture and social cognition. Trends in Cognitive Sciences, 13, 511–516. Wang, Q., & Conway, M. A. (2004). The stories we keep: Autobiographical memory in American and Chinese middle-aged adults. Journal of Personality and Social Psychology, 72, 938–991. Wang, Q., Shao, Y., & Li, Y. J. (2010). ‘‘My way or mom’s way?” The bilindual and bicultural self in Hong Kong Chinese children and adolescents. Child Development, 81, 555–567. Weir, A. A. S., Chappell, J., & Kacelnik, A. (2002). Shaping of hooks in New Caledonian crows. Science, 297, 981. 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). Towards 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. Wittling, W. (1997). Brain asymmetry and autonomic control of the heart. European Psychologist, 2, 313–327. Wittling, W., Block, A., Genzel, S., & Schweiger, E. (1998). Hemisphere asymmetry in parasympathetic control of the heart. Neuropsychologia, 36, 461–468. Yaffe, K., Vittinghoff, E., Lindquist, K., Barnes, D., Covinsky, K. E., Neylan, T., et al. (2010). Posttraumatic stress disorder and risk of dementia among US veterans. Archives of General Psychiatry, 67, 608–613. Yonelinas, A. P., Kroll, N. E. A., Dobbins, I., Lazzara, M., & Knight, R. T. (1998). Recollection and familiarity deﬁcits in amnesia: Convergence of rememberknow, process dissociation, and receiver operating characteristic data. Neuropsychology, 12, 323–339. Zhu, Y., Zhang, L., Fan, J., & Han, S. (2007). Neural basis of cultural inﬂuence on self-representation. Neuroimage, 34, 1310–1316. Zöllig, J., West, R., Martin, M., Altgassen, M., Lemke, U., & Kliegel, M. (2007). Neural correlates of prospective memory across the lifespan. Neuropsychologia, 45, 3299–3314.

Consciousness and Cognition 20 (2011) 40–51

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The neural correlates of visual self-recognition q Christel Devue a,b,⇑, Serge Brédart a a b

Centre des Neurosciences Cognitives et Comportementales, Université de Liège, Belgium Cognitive Psychology, Vrije Universiteit Amsterdam, Van der Boechorststraat 1, 1081 BT Amsterdam, The Netherlands

a r t i c l e

i n f o

Article history: Available online 28 September 2010 Keywords: Self-recognition Face perception Self-awareness Neural correlates

a b s t r a c t This paper presents a review of studies that were aimed at determining which brain regions are recruited during visual self-recognition, with a particular focus on self-face recognition. A complex bilateral network, involving frontal, parietal and occipital areas, appears to be associated with self-face recognition, with a particularly high implication of the right hemisphere. Results indicate that it remains difﬁcult to determine which speciﬁc cognitive operation is reﬂected by each recruited brain area, in part due to the variability of used control stimuli and experimental tasks. A synthesis of the interpretations provided by previous studies is presented. The relevance of using self-recognition as an indicator of self-awareness is discussed. We argue that a major aim of future research in the ﬁeld should be to identify more clearly the cognitive operations induced by the perception of the self-face, and search for dissociations between neural correlates and cognitive components. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Our own face is an important component of our identity, besides other self-related information such as our own name, hometown, occupation, and preferences. However, contrary to other self-related information our face is a unique self-referential stimulus, presumably our most distinctive physical feature (Tsakiris, 2008). Indeed, our own face is a property that we do not share with other people (with the exception of twins), whereas it is quite common to share properties such as the occupation, the hometown or even the name with other people (Devue & Brédart, 2008). During the last 40 years, the ability of self-recognition in a mirror has been extensively investigated by researchers searching for signs of ‘self-consciousness’ or some sense of personal identity in infants and in animals (Amsterdam, 1972; Gallup, 1970; for a review see Keenan, Gallup, & Falk, 2003a). The face is often seen as the emblem of the self (McNeill, 1998), and Cole (1998) has described in his book ‘About face’ how life of people suffering from problems touching their own face (e.g., facial paralysis or disﬁgurement) is deeply affected. More recently, with the occurrence of face transplants procedures, psychologists started to consider the possible deleterious effects of such surgical procedures on the patients’ sense of identity (Bluhm & Clendenin, 2009). A much more mundane but signiﬁcant example of the importance of the face in deﬁning a person is illustrated by the fact that since the propagation of photography, passports or driving licenses from all over the world contain an identity picture in addition to the owner’s name. Over the last 10 years, the interest towards visual self-recognition has grown among the neuroscientiﬁc community. One of the reasons is that a lot of researchers in that ﬁeld have made the assumption that presenting people with stimuli depict-

q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap

⇑ Corresponding author. Address: Department of Cognitive Science, Université de Liège, Bd du Rectorat, 5 (Bât. B32), B-4000 Liège, Belgium. Fax: +32 (0)4 3662859. E-mail address: [email protected] (C. Devue). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.007

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

41

ing themselves is a clear and straightforward way to study the neural correlates of the self and of self-awareness. The present review addresses visual self-recognition and its neural correlates. More speciﬁcally, the issues of determining the hemispheric dominance of self-recognition (the lateralization issue) and determining which brain areas are consistently involved during self-recognition (the localization issue) will be examined here. Note that this topic is clearly much more restricted than ‘‘self-processing” in general. The important point of the relationship between visual self-recognition and self-awareness will be addressed in the discussion.

2. The lateralization of visual self-recognition A number of studies compared reaction times for contralateral left or right hand use as a measure of hemispheric dominance of self-face recognition. People tend to respond faster to their own face than other familiar faces, either famous (Miyakoshi, Kanayama, Iidaka, & Ohira, 2010; Tacikowski & Nowicka, 2010) or friend’s faces (Keyes & Brady, 2010; Sugiura et al., 2008; Sui, Liu, & Han, 2009; Sui, Zhu, & Han, 2006) in explicit or incidental face identiﬁcation tasks. However several studies showed that this speed advantage for the self-face only occurs when participants respond with their left hand (Keenan et al., 1999; Ma & Han, 2010; Platek & Gallup, 2002; Platek, Thomson, & Gallup, 2004). Because of contralateral motor control, this ﬁnding was interpreted as reﬂecting right hemisphere dominance of self-face recognition. This left-hand advantage was demonstrated with another dependent measure using an innovative experimental procedure in which the participants’ task was to stop a movie of a morphed face that transitioned between a famous face and self-face, or between a famous face and a co-worker’s face, as soon as they thought that the image looked more like self, or the co-worker, than the famous person (Keenan, Freund, Hamilton, Ganis, & Pascual-Leone, 2000; Keenan, Ganis, Freund, & Pascual-Leone, 2000). In these studies participants stopped the ‘‘famous to self” movie sooner when responding with the left hand than when responding with the right hand. Such hand difference did not occur for the ‘‘famous to co-worker” movie. In other words, participants were more sensitive to self when they responded with their left hand than when they did it with the right hand. Although the left-hand advantage seems to provide a ﬁrst support for a preferential role of the right hemisphere in processing the self-face, other behavioral studies reported data supporting a left hemisphere bias for self-recognition. In these studies, participants were presented with composite faces of themselves and a friend. They were asked to choose which of two symmetric self-faces (one made from the left half and one made of the right half) looked more like themselves. They preferentially chose the composite made of the right half face, i.e. the half face that lies in their right visual ﬁeld when they look at themselves in the mirror. When asked to choose which symmetric face was more representative of their friend, participants chose the composite made of the right half of their friend’ face. This choice represents the opposite bias since that half face lies in their left visual ﬁeld when they look at their friend (Brady, Campbell, & Flaherty, 2004). These results suggest that the left hemisphere is dominant for self-recognition, and the right hemisphere is dominant for the recognition of other familiar faces. The study of split-brain patients also provided important data for the assessment of brain lateralization of self-recognition. Sperry, Zaidel, and Zaidel (1979) examined two split-brain patients and reported that both hemispheres were capable of self-recognition. More recently, several studies used the technique of presenting split-brain patients with varying levels of images of self and familiar faces being morphed together (Turk et al., 2002), or levels of self and familiar faces each being morphed with another face (Uddin, Rayman, & Zaidel, 2005b). Stimuli were presented laterally to each hemisphere. The patients’ task was to choose whether a given image portrayed her/himself or a familiar person. Turk et al. (2002) found that the proportion of either self or other responses increased, until becoming excellent, as the images approached completeness (i.e. 100% of self or familiar face in the morph). This ﬁnding suggests that both hemispheres are capable of self-recognition and familiar face recognition. These authors also found that the patient’s proportion of positive responses while he judged whether or not the image was self was signiﬁcantly higher when the images were presented to the left than to the right hemisphere. The opposite bias was observed when the task was to determine whether the presented face was that of a familiar person. These results support the view that the left hemisphere plays a dominant role in self-recognition. Uddin et al. (2005b) also examined a split-brain patient’s performances using a similar procedure but they obtained quite different results. They also found that both hemispheres are capable of self-recognition. Their results, contrary to those reported by Turk et al. (2002), showed no indication of a hemispheric specialization for self-recognition. Moreover, this patient was able to recognize the familiar face with her right hemisphere only. The same kind of task of self/other judgment from morphs was also used with another split-brain patient (Keenan, Wheeler, Platek, Lardi, & Lassonde, 2003). In that study, stimuli were presented centrally and the patient carried out the self-search and the familiar other search task with either the left or the right hand. In the self-search condition, the proportion of correct identiﬁcations was higher and the proportion of false positives was lower when responding with left hand than with the right hand. No hand advantage occurred when searching for the familiar face. These results are consistent with the hypothesis of right hemisphere dominance of self-recognition. Keenan and colleagues drew the same conclusion from another study in which patients with intractable epilepsy underwent an intracarotid amobarbital procedure (Wada test) consisting in anaesthetizing one cerebral hemisphere at a time (Keenan, Nelson, O’Connor, & Pascual-Leone, 2001). During anaesthesia, ﬁve patients were presented with a picture of their own face morphed with a celebrity’s face (e.g. Marilyn Monroe). After they recovered from anaesthetization, patients were given a force-choice task in which they had to choose the picture they had been shown earlier. After anaesthetization of the right

42

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

hemisphere patients were more likely to choose the famous face. By contrast, when the left hemisphere had been anaesthetized (and the right hemisphere was still active), all patients chose their own face. Hemispheric biases for self-face recognition have also been studied with repeated transcranial magnetic stimulation (rTMS). This technique allows the creation of virtual lesions in chosen regions and an assessment of their effects. Uddin, Molnar-Szakacs, Zaidel, and Iacobini (2006) demonstrated an implication of the right (but not left) inferior parietal lobule during self-face recognition. Indeed, an inhibition of this region decreased the sensitivity of participants to detect their own face among morphed images of themselves and another highly familiar person. More recently, Heinisch, Dinse, Tegenthoff, Juckel, and Brüne (2010) showed that the right temporo-parietal junction was important in self-other distinction and that the prefrontal structures are involved in self-evaluation. Further evidence of a crucial intervention of the right hemisphere in self-face recognition comes from studies of patients with a deﬁcit which involves misidentiﬁcation of one’s own mirror reﬂection as another person. Extensive right hemisphere damage is usually observed in patients suffering from such mirrored-self misidentiﬁcation (Breen, Caine, & Coltheart, 2001; Feinberg & Keenan, 2005). A conclusion that is obvious from studies reported in the present section is that hemispheric dominance for self-face recognition has been a matter of controversy. However an important result reported in all studies of split-brain patients is that both cerebral hemispheres are capable of self-face recognition (Keenan, Wheeler, Platek, Lardi, & Lassonde, 2003b; Sperry et al., 1979; Turk et al., 2002; Uddin et al. (2005b)). However, as far as hemispheric dominance of self-face recognition is concerned, the conclusions of these studies are contradictory: Turk et al. (2002) concluded that the left hemisphere is dominant, Keenan et al. (2003b) claimed that the right hemisphere is dominant, and Uddin et al. (2005b) found no hemispheric dominance. Besides these studies of split-brain patients, the observation of a left-hand advantage observed in self-recognition tasks only, the results from rTMS studies, as well as the preferential interpretation of a morph being self-face after anaesthetization of the left but not the right hemisphere support the hypothesis of right hemisphere dominance. By contrast, results from Brady et al.’s (2004) experiments with composite faces are more consistent with a left hemisphere hypothesis. In the next section, we will address the issue of which brain areas, in the left or the right hemisphere, are selectively recruited in self-face recognition.

3. Neuroimaging studies A summary of the main characteristics of the methods used in the studies reviewed here can be found in Table 1. In previous studies, the cerebral regions involved during self-face recognition have been compared to those involved during the processing of faces varying in familiarity, namely from unfamiliar faces, to famous or personally familiar faces (i.e. coworker, friend, partner, mother, sibling). The studies using familiar faces as control offer a better guarantee that activity found with the self-face is due to its self-relevance and not to its familiarity. Therefore we present results obtained with familiar and unfamiliar control faces separately. Fig. 1 shows the percentage of activation reported in the different lobes in all studies comparing the neural correlates of the self-face to that of another face. When the same region within a lobe was activated several times using different contrasts in the same study (e.g. when different types of visual material such as pictures and movies or different types of tasks such as passive or active viewing were used), we only counted it as one ‘‘hit” on condition that the various contrasts included the same category of face of comparison (i.e. familiar versus unfamiliar). In other words, if the same region was found in a contrast comparing the self-face to another familiar face and another comparing the self-face to an unfamiliar face in the same study, the number of ‘‘hits” for that region would equal 2 in the ﬁrst panel of that ﬁgure. Fig. 1b is based on peaks of activation reported during the processing of the self-face by comparison with other highly familiar faces (famous or personally familiar) and Fig. 1c on regions activated when the self-face processing was compared with that of unfamiliar faces. Fig. 1a conﬁrms the view obtained from the examination of behavioral performance in the previous section: both hemispheres seem to be implied during self-face recognition, with an advantage for the right one, particularly in the frontal and in the parietal lobes. Fig. 1b indicates that the self-referential aspects of self-face processing seem to involve the right hemisphere even more preferentially because the contribution of the frontal and parietal lobe is then much more lateralized than when the self-face is compared with unfamiliar faces (see Fig. 1c). Table 2 shows the peaks of activations reported in previous studies and gives a more precise idea of the speciﬁc regions involved during self-face processing by comparison with that of familiar and unfamiliar faces. Hereafter, we try to sketch the respective role of the regions implied during self-recognition. We adopted the criterion used by Platek, Wathne, Tierney, and Thomson (2008) in discussing only regions activated in ﬁve or more studies in total. 3.1. Prefrontal cortex 3.1.1. Inferior frontal gyrus Activation of the inferior frontal cortex have been reported in six studies comparing the processing of the self-face to that of a familiar face (Devue et al., 2007; Kaplan, Aziz-Zadeh, Uddin, & Iacoboni, 2008; Platek, Keenan, Gallup, & Mohamed, 2004; Sugiura et al., 2006, 2008; Uddin, Kaplan, Molnar-Szakacs, Zaidel, & Iacoboni, 2005a), all on the right side, and in three studies comparing the processing of the self-face to that of unfamiliar faces, one on the right side (Sugiura et al., 2000) and two on

43

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51 Table 1 Summary of methods used in previous neuroimaging studies. Study

Aim

Technique

N

Stimuli used

Task

Self related analysis (faces of comparison)

Sugiura et al. (2000)

Self-face recognition

PET-SRC

9,7

Pictures of faces

Self vs. unfamiliar

Kircher et al. (2000)

Self-judgments

fMRI

6

Head orientation judgment – passive active recognition Identiﬁcation of faces – evaluation of adjectives

Kircher et al. (2001)

Self-face recognition

fMRI

6

Change of identity detection

Self vs. unfamiliar

Uddin et al. (2005a) Platek et al. (2006)

Self-face recognition Self-judgments (faces)

fMRI

10

Sensitivity judgment

Self vs. familiar (friend)

fMRI

5

Self vs. familial (famous)

Sugiura et al. (2005)

Self-face recognition

fMRI

34

Pictures of faces

Think about the person depicted OR her mental state Familiarity judgment

Sugiura et al. (2006) Platek et al. (2006)

Self-face body recognition Self-face recognition Self-face body recognition

fMRI

42

Familiarity judgment

fMRI

12

Pictures/movies of faces bodies Pictures of faces

fMRI

20

‘‘Intact/altered” judgment

Sui and Han (2007) Morita et al. (2008)

Self-construal priming Self-face evaluation

fMRI

12

Altered/intact pictures of faces bodies Pictures of faces

fMRI

19

Sugiura et al. (2008) Kaplan et al. (2008) Platek et al. (2009)

Self-face/name recognition Self-face/voice recognition Self resemblance

fMRI

35

fMRI

12

fMRI

Hodzic et al. (2009) Taylor et al. (2009)

Self-body recognition Self-face recognition Self-face recognition Self similarity

Devue et al. (2007)

Platek and Kemp (2009) Verosky and Todorov (2010)

Morphed pictures – self-descriptive words Movie of a face changing into another Morphed pictures – scrambled Pictures of faces

Identiﬁcation task

Self vs. unfamiliar

Self vs. familiar (friend + researcher, familiarised) – self vs. unfamiliar Self vs. familiar (friend) masked by familiar vs. unfamiliar Self vs. familiar (fraternity brother) – self vs. unfamiliar Self vs. familiar (friend colleague)

Head orientation judgment Photogenic judgment – embarrassment ratings

Self vs. familiar (colleague) vs. scrambled face Self vs. unfamiliar

Familiarity judgment

Self vs. familiar (friend)

‘‘Self/other” judgment

Self vs. familiar (friend/colleague)

15

Pictures of faces – voice recording Morphed pictures

Race judgment

fMRI

10

Pictures of bodies

Familiarity judgment

Amygdala activity for four types effaces Self vs. unfamiliar

fMRI

10

Pictures of faces

Passive viewing

Self vs. unfamiliar

fMRI

12

Morphed pictures

Identiﬁcation talk

Self vs. unfamiliar

fMRI

30

Morphed pictures

Sensitivity–judgment

Activation to faces more or less similar to the self-face

Bad/good pictures extracted from videos Faces/names

the left side (Kircher et al., 2000, 2001). In one study, the processing of the self-body by comparison with that of a familiar body also elicited activity in the inferior frontal gyrus, but on the left side (Devue et al., 2007). The right inferior frontal cortex has been hypothesised to be implied in self-other differentiation (Devue et al., 2007; Uddin et al., 2005a) and in the attention to the representation of one’s own face (Sugiura et al., 2000). For instance, this region reacts to morphed faces that contain more ‘‘self” (Uddin et al., 2005a) and to an intact picture of the self-face during a task requiring to discriminate intact from altered pictures of the self-face and another highly familiar face (Devue et al., 2007). A study presenting participants with pictures of their own face but also with recordings of their own voice showed activity in the inferior frontal gyrus with the two kinds of stimuli suggesting that this structure processes self-referential stimuli of different modalities and that it could contribute to an abstract representation of the self (Kaplan et al., 2008). Recent studies also indicate that this structure might be involved in evaluative judgments about the own face (Kita et al., 2010; Morita et al., 2008; see also the discussion of the present paper). 3.1.2. Medial and middle frontal gyrus Two studies comparing self-face processing to a familiar face processing found activity in the medial frontal gyrus, one on the right (Platek et al., 2006) and one on the left (Platek & Kemp, 2009). With unfamiliar control faces, two studies reported activity in the right (Platek & Kemp, 2009; Sugiura et al., 2000) and one bilateral activity (Taylor et al., 2009) in the medial frontal gyrus. In a study comparing recognition of the self to recognition of personally familiar faces related (i.e. sibling) or not to the self (i.e. friend), Platek and Kemp (2009) suggested that this region is implied, along with the anterior cingulate gyrus, in complex forms of distinction between self and other (e.g. kinship) and related decisions towards the stimuli (e.g. cooperation vs. aversion).

44

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

Percentage of reported activations

(a)

Self-face vs. other faces (overall) 35

Percentage of reported activations

Right

25 20 15 10 5 0

(b)

..

Self-face vs. familiar faces 35

Left

30

Right

25 20 15 10 5 0

(c) Percentage of reported activations

Left

30

Self-face vs. unfamiliar faces 35 30 25

Left Right

20 15 10 5 0

Fig. 1. Percentage of activation reported in cerebral areas in each hemisphere on all activations reported in previous neuroimaging studies: (a) all studies comparing self-face processing to another face processing collapsed; (b) in studies comparing self-face processing to familiar face processing; (c) in studies comparing self-face processing to unfamiliar face processing.

The middle frontal gyrus was activated on the right side in two studies contrasting self-face and familiar faces (Platek et al., 2004, 2006). When using unfamiliar control faces, three studies reported activity on the left side (Kircher et al., 2000, 2001; Taylor et al., 2009), one on the right side (Platek & Kemp, 2009) and one bilaterally (Sugiura et al., 2000). In their recent meta-analysis of self-face recognition Platek et al. (2008) asked the question of the origin of the activity in this region which is also often associated with the empathic processing of pain, whereas most self-recognition studies used neutral faces. They suggest that ‘‘neutral faces may possess a level of affective information that is not yet understood” and that activity reported in ‘‘left and/or right middle frontal gyrus (. . .) may be a consequence of variations in affective nature of the respective neutral faces” (p. 179). Recent studies introducing tasks that involve more evaluative and social aspects besides the classical perceptual task of distinguishing between self and other might help answering that question. Platek, Krill, and Wilson (2009) showed that activity in the ventral inferior, middle and medial frontal gyri was related to trustworthiness ratings of faces resembling oneself. Moreover, a study by Sui and Han (2007) involving Chinese participants showed that activity in the right middle frontal gyrus increased when they viewed their own face after being primed with an independent pronoun (e.g. ‘‘I”, ‘‘mine”) compared to when they were primed with an interdependent pronoun (e.g. ‘‘we”, ‘‘ours”). The authors suggest that stressing the independence of oneself increases the self-other distinction. However, the study by Sugiura et al. (2008) in which they presented their participants with faces and names suggests that the activity in the medial cortical structures is not relevant to distinguish self from others during face or name recognition. These regions thus seem to play a role in the distinction between self and other on a level involving evaluation and where the self is embedded within a social context rather than on a simpler perceptual level. 3.2. Insula Four studies comparing the processing of the self-face to that of unfamiliar faces reported activity in the insula, mainly of its anterior part, two on the right side (Kircher et al., 2000, 2001) and two on the left side (Morita et al., 2008; Sugiura et al.,

45

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

Table 2 Number of cerebral areas reported in neuroimaging studies of self-face recognition as a function of the type of contrast used (self-face vs. familiar faces and selfface vs. unfamiliar faces). Left

Right

Total

8 1 0

33 14 5

41 15 5

Mid-inferior frontal gyrus Middle frontal gyrus Medial frontal gyrus Superior frontal gyrus Precentral sulcus

0 0 1 0 0

1 2 1 3 2

1 2 2 3 2

Devue et al. (2007), Kaplan et al. (2008), Platek et al. (2004), Sugiura et al. (2008), Uddin et al. (2005a) Sugiura et al. (2006) Platek et al. (2004, 2006) Platek and Kemp (2009), Platek et al. (2006) Platek et al. (2004, 2006), Platek and Kemp (2009) Sugiura et al. (2006, 2008)

Inter lobes Occipito-parietal junction

0 0

1 1

1 1

Sugiura et al. (2006)

Limbic Anterior cingulate cortex Anterior cingulate/paracingulate Parahippocampal gyrus

1 1 0 0

3 1 1 1

4 2 1 1

Devue et al. (2007), Platek and Kemp (2009) Platek and Kemp (2009) Platek and Kemp (2009)

Occipital Inferior occipital gyrus

0 0

2 2

2 2

Kaplan et al. (2008), Uddin et al. (2005a)

Parietal Intraparietal sulcus Inferior parietal lobule Superior parietal Posterior superior parietal lobule Precuneus Supramarginal gyrus

3 0 0 1 0 1 1

8 1 3 1 1 0 2

11 1 3 2 1 1 3

Sub-lobar Anterior insula Insula

0 0 0

2 1 1

2 1 1

Devue et al. (2007) Devue et al. (2007)

Temporal Fusiform gyrus/inferior temporal gyrus Inferior temporal gyrus Middle temporal gyrus Superior temporal gyrus

3 1

3 1

6 2

Sugiura et al. (2006)

1 1 0

1 0 1

2 1 1

Sugiura et al. (2008) Platek et al. (2006) Platek and Kemp (2009)

Self-face vs. familiar faces Frontal Inferior frontal gyrus

Self-face vs. unfamiliar faces

References

Sugiura et al. (2006) Kaplan et al. (2008), Platek et al. (2006), Uddin et al. (2005a) Platek and Kemp (2009), Uddin et al. (2005a) Sugiura et al. (2006) Platek and Kemp (2009) Platek and Kemp (2009), Sugiura et al. (2006, 2008)

32

48

80

Cerebellum Cerebellum

1 1

2 2

3 3

Frontal Frontal operculum Inferior frontal gyrus Inferior frontal gyrus/DLPFC Middle frontal gyrus

7 0 1 1 4

8 1 1 0 2

15 1 2 1 6

1 0

3 1

4 1

Sugiura et al. (2005) Kircher et al. (2000), Sugiura et al. (2000) Kircher et al. (2001) Kircher et al. (2000, 2001), Platek and Kemp (2009), Sugiura et al. (2000), Taylor et al. (2009) Platek and Kemp (2009), Sugiura et al. (2000), Taylor et al. (2009) Morita et al. (2008)

Inter lobes Occipito-temporo-parietal junction Postcingulate cortex/ parahippocampal gyrus Mid-inferior frontal gyrus/insula

1 0 1

4 2 1

5 2 2

Morita et al. (2008), Sugiura et al. (2005) Sugiura et al. (2005)

0

1

1

Morita et al. (2008)

Limbic Anterior cingulate cortex Anterior cingulate gyrus Anterior cingulate sulcus Cingulate cortex Hippocampal formation

3 1 1 0 1 0

9 4 1 1 1 2

12 5 2 1 2 2

Occipital Anterior occipital cortex Inferior occipital gyrus Occipital cortex Posterior occipital cortex Cuneus

4 0 1 1 0 1

5 1 1 0 1 1

9 1 2 1 1 2

Medial frontal gyrus Precentral gyrus

Kircher et al. (2000, 2001), Platek et al. (2006)

Kircher et al. (2000, 2001), Morita et al. (2008), Taylor et al. (2009) Sugiura et al. (2000) Sugiura et al. (2000) Taylor et al. (2009) Kircher et al. (2000, 2001) Morita et al. (2008) Taylor et al. (2009) Morita et al. (2008) Morita et al. (2008) Taylor et al. (2009) (continued on next page)

46

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

Table 2 (continued) Left Lingual gyrus

Right

Total

References

1

1

2

Taylor et al. (2009)

Parietal Postcentral gyrus Precuneus Superior parietal Supramarginal gyrus Supramarginal gyrus/Inferior parietal lobe

3 0 1 0 0 2

8 1 3 1 3 0

11 1 4 1 3 2

Platek et al. (2006) Kircher et al. (2000, 2001), Sugiura et al. (2000), Taylor et al. (2009) Sugiura et al. (2000) Platek et al. (2006), Platek and Kemp (2009), Sugiura et al. (2000) Kircher et al. (2000, 2001)

Sub-lobar Anterior insula Insula Anterior and mid-posterior insula Caudate nucleus Globus pallidus Hypothalamus Lenticular/subthalamic nucleus Lentiform nucleus Midbrain Pulvinar Putamen Thalamus

6 1 1 0 1 0 0 0 1 0 1 1 0

10 0 0 2 0 1 1 2 1 1 1 0 1

16 1 1 2 1 1 1 2 2 1 2 1 1

Sugiura et al. (2000) Morita et al. (2008) Kircher et al. (2000, 2001) Sugiura et al. (2000) Sugiura et al. (2000) Sugiura et al. (2000) Kircher et al. (2000, 2001) Platek et al. (2006) Sugiura et al. (2005) Sugiura et al. (2000) Sugiura et al. (2000) Platek et al. (2006)

Temporal Fusiform gyrus Superior temporal gyrus

7 5 2

2 1 1

9 6 3

Kircher et al. (2000, 2001), Sugiura et al. (2000, 2005), Taylor et al. (2009) Kircher et al. (2000, 2001), Platek et al. (2006)

2000). Morita et al. (2008) suggest that activity in the insula reﬂects an automatic arousal resulting from self-face recognition. Our study using familiar controls also found an implication of the right anterior insula (Devue et al., 2007) during self-face as well as during self-body processing and when collapsing data from self-face and self-body processing. This leads to the idea that this region might be involved in an integrative self-processing, that is, independent of the stimuli used. This idea is supported by the study by Kircher et al. (2000) that showed common activation in the right insula with the self-face and self-descriptive adjectives. In line with this hypothesis, Platek, Keenan, and Mohamed (2005) suggested that the insula plays a role in making decisions about self-referential information. Indeed, the insula has also been found to be implicated in different aspects of self-processing such as self-agency (for a recent review, see Karnath & Baier, 2010) or autobiographical episodic memory retrieval (Fink et al., 1996). 3.3. Cingulate cortex Similarly, the cingulate gyrus has been suggested to be involved during abstract self-processing (Devue et al., 2007; Northoff & Bermpohl, 2004). Involvement of different parts of the cingulate cortex have been consistently reported during self-face recognition by comparison with familiar faces recognition, namely, the anterior part in the right (Devue et al., 2007; Platek & Kemp, 2009) and a part extending to the paracingulate in the left (Platek & Kemp, 2009) hemisphere. Studies comparing the self-face to unfamiliar faces also found implication of the right (Kircher et al., 2000, 2001; Morita et al., 2008) or bilateral (Sugiura et al., 2000; Taylor et al., 2009) cingulate cortex, again mainly of its anterior part. As with the insula, our study involving self-face and self-body processing showed activity in the right anterior cingulate with each kind of stimuli but also when data from the two types of stimuli were collapsed (Devue et al., 2007). This region thus also seems to play a role in the integration of information about oneself independently of the stimulus domain. This is in agreement with the claim that this structure might be generally involved during abstract self-processing (i.e., independent of the stimulus domain or of the sensorial modality) or when making decisions about self-referential information (Northoff & Bermpohl, 2004; Platek et al., 2005). 3.4. Temporal cortex 3.4.1. Fusiform gyrus and inferior temporal gyrus Activations of the fusiform gyrus have mainly been reported in the left hemisphere and when comparing the processing of the self-face to that of unfamiliar faces (Kircher et al., 2000, 2001; Sugiura et al., 2000, 2005) but in one case, activity was bilateral (Taylor et al., 2009). When the self-face was compared to familiar faces, but also when stimuli representing the self-body were used during a familiarity judgment task, the activity in the fusiform gyrus extended to the inferior part of the temporal gyrus, bilaterally (Sugiura et al., 2006). Implication of the bilateral inferior temporal gyrus was also found in another study using a familiarity judgment task (Sugiura et al., 2008). In sum, these ﬁndings indicate that the fusiform gyrus and the inferior temporal gyrus would not differentiate the self-face from other faces in terms of self properties but rather that they would allow a ﬁrst access to the familiarity of a face. Increased activity for the self-face has also been reported in

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

47

bilateral occipital cortices extending to the inferior temporal gyri and to the fusiform gyrus (Morita et al., 2008) and was interpreted as reﬂecting a higher attention to the emotional salience of one’s own face. 3.5. Parietal cortex 3.5.1. Inferior parietal lobule An implication of the right inferior parietal lobule has been reported in three studies contrasting the processing of the self-face to that of a familiar face (Kaplan et al., 2008; Platek et al., 2006; Uddin et al., 2005a). In two studies using unfamiliar faces as control, the activity in the left inferior parietal lobule extended in the supramarginal gyrus (Kircher et al., 2000, 2001). It was suggested that activity in this region might reﬂect the representation of the self-face as part of a more general awareness of the self-body (Platek et al., 2006; Uddin et al., 2005a) and a self-other discrimination across different sensorial modalities (Uddin et al., 2005a). A study showing speciﬁc activation in the right inferior parietal lobe while viewing one’s own body compared to a familiar body supports this idea (Hodzic, Kaas, Muckli, Stirn, & Singer, 2009). 3.5.2. Supramarginal gyrus According to three studies, the supramarginal gyrus is implied during self-face recognition by comparison with familiar faces recognition, on the right (Sugiura et al., 2006, 2008) but also on the left (Platek & Kemp, 2009). An implication of this area has also been found when contrasting the self-face with unfamiliar faces, on the right (Platek & Kemp, 2009; Platek et al., 2006; Sugiura et al., 2000) and on the left (Kircher et al., 2000, 2001). Based on evidence that this region is damaged in patients showing asomatognosia, some authors have suggested that this region is involved in representing the own face as part of the own body (Sugiura et al., 2000; see also Platek et al., 2006). 3.5.3. Precuneus The precuneus was activated in four studies comparing the self-face processing to that of unfamiliar faces, in three cases on the right (Kircher et al., 2000, 2001; Sugiura et al., 2000) and in one on the left (Taylor et al., 2009). One study using familiar control faces also reported activity in the left precuneus (Platek & Kemp, 2009). As noted in two recent meta-analyses (Northoff et al., 2006; Platek et al., 2008), this region reacts to different kinds of self-referential stimuli and might be devoted to the integration of different types of self-processing (see also Uddin, Iacobini, Lange, & Keenan, 2007).

4. Discussion 4.1. Summary and functions of the self-related areas Results from behavioral studies, examination of split-brain patients, TMS and neuroimaging studies point out to a high implication of the right hemisphere but also show clear evidence of an implication of both hemispheres during self-recognition (see Fig. 1b), consistent with the idea of a bilateral network for self-recognition (Kircher et al., 2001; Sugiura et al., 2005). Searching for the cerebral areas involved during self-recognition from neuroimaging studies, some areas seem to emerge consistently. For instance, there is an extensive implication of the prefrontal and parietal cortices (see also Uddin et al., 2005a, 2007). More speciﬁcally, within the prefrontal cortex, the inferior frontal gyrus, the medial and middle frontal gyri are often implicated, mostly on the right side. Within the parietal cortex, there is an involvement of the inferior parietal lobule, the supramarginal gyrus and the precuneus, also mainly in the right hemisphere. In addition, other regions such as the anterior cingulate cortex (mainly on the right), the bilateral insula, the fusiform gyrus (mainly on the left), and the bilateral inferior temporal gyrus seem highly involved during self-recognition. To date, the deﬁnition of the role of each speciﬁc region remains highly hypothetical. Indeed, so far the processing of the self-face has been compared to that of very different ‘‘other” faces (from unfamiliar to highly familiar faces) and by means of a variety of tasks. Each of these tasks probably involved a variety of processes, going from a perceptual analysis of facial features allowing face recognition to more unpredictable subjective evaluations and emotions triggered by the vision of the face. Nonetheless, hereafter we will attempt to provide a synthesis of the functions of the neural correlates of self-recognition as proposed by reviewed studies. The activations in the most posterior parts of the brain, such as in the occipital cortices (Kaplan et al., 2008; Uddin et al., 2005a) presumably reﬂect perceptual processing. A recent ERP study has also shown that the self-face can already be differentiated from other familiar or unfamiliar faces on the N170 component over occipital regions (Keyes, Brady, Reilly, & Foxe, 2010). This early component is supposed to reﬂect structural encoding of faces. This occipital activity extends to the inferior temporal gyri and the fusiform gyrus (e.g. Morita et al., 2008) which might result from a ﬁrst differentiation between faces of different levels of familiarity (Devue et al., 2007) before the identiﬁcation of the faces. Indeed, activity in the left fusiform gyrus was mainly found when comparing the self-face to unfamiliar faces (Kircher et al., 2000, 2001; Sugiura et al., 2000, 2005; Taylor et al., 2009). One might wonder why perceptual analysis of one’s own face would differ from that of other faces. These self-speciﬁc posterior activations might result from the fact that we do not process the same kind of structural

48

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

information when seeing our own face and other faces because processing goals are different in each case. The aim of processing another person’s face is typically to identify that person or to interpret her emotional expressions whereas the goal of processing our own face is normally not identiﬁcation but an inspection of facial features, for instance when grooming (Greenberg & Goshen-Gottstein, 2009; Keyes & Brady, 2010). Regions of the parietal cortex such as the inferior parietal lobule and the supramarginal gyrus have often been attributed a role in terms of spatial representation of oneself (e.g. Platek et al., 2006; Sugiura et al., 2000; Uddin et al., 2005a). The supramarginal gyrus, along with the occipito-parietal junction might be part of an occipito-parietal network (dorsal pathway) implied in a visuo-spatial representation of oneself, in other words, with the self being represented as an object in the space and with the self-face being represented as a part of the own body (Platek et al., 2006; Sugiura et al., 2000; Uddin et al., 2005a). There would be direct connections between the parietal and the frontal cortices during self-recognition. In their recent review about the self and social cognition, Uddin et al. (2007) suggested that right frontoparietal areas activated during self-recognition overlap with the mirror neuron system, comprising the inferior frontal cortex and the inferior parietal lobule. They posit that this network is involved in representing different aspects of the physical self, and in relating the physical self to others through motor simulation mechanisms. During self-recognition, one would process the perceived physical self (e.g. the self-face) using a similar mechanism. The perceived self would be mapped onto the perceiver’s own motor repertoire. Uddin et al. (2007) suggested that a second network involving the midline cortical structures and overlapping with some areas of a ‘‘default-mode” network including the ventral and dorsal prefrontal cortex, the precuneus and posterior lateral cortices would play a role in more abstract self- and other-related processing, in terms of mental states attribution and evaluation. Activations within the frontal cortex have been attributed to more complex forms of self-other differentiation and selfevaluation and this region would be of particular importance for social cognition. For instance, the right inferior frontal gyrus would contribute to complex perceptual judgments in terms of comparison to an ‘‘ideal” self (Morita et al., 2008). Its activation might also correspond to the sustained attention to one’s own face (Sugiura et al., 2000) which might itself allow or reﬂect evaluative processes (Kita et al., 2010; Morita et al., 2008; see also Heinisch et al., 2010). Accordingly, ERP studies indicate that the self-face is a particularly salient stimulus because it elicits ampler P300 than other faces (Miyakoshi, Nomura, & Ohira, 2007; Ninomiya, Onitsuka, Chen, Sato, & Tashiro, 1998). Similarly the medial frontal gyrus would allow the detection of self-resemblance in terms of kinship along with the cingulate cortex (Platek & Kemp, 2009) or during judgment of trustworthiness along with the ventral inferior and with the middle frontal gyri (Platek et al., 2009; but see Verosky & Todorov, 2010). Various regions within the frontal cortex as well as the insula and the cingulate cortex have also been hypothesized to play a role in the construction of an abstract representation of oneself (i.e. a sense of ‘‘me”), independently of the type of stimulus presented or of the sensorial modality stimulated (see e.g. Devue et al., 2007; Kaplan et al., 2008; Kircher et al., 2000, 2001). According to Uddin et al. (2007), because it is part of the cortical midline structures network and because it has direct connections with the posterior component of the mirror neurons system (i.e. the inferior parietal lobule), the precuneus would allow interactions between these two networks and would be of high importance in the elaboration of integrated information about oneself. There are thus at least four candidate regions for processing self-related stimuli at a higher, integrated level, namely, the inferior frontal gyrus (Kaplan et al., 2008), the anterior insula (Devue et al., 2007; Kircher et al., 2000), the anterior cingulate cortex (Devue et al., 2007; Northoff & Bermpohl, 2004; Platek et al., 2005) and the precuneus (Northoff et al., 2006; Platek et al., 2008; Uddin et al., 2007). However, this idea of integration might be confounded with other kinds of responses that could be common to different self-related material, such as evaluative judgments or emotional reactions. The notion of general ‘‘decision making about oneself” used by some authors could correspond to such responses. For instance, the task used in the Devue et al. (2007) study involved detection of alterations affecting the faces (interocular distance) or the bodies (waist to hip ratio) of the participant herself or of her colleague. We hypothesized that activity found in the insula and the cingulate cortex reﬂected an abstract representation of oneself but it might actually originate from subjective judgments (e.g. evaluation) or a subsequent emotional reaction in response to one’s own pictures. The same can be said from inferior frontal gyrus activation following the presentation of the own face and of the own voice in the study by Kaplan et al. (2008). This activity might reﬂect some subjective judgments emitted by people when seeing their own face (e.g. ‘‘I look tired on that picture”) and when hearing their own voice (e.g. ‘‘I sound funny”), judgments that might be followed by some emotional reactions (e.g. feeling ashamed), rather than an abstract integration of information. As a consequence, the exact role of these ‘‘higher-level” regions still needs to be clariﬁed. When considering self-processing in more general terms, Northoff and Bermpohl (2004) suggested that integration processes would take place in the posterior cingulate whereas evaluative processes would be subtended by the dorsolateral prefrontal cortex. Regarding self-recognition, the distinction between the integration of the physical aspects of the self and the evaluative judgments about these physical aspects is very difﬁcult. Indeed, it is likely that even if the task used only requires a perceptual judgment, participants also automatically emit speciﬁc judgments about themselves which in turn triggers activity linked to these processes. Researchers in this ﬁeld probably all know that participants in their study are almost never indifferent about the way they look on the pictures (see e.g. Kita et al., 2010). The next step in understanding the role of selfspeciﬁc regions is thus to take more subjective and emotional variables and the variety of the cognitive processes involved when looking at one’s own face into account. We will now consider the more speciﬁc issue of the relationship between selfface recognition and self-awareness.

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

49

4.2. Relation between self-recognition and self-awareness Does self-face recognition equal self-awareness? There is no consensual answer to that question. For Keenan and his colleagues self-awareness includes knowing that one is a separate entity. Self-awareness also involves being aware of one’s thoughts, or mental state, and encompasses the ability to imagine oneself in the future or in the past (e.g. Keenan, Gallup, & Falk, 2003a; Keenan, Rubio, Racioppi, Johnson, & Barnacz, 2005). These researchers argued that self-face recognition appears to be related to high-order awareness and that ‘‘the self-face may be an ideal stimulus to begin the investigations of higherorder consciousness and the brain” (e.g. Keenan et al., 2005, p. 697). By contrast, other researchers considered that self-face recognition does not imply such a sophisticated kind of self-awareness. In several papers, Mitchell (1993) and Morin (2002, 2007) admitted that self- recognition requires self-attention and a certain kind of self-knowledge. However, according to these authors, the kind of self-information that is required for recognizing oneself in a mirror is a kinesthesic representation of one’s own body. Brieﬂy speaking, the participant (in the context of these papers, participants of interest were usually chimpanzees or infants) matches the representation of the physical self with the reﬂection seen in the mirror and concludes ‘‘it’s me” (Morin, 2007, p. 1068). This argumentation is relevant as far as self-mirror recognition is concerned. In the studies reviewed in the present paper, participants were not placed in front of a mirror but were presented with pictures. Therefore, kinesthesic information was not available. We think that, like self-mirror recognition, self-face recognition from pictures does not require access to one’s own mental states or thoughts. According to models of face recognition the intervention of such high-level cognitive processes is not necessary for recognizing a face. Face recognition requires matching the current representation of the seen face’s surface structure with a stored representation in the perceptual memory system devoted to faces (Bruce & Young, 1986). After a face has been recognized, identity- speciﬁc semantic information about the person may be retrieved. However, the fact that face recognition does not require high-level cognitive processes does not mean that recognizing a face may not be accompanied with such more complex processes. For instance, recent research demonstrated that conscious recollection, which is associated with autonoetic awareness (Tulving, 1985), may accompany the recognition of famous (Damjanovic & Hanley, 2007) and personally familiar faces (Barsics & Brédart, in press). It is therefore highly probable that such conscious recollection also accompanies self-recognition in some circumstances. Neuroscientiﬁc research on self-face recognition has focused mostly on localizing its neural substrates rather than on determining cognitive operations induced by the perception of the self-face (Tsakiris, 2008). At this stage of research, it seems necessary to clarify what we are searching for when we use self-face as stimuli in experiments. We mentioned here above that some researchers consider that the self-face may be an ideal stimulus to investigate the brain correlates of higherorder consciousness (Keenan et al., 2003a, 2005). We think that if a study is aimed at determining which brain regions are selectively involved in complex cognitive operations such as mental travel in the past and the future, or mentalizing, then stimuli and procedures that are speciﬁcally designed to tap these functions should be used rather than using a self-face recognition task. A number of works did already resort to such a more targeted strategy for studying the brain correlates of mental travel in the past and future (for a recent review, see Spuznar, 2010) or intention understanding (for a review, see Van Overwalle, 2009). Although a relationship between self-recognition and self-awareness has been assumed in the literature, it is only recently that neuroimaging studies started to examine whether separate brain regions recruited during self-recognition were speciﬁcally associated with behavioral measures of different aspects of self-awareness, for instance, public selfawareness induced by the perception of the self-face vs. evaluation of the self-face toward a standard (see Morita et al., 2008). Perceiving the self-face is likely to direct attention on one’s own appearance, i.e. public self-awareness (Morin, 2006). In turn, focusing on the self may yield self-evaluation, i.e. a comparison of the seen self-face against a mental representation an ideal self-face (for a review see Carver, 2003). If a discrepancy is found between the perceived and the ideal representation of the self-face, then the observer can experience negative emotions or feelings such as embarrassment. Brain correlates of public self-awareness induced in the participant by the perception of her/his own face could be different from those of self-evaluation following self-recognition. Recently, Morita et al. (2008) reported data supporting this hypothesis by scanning participants whose task was to evaluate how photogenic seen faces of self and friends were. After scanning, participants completed a questionnaire evaluating their public self-consciousness (i.e. their tendency to be aware of their own appearance) as well as a task consisting in rating the intensity of the embarrassment they experienced while seeing each presented face. Results showed that the right precentral gyrus was more strongly activated when participants with high public self-consciousness viewed their own faces compared with participants with low public self-consciousness. However, activity in the right precentral gyrus was not correlated with the level of embarrassment experienced by seeing non-photogenic self-pictures. In addition, Morita et al. (2008) showed that, conversely, activity in the right middle inferior frontal gyrus was not related with levels of public self-consciousness but was modulated by the extent of embarrassment (the outcome of self-evaluation). These results demonstrated a functional dissociation between two brain correlates of self-recognition: the activation of the right precentral gyrus is associated with public self-awareness while the right middle inferior frontal gyrus is associated with self-evaluation induced by the perception of one’s own face. 5. Conclusion During the last 10 years a considerable number of studies were aimed at localizing the neural correlates of self-face recognition. The real objective of most of these studies was, in fact, to gain knowledge about self-awareness rather than

50

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

self-face recognition per se. Self-recognition was considered as an indicator of self-awareness (e.g. Feinberg & Keenan, 2005; Keenan et al., 2005; Sugiura et al., 2005; Sui et al., 2006; Uddin et al., 2005a, 2005b; for a critical view see Morin, 2002, 2007), the level of such self-awareness being often left unspeciﬁed. As shown in the present paper and in other recent reviews (e.g. Platek et al., 2008), researchers discovered many different cerebral areas that are associated with self-face recognition. Unfortunately, it remains an extremely difﬁcult and hazardous task to relate speciﬁc cognitive operations induced by self-perception and brain regions recruited. We do not think that understanding the neural correlates of self-face recognition (and some forms of related self-awareness) will signiﬁcantly improve without specifying more clearly what happens when a participant sees her or his face on the screen, and what is speciﬁc to self-recognition in comparison with the recognition of another highly familiar face. Intuitively, the perception of one’s own face should trigger the structural representation of the self-face in the perceptual representational system, and allow self-recognition. This recognition is likely to induce public selfawareness which in turn may yield some kind of self-evaluation. The result of such an evaluation may be accompanied with emotional responses. Similar hypothetical sequences of operations following self-perception have earlier been proposed by others (e.g. Carver, 2003; Morita et al., 2008), however ﬁne-grained empirical evaluation of such proposals is needed. Otherwise, we mentioned earlier that it has been demonstrated that the recognition of familiar faces is prone to produce a retrieval of episodic information (autonoetic awareness). It is highly probable that such memories may accompany self-recognition. We think that one of the major aims of research on the neural correlates of self-recognition should be to identify more clearly the cognitive components induced by the perception of the self-face, and to relate them with the brain regions that have been shown to be associated with self-face recognition. Acknowledgements C.D. is a Postdoctoral Researcher of the Belgian National Fund for Scientiﬁc Research (FRS-FNRS). S.B. was funded by the French-speaking community of Belgium (ARC 06/11-340). References Amsterdam, B. (1972). Mirror self-image reactions before age two. Developmental Psychobiology, 5, 297–305. Barsics, C., & Brédart, S. (in press). Recalling episodic information about personally known faces and voices. Consciousness and Cognition, doi:10.1016/ j.concog.2010.03.008. Bluhm, C., & Clendenin, N. (2009). Someone else’s face in the mirror: Identity and the new science of face transplants. Westport, CT: Praeger. Brady, N., Campbell, M., & Flaherty, M. (2004). My left brain and me: A dissociation in the perception of self and others. Neuropsychologia, 42, 1156–1161. Breen, N., Caine, D., & Coltheart, M. (2001). Mirrored-self misidentiﬁcation: Two cases of focal onset dementia. Neurocase, 7, 239–254. Bruce, V., & Young, A. W. (1986). Understanding face recognition. British Journal of Psychology, 77, 305–327. Carver, C. S. (2003). Self-awareness. In M. R. Leary & J. P. Tangney (Eds.), Handbook of self and identity (pp. 179–196). New York: Guilford Press. Cole, J. (1998). About face. Cambridge, MA: MIT Press. Damjanovic, L., & Hanley, J. R. (2007). Recalling episodic and semantic information about famous faces and voices. Memory & Cognition, 35, 1205–1210. Devue, C., & Brédart, S. (2008). Attention to self-referential stimuli: Can I ignore my own face? Acta Psychologica, 128, 290–297. 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. Feinberg, T. E., & Keenan, J. P. (2005). Where in brain is the self. Consciousness and Cognition, 14, 661–678. 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. Gallup, G. G. (1970). Chimpanzees: Self-recognition. Science, 167, 86–87. Greenberg, S. N., & Goshen-Gottstein, Y. (2009). Not all faces are processed equally: Evidence for featural rather than holistic processing of one’s own face in a face-imaging task. Journal of Experimental Psychology: Learning, Memory, and Cognition, 55, 499–508. Heinisch, C., Dinse, H. R., Tegenthoff, M., Juckel, G., & Brüne, M. (2010). An rTMS study into self-face recognition using video-morphing technique. Social Cognitive and Affective Neuroscience, 5, 356–361. Hodzic, A., Kaas, A., Muckli, L., Stirn, A., & Singer, W. (2009). Distinct cortical networks for the detection and identiﬁcation of human body. NeuroImage, 45, 1264–1271. Kaplan, J. T., Aziz-Zadeh, L., Uddin, L., & Iacoboni, M. (2008). The self across the senses: The neural response to one’s own face and voice. Social, Cognitive, and Affective Neuroscience, 3, 218–223. Karnath, H. O., & Baier, B. (2010). Right insula for our sense of ownership and self-awareness of actions. Brain Structure and Function, 214, 411–417. Keenan, J. P., Freund, S., Hamilton, R. H., Ganis, G., & Pascual-Leone, A. (2000). Hand response differences in a self-face identiﬁcation task. Neuropsychologia, 38, 1047–1053. Keenan, J. P., Gallup, G. G., & Falk, D. (2003a). The face in the mirror: The search for the origins of consciousness. New York: HarperCollins/Ecco. Keenan, J. P., Ganis, G., Freund, S., & Pascual-Leone, A. (2000). Self-face identiﬁcation is increased with left hand responses. Laterality, 5, 259–268. Keenan, J. P., McCutcheon, B., Freund, S., Gallup, G. G., Sanders, G., & Pascual-Leone, A. (1999). Left hand advantage in a self-face recognition task. Neuropsychologia, 37, 1421–1425. Keenan, J. P., Nelson, A., O’Connor, M., & Pascual-Leone, A. (2001). Self-recognition and the right hemisphere. Nature, 409, 305. Keenan, J. P., Rubio, J., Racioppi, C., Johnson, A., & Barnacz, A. (2005). The right hemisphere and the dark side of consciousness. Cortex, 41, 695–704. Keenan, J. P., Wheeler, M., Platek, S. M., Lardi, G., & Lassonde, M. (2003b). Self-face processing in a callosotomy patient. European Journal of Neuroscience, 18, 2391–2395. Keyes, H., & Brady, N. (2010). Self-face recognition is characterized by ‘‘bilateral gain” and by faster, more accurate performance which persists when faces are inverted. The Quarterly Journal of Experimental Psychology, 63, 840–847. Keyes, H., Brady, N., Reilly, R. B., & Foxe, J. J. (2010). My face or yours? Event-related potential correlates of self-face processing. Brain and Cognition, 72, 244–254. Kircher, T. T. J., 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. Cognitive Brain Research, 10, 133–144. Kircher, T. T. J., Senior, C., Phillips, M. L., Rabe-Hesketh, S., Benson, P. J., Bullmore, E. T., et al (2001). Recognizing one’s own face. Cognition, 78, B1–B15. Kita, Y., Gunji, A., Sakihara, K., Inagaki, M., Kaga, M., Nakagawa, E., et al (2010). Scanning strategies do not modulate face identiﬁcation: Eye-tracking and near-infrared spectroscopy study. PLoS ONE, 5, e11050.

C. Devue, S. Brédart / Consciousness and Cognition 20 (2011) 40–51

51

Ma, Y., & Han, S. (2010). Why we respond faster to the self than to others? An implicit positive association theory of self-advantage during implicit face recognition. Journal of Experimental Psychology: Human Perception and Performance, 36, 619–633. McNeill, D. (1998). The face. Boston: Little, Brown and Company. Mitchell, R. W. (1993). Mental models of mirror self-recognition: Two theories. New Ideas in Psychology, 11, 295–325. Miyakoshi, M., Kanayama, N., Iidaka, T., & Ohira, H. (2010). EEG evidence for face-speciﬁc visual self-representation. NeuroImage, 50, 1666–1675. Miyakoshi, M., Nomura, M., & Ohira, H. (2007). An ERP study on self-relevant object recognition. Brain and Cognition, 63, 182–189. Morin, A. (2002). Right hemispheric self-awareness: A critical assessment. Consciousness and Cognition, 11, 396–401. Morin, A. (2006). Levels of consciousness and self-awareness: A comparison and integration of various neurocognitive view. Consciousness and Cognition, 15, 358–371. Morin, A. (2007). Self-awareness and the left hemisphere: The dark side of selectively reviewing the literature. Cortex, 43, 1068–1073. Morita, T., Itakura, S., Saito, D. N., Nakashita, S., Harada, T., Kochiyama, T., et al (2008). The role of the right prefrontal cortex in self-evaluation of the face: A functional magnetic resonance imaging study. Journal of Cognitive Neuroscience, 20, 342–355. Ninomiya, H., Onitsuka, T., Chen, C. H., Sato, E., & Tashiro, N. (1998). P300 in response to the subject’s own face. Psychiatry and Clinical Neurosciences, 52, 519–522. Northoff, G., & Bermpohl, F. (2004). Cortical midline structures and the self. Trends in Cognitive Sciences, 8, 102–107. 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, 440–457. Platek, S. M., & Gallup, G. G. (2002). Self-face recognition is affected by schizotypal personality traits. Schizophrenia Research, 57, 81–85. Platek, S. M., Keenan, J. P., Gallup, G. G. Jr., & Mohamed, F. B. (2004). Where am I? The neurological correlates of self and other. Cognitive Brain Research, 19, 114–122. Platek, S. M., Keenan, J. P., & Mohamed, F. B. (2005). Sex differences in the neural correlates of child facial resemblance: An event related fMRI study. NeuroImage, 25, 1336–1344. Platek, S. M., & Kemp, S. M. (2009). Is family special to the brain? An event-related fMRI study of familiar, familial, and self-face recognition. Neuropsychologia, 47, 849–858. Platek, S. M., Krill, A. L., & Wilson, B. (2009). Implicit trustworthiness ratings of self-resembling faces activate brain centers involved in reward. Neuropsychologia, 47, 289–293. Platek, S. M., Loughead, J. W., Gur, R. C., Busch, S., Ruparel, K., Phend, N., et al (2006). Neural substrates for functionally discriminating self-face from personally familiar faces. Human Brain Mapping, 27, 91–98. Platek, S. M., Thomson, J. W., & Gallup, G. G. (2004). Cross-modal self-recognition: The role of visual, auditory, and olfactory primes. Consciousness and Cognition, 13, 197–210. Platek, S. M., Wathne, K., Tierney, N. G., & Thomson, J. W. (2008). Neural correlates of self-face recognition: An effect-location meta-analysis. Brain Research, 1232, 173–184. Sperry, R. W., Zaidel, E., & Zaidel, D. (1979). Self-recognition and social awareness in the disconnected minor hemisphere. Neuropsychologia, 17, 153–166. Spuznar, K. K. (2010). Episodic future thought: An emerging concept. Perspectives in Psychological Science, 5, 142–162. Sugiura, M., Kawashima, R., Nakamura, K., Okada, K., Kato, T., Nakamura, A., et al (2000). Passive and active recognition of one’s own face. NeuroImage, 11, 36–48. Sugiura, M., Sassa, Y., Jeong, H., Horie, K., Sato, S., & Kawashima, R. (2008). Face-speciﬁc and domain-general characteristics of cortical responses during selfrecognition. NeuroImage, 42, 414–422. Sugiura, M., Sassa, Y., Jeong, H., Miura, N., Akitsuki, Y., Horie, K., et al (2006). Multiple brain networks for visual self-recognition with different sensitivity for motion and body part. NeuroImage, 32, 1905–1917. Sugiura, M., Watanabe, J., Maeda, Y., Matsue, Y., Fukuda, H., & Kawashima, R. (2005). Cortical mechanisms of visual self-recognition. NeuroImage, 24, 143–149. Sui, J., & Han, S. (2007). Self-construal priming modulates neural substrates of self-awareness. Psychological Science, 18, 861–866. Sui, J., Liu, C. H., & Han, S. (2009). Cultural differences in neural mechanisms of self-recognition. Social Neuroscience, 4, 402–411. Sui, J., Zhu, Y., & Han, S. (2006). Self-face recognition in attended and unattended conditions: An event-related brain potential study. Neuroreport, 17, 423–427. Tacikowski, P., & Nowicka, A. (2010). Allocation of attention to self-name and self-face: An ERP study. Biological Psychology, 84, 318–324. Taylor, M. J., Arsalidou, M., Bayless, S. J., Morris, D., Evans, J. W., & Barbeau, E. J. (2009). Neural correlates of personally familiar faces: Parents, partner and own faces. Human Brain Mapping, 30, 2008–2020. Tsakiris, M. (2008). Looking for myself: Current multisensory input alerts self-face recognition. PLoS ONE, 3, e4040. Tulving, E. (1985). Memory and consciousness. Canadian Psychology, 26, 1–12. Turk, D. J., Heartherton, T. F., Kelley, W. M., Funnell, M. G., Gazzaniga, M. S., & Macrae, C. N. (2002). Mike or me? Self-recognition in a split-brain patient. Nature Neuroscience, 5, 841–842. Uddin, L. Q., Iacobini, M., Lange, C., & Keenan, J. P. (2007). The self and social cognition: The role of cortical midline structures and mirror neurons. Trends in Cognitive Sciences, 11, 153–157. Uddin, L. Q., Kaplan, J. T., Molnar-Szakacs, I., Zaidel, E., & Iacoboni, M. (2005a). Self-face recognition activates a frontoparietal ‘‘mirror” network in the right hemisphere: An event-related fMRI study. NeuroImage, 25, 926–935. Uddin, L. Q., Molnar-Szakacs, I., Zaidel, E., & Iacobini, M. (2006). RTMS to the right inferior parietal lobule disrupts self–other discrimination. Social Cognitive and Affective Neuroscience, 1, 65–71. Uddin, L. Q., Rayman, J., & Zaidel, E. (2005b). Split-brain reveals separate but equal self-recognition in the two cerebral hemispheres. Consciousness and Cognition, 14, 633–640. Van Overwalle, F. (2009). Social cognition and the brain: A meta-analysis. Human Brain Mapping, 30, 829–858. Verosky, S. C., & Todorov, A. (2010). Differential neural responses to faces physically similar to the self as a function of their valence. NeuroImage, 49, 1690–1698.

Consciousness and Cognition 20 (2011) 52–63

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Brain imaging of the self – Conceptual, anatomical and methodological issues q Georg Northoff a,⇑,1, Pengmin Qin a,1, Todd E. Feinberg b,c,d a

Mind, Brain Imaging and Neuroethics, University of Ottawa Institute of Mental Health Research (IMHR), 1145 Carling Avenue Ottawa, ON, Canada K1Z 7K4 Albert Einstein College of Medicine, United States c Yarmon Neurobehavior Center, United States d Beth Israel Medical Center, United States b

a r t i c l e

i n f o

Article history: Received 13 September 2010 Available online 6 October 2010 Keywords: The self Brain image Cortical midline structures

a b s t r a c t In this paper we consider two major issues: conceptual–experimental approaches to the self, and the neuroanatomical substrate of the self. We distinguish content- and processed-based concepts of the self that entail different experimental strategies, and anatomically, we investigate the concept of midline structures in further detail and present a novel view on the anatomy of an integrated subcortical–cortical midline system. Presenting meta-analytic evidence, we show that the anterior paralimbic, e.g. midline, regions do indeed seem to be speciﬁc for self-speciﬁc stimuli. We conclude that future investigation of the self need to develop novel concepts that are more empirically plausible than those currently in use. Different concepts of self will require novel experimental designs that include, for example, the brain’s resting state activity as an independent variable. Modiﬁcations of both conceptual and anatomical dimensions will allow an empirically more plausible account of the relationship between brain and self. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction The problem of the self has been investigated extensively in neuroscience using most notably brain imaging (Gillihan & Farah, 2005; Legrand & Ruby, 2009; Metzinger & Gallese, 2003; Northoff & Bermpohl, 2004; Northoff et al., 2006). Comparing self- vs. non-self-speciﬁc stimuli, brain imaging studies observed neural activity changes in various medial cortical regions including the perigenual anterior cingulate cortex (pACC), dorsomedial prefrontal cortex (MPFC) and the posterior cingulate cortex (PCC) (Kelley et al., 2002; Mitchell, Banaji, & Macrae, 2005; Northoff & Bermpohl, 2004; Northoff et al., 2006; Platek et al., 2006; Uddin, Iacoboni, Lange, & Keenan, 2007; Yaoi, Osaka, & Osaka, 2009; Zhu, Zhang, Fan, & Han, 2007). The results obtained in single studies were however contradicted by recent meta-analyses on imaging studies of the self (Gillihan & Farah, 2005; Legrand & Ruby, 2009) which do not support the speciﬁc association of medial cortical regions, i.e., the cortical midline structures (CMS) (Northoff & Bermpohl, 2004) with self-speciﬁc stimuli. Instead, they demonstrated that theses regions may also be implicated in processing non-self-speciﬁc stimuli as, for instance personal familiar stimuli (Gillihan & Farah, 2005; Seger, Stone, & Keenan, 2004) or task-speciﬁc requirements like general evaluation (Legrand & Ruby, 2009).

q

This article is part of a special issue of this journal on ‘‘Brain and Self: Bridging the Gap”.

⇑ Corresponding author. Address: Mind, Brain Imaging and Neuroethics, University of Ottawa Institute of Mental Health Research (IMHR), Room 6959, 1145 Carling Avenue, Ottawa, ON, Canada K1Z 7K4. E-mail address: [email protected] (G. Northoff). 1 These authors have contributed equally. 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.011

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

53

However, the association of familiarity and task-speciﬁc requirements with the cortical midline structures during presentation of self- and non-self-speciﬁc stimuli remains to be investigated. In addition to the cortical midline structures, neural processing of self-speciﬁc stimuli has recently been also associated with resting state activity in the default-mode network (DMN) (Buckner, Andrews-Hanna, & Schacter, 2008; Raichle et al., 2001). Since the regions of the DMN strongly overlap with those of the cortical midline structures, some authors speak even of ‘default-self’ arguing that the self may be more or less identical with the high resting state activity observed in these regions (Boly et al., 2008; Christoff, Ream, Geddes, & Gabrieli, 2003; David et al., 2007; Golland et al., 2007; Gusnard, Akbudak, Shulman, & Raichle, 2001; Wicker, Ruby, Royet, & Fonlupt, 2003). If so the self may be assumed to be based purely on internal processing, i.e., the brain’s resting state activity, and thus distinguished from external processing, i.e., stimulus-induced activity. While there is some indirect support for the regional overlap in especially the pACC between resting state activity and neural activity induced by self-speciﬁc stimuli (D’Argembeau et al., 2005; Schneider et al., 2008), direct experimental demonstration is thus far lacking. This may also be due in part to the methodological difﬁculty of measuring resting state activity in relation to self-speciﬁc stimuli with the latter violating the former. There may be several reasons why the data regarding the neuroanatomical basis of the self conﬂict. One of them may be that differing neuroscientiﬁc and philosophical concepts of the self. (Legrand & Ruby, 2009; Northoff et al., 2006) may yield different results regarding its underlying anatomy. What exactly is meant by self when we for instance compare the neural effects of judging self- and non-self-speciﬁc stimuli? There may be other concepts of self which require a different experimental design. One possibility would be to determine if the self as judged by its internal contents is already present in the resting state independent of externally presented stimuli. This is suggested by the strong overlap between cortical midline structures (CMS) and the DMN. Another issue is that how we organize and interpret the hierarchical structure of brain anatomy with reference to the concept of anatomically ‘‘midline structures” has special relevance for our understanding of the neuroanatomy of the self Feinberg (2009, this issue) suggests that the neural hierarchy is organized along an anatomically medial–lateral or central–peripheral dimension. This results in anatomically concentric rings that extend the length of the neuroaxis from subcortical to cortical zones (Feinberg, this issue). According to this organization, the medial (interoself) system corresponds to the innermost rings and is related to self-related and homeostatic processes, the lateral (peripheral) system is anatomically related to the outer rings and subserves exterosensorimotor processes, and the middle rings represent an integrative system that mediates between the other two. Rostrally, the cortical midline regions (Northoff & Bermpohl, 2004; Northoff et al., 2006) are in part paralimbic regions that correspond to the medial- inner rings It would be of interest to see how the imaging data map onto this medial-inner and lateral outer distinction. Third, in addition to conceptual and anatomical issues, we need to consider some methodological issues. When observing the effects of self- and non-self-speciﬁc stimuli current imaging studies treat the self as independent variable on neural activity, while the neural activity itself as measured with fMRI is considered the dependent variable. This informs us about neural activity related to the stimulus itself, e.g., the stimulus-induced activity but it may not provide any insight into the brain’s intrinsic activity, e.g., its resting state activity, and how it modulates the stimulus-induced activity, e.g., rest–stimulus interaction (Northoff, Qin, & Nakao, 2010) However as there is strong overlap between stimulus-induced activity in CMS and resting state activity in the DMN, one may need to consider the latter, e.g., the resting state activity, in experimental designs. To do that, however, we may need to modify our current methodological and experimental approaches to the self in brain imaging studies. 2. Concepts of the self and their experimental realization 2.1. Content-based concepts of the self The question of the self has been one of the most salient problems throughout the history of philosophy, psychology and neuroscience (Gallagher, 2000; Gallagher & Frith, 2003; Metzinger & Gallese, 2003; Northoff, 2004). For example, William James distinguished between a physical self, a mental self, and a spiritual self. These distinct selves even may be related to distinct brain regions (Churchland, 2002; Dalgleish, 2004; Damasio, 1999, 2003a, 2003b; Gallagher, 2000; Gallagher & Frith, 2003; Keenan, Wheeler, Platek, Lardi, & Lassonde, 2003; Kelley et al., 2002; Kircher & David, 2003; Lambie & Marcel, 2002; LeDoux, 2002; Marcel & Lambie, 2004; Northoff & Bermpohl, 2004; Panksepp, 1998a, 2003; Stuss, Gallup, & Alexander, 2001; Turk, Heatherton, Macrae, Kelley, & Gazzaniga, 2003; Turk et al., 2002; Vogeley & Fink, 2003). Damasio (1999) and Panksepp (1998b), Panksepp (2003) suggest a ‘‘proto-self” that corresponds more or less to James’s physical self. The ‘‘proto-self” is supposed to outline one’s body in affective and sensory-motor terms and is associated with subcortical regions like the PAG, the colliculi and the tectum (Panksepp, 2007). Such bodily self-related sensorimotor contents resemble William James’s description of the physical self. A variant of such sensorimotor-based self has recently been suggested by Legrand and Ruby (2009). Based on the phenomenological distinction between reﬂexive, e.g., cognitive, and pre-reﬂexive, e.g., pre-cognitive self-awareness, they associate the latter with sensorimotor rather than cognitive contents (Legrand & Ruby, 2009). This emphasis on sensormotor functions is in agreement with their assumption of embodiment as crucial for reﬂexive and thus cognitive functions (Legrand, 2005). Following their sensorimotor-based concept of self, they assume that the neural structures underlying sensorimotor functions including sensorimotor feedback loops are crucially involved in generating a sense of self, e.g., pre-reﬂexive self-awareness. This hypothesis remains to be experimentally tested.

54

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

In addition to sensorimotor and bodily contents, mental contents are regarded as speciﬁc for the self. What has recently been described as the ‘‘minimal self” (Gallagher, 2000; Gallagher & Frith, 2003) or ‘‘core or mental self” (Damasio, 1999) might roughly correspond to James’ concept of mental self. The ‘‘core or mental self” builds upon the ‘‘proto-self” in mental terms and is associated with regions including the thalamus and the ventromedial prefrontal cortex (see for instance Damasio, 1999, 2003a). Instead of sensorimotor and bodily contents of the ‘proto-self’, the mental self is deﬁned by its mental contents and their associated cognitive contents. For instance, one’s own name may be considered a mental content that is speciﬁcally related to the self as part of the mental self, and therefore the mental self is not restricted to parts of one’s own body nor their underlying substrate. Instead, the mental self may also concern stimuli from the outside of the body and person. The central feature is not ownership (as in the case of the body) but rather the designation of certain stimuli as being either self or non-self-speciﬁc. Since the judgment of stimuli as either self- or non-self-speciﬁc is the guiding experimental paradigm in most current imaging studies, they presuppose in part the concept of the mental self (see below). Finally, a more extended concept of the self may be deﬁned. Here the self is not based on either sensorimotor or mental contents as the ‘proto- and the mental self’ but rather on its autobiographical contents. The inclusion of autobiographical memories further entails the concept of time, more speciﬁcally the subjective experience of time and episodic memory with its extension into past, present and future. Philosophically, the concept of the autobiographical self overlaps with the concept of personal identity and the question of temporal continuity. This is reﬂected in, for instance, Damasio’s (1999) ‘‘autobiographical self” and Gallagher’s (Gallagher, 2000; Gallagher & Frith, 2003) ‘‘narrative self” in that both rely on linking past, present, and future events thereby resembling James’ concept of a spiritual self. Since the autobiographical dimension impacts the ability to judge speciﬁc stimuli as either self- or non-self-speciﬁc, many of the current paradigms in brain imaging presuppose an ‘autobiographical self’. The concepts of self assumed in many imaging studies therefore amounts to an admixture of mental and autobiographical self. Taken together, the self is deﬁned here on the basis of various contents. The proto-self presupposes bodily contents. The mental self is determined by one’s own mental contents. Finally, the autobiographical self presupposes autobiographical contents and distinguishes them from heterobiographical contents. These different contents provide the basis for current neuroscience to ‘neuronalize’ the self, and provide the basis to investigate whether different self-speciﬁc contents are associated with different brain regions. 2.2. Process-based concepts of the self What remains unclear, however, is what unites the different content-based concepts of self? One common denominator is that the stimuli are often characterized as self-referential and entail self-referential processing that is considered common to the aforementioned distinct concepts of self. This has also been described as ‘self-related’ or ‘self-relevant’ processing (Churchland, 2002; Dalgleish, 2004; Gallagher, 2000; Gallagher & Frith, 2003; Keenan et al., 2003; Kelley et al., 2002; Lambie & Marcel, 2002; LeDoux, 2002; Marcel & Lambie, 2004; Northoff & Bermpohl, 2004; Turk et al., 2002, 2003). In this particular group of studies subjects were presented pictures, faces, words, or tones, and had to evaluate whether they were personally related to them or not. Faces, for instance, were presented from the own person, relatives, family members, and other nonrelated famous and non-famous persons. Subjects had to decide upon the degree of the stimuli’s closeness to the own person. Another example is the way we perceive pictures of ourselves or close friends vs. pictures of completely unknown people or pictures of our houses where we spent our childhood vs. pictures of any unknown house, etc. Such comparisons are possible in different sensory modalities. Self-relatedness is here understood and presupposed in a cognitive sense that implies that one becomes aware of one’s self once one sees the stimulus. The experimental designs in current imaging studies focus on the judgment of speciﬁc contents, be they sensorimotor/ bodily, mental or autobiographical. This judgment task implicates self-awareness or self-consciousness, the ability to become aware of that stimulus being speciﬁc or non-speciﬁc. Imaging studies thus combine a content-based view of the self, be it bodily, mental or autobiographical, with the recruitment of higher-order cognitive functions required in the task. Legrand and Ruby (2009) have criticized the latter. They argue that the imaging results may be confounded by a general non-speciﬁc evaluation function and inﬂuenced by the judgment required in these imaging studies. The focus on the role of making judgments raises the role of self-consciousness or self-awareness. This is because the various tasks applied in these studies required subjects to make explicit reference to some aspects of themselves and to consciously access and monitor representational content about one’s self. Since subjects must reference themselves in self-consciousness or self-awareness, one may speak of ‘self-referential processing’. Due to the fact that it requires selfconsciousness or self-awareness, self-referential is assumed to involve higher-order cognitive functions, the ‘‘highest” and most advanced forms of cognitive processing, out of which the self emerges at the pinnacle of the psychological and neural hierarchy (Feinberg, 2000, 2001a, 2001b). On the philosophical level, such higher-order view of self-referential processing corresponds to predominantly cognitive accounts of the self and subjectivity as it has for example been advanced by Kant and the German school of idealism. What are the alternatives? Experimentally, we need to replace ‘‘judgment” by a less cognitive task as for instance by mere perception of self-speciﬁc and non-self-speciﬁc stimuli. This strategy has been pursued by some single studies (Northoff et al., 2009; Qin et al., 2010; Schneider et al., 2008). In one, subjects were shown either emotional pictures (Northoff et al., 2009; Schneider et al., 2008) or their own name (Qin et al., 2010) and instructed to not make any judgment. This experimental design therefore does presuppose a judgment or general evaluation function. Interestingly, in both cases various cortical midline structures as well as subcortical regions were found to be active during the self-speciﬁc stimuli. This

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

55

indicates that the neural activity in these regions may not be related to the general evaluation function or judgment itself (Legrand & Ruby, 2009). In addition, neural activity in the CMS may also be independent of the consciousness of the self, e.g., self-consciousness. Qin et al. (2010) demonstrated neural activity in various cortical midline regions during perception of the subject’s own name in patient’s in vegetative state (Qin et al., 2010). These results indicate ﬁrst that the self may be processed independent of full consciousness, and second that the neural activity in the CMS may not be related to consciousness itself whether it is the consciousness of the self or the non-self. In summary, imaging results have demonstrated that the neural activity in the CMS is not speciﬁc for self-speciﬁc stimuli. Hence, distinction between self- and non-self-speciﬁc contents could not be mapped onto a corresponding anatomical distinction in the cortex. At the same time however, neural activity in the CMS may not be associated with judgment/general evaluation function or consciousness either. This means that the neural activity in the CMS cannot be accounted for by a speciﬁc function be it judgment/general evaluation or consciousness. This raises the question: What is the neural activity in the CMS speciﬁc for if it is neither self-speciﬁc contents nor a general evaluation function nor consciousness? Rather than being speciﬁc for a speciﬁc content (bodily, mental, autobiographical) or a speciﬁc function (judgment/general evaluation, consciousness), neural activity in the CMS may be assumed to be speciﬁc for a speciﬁc process. Conceptually, this entails a shift from a content- or function-based concept of self to a process-based view of the self. Neural activity in the CMS may then be determined by a speciﬁc process that is instantiated when being confronted with both self- and non-self-speciﬁc stimuli. What could this speciﬁc process be? 2.3. Psychological and experimental characterization of self-related processing Any stimuli, be they bodily, mental or autobiographical, must ﬁrst be related to the organism in order for the latter to be able to access the former as a speciﬁc content be it self- or non-self-speciﬁc in perception or judgment. The constitution of any content thus may be traced back to a speciﬁc relation between the stimulus and the organism which must itself be mediated by a speciﬁc process in order to yield any kind of content, be it bodily, mental, or autobiographical. The process that establishes a relation between the organism and a stimulus is called self-related processing. It is distinguished from its cognitive counterpart, self-referential processing, that takes the contents be they bodily, mental or autobiographical as given (and preexisting). Self-related processing concerns stimuli that are experienced as strongly related to one’s own person. A deﬁnition of self-related processing by experience implies a focus on the implicit, subjective, and phenomenal aspects (to feel or experience self-referential stimuli) what Kircher and David (2003) call ‘‘self-qualia” and Zahavi (2005) and others’ (Legrand, 2005; Legrand & Ruby, 2009) describe as ‘‘prereﬂective” whereas our current focus is less on associated cognitive and reﬂective functions. As such we distinguish self-related processing from what is commonly called ‘‘insight” which we consider to presuppose cognitive and reﬂective functions rather than simply pure subjective and phenomenal aspects (Metzinger & Gallese, 2003; Zahavi, 2005). Self-related processing (SRP), can neither be associated with the ‘‘self-as-object” nor the ‘‘self-as-subject”; instead, it makes this distinction ﬁrst and foremost possible in that it allows to distinguish between subject and object and hence between both concepts of the self. SRP must consequently be regarded more basic and fundamental than both subjective, i.e., phenomenological, and objective, i.e., neuroscientiﬁc, concepts of the self. As we will see in the following, characterization of SRP as non-cognitive, affective, basic and fundamental is central in constituting subjectivity and objectivity. Neither SRP nor the implied sense of self can be equated with any kind of content like self-speciﬁc contents as distinguished from non-self-speciﬁc ones or subjective-experiential contents as distinguished from objective-observational ones. Instead, SRP may conceptually be determined rather as process that ﬁrst and foremost makes the distinction between different kinds of contents with different degrees of self-relatedness possible. Considered in this way, the neural mechanisms underlying SRP can no longer be regarded the neural correlates, e.g., the sufﬁcient conditions, of the self. Instead, the neural mechanisms underlying SRP may only be considered a necessary condition which is not sufﬁcient by itself to constitute a self with its self-speciﬁc contents. SRP may only be a necessary but non-sufﬁcient condition of the self that as such enables and predisposes but not executes the self. One may consequently characterize the neuronal mechanisms underlying SRP no longer as neural correlates but rather as ‘neural predisposition’ of the self. This entails that methodologically we may need to tap into those neural mechanisms and processes that precede those we currently focus our attention within the context of our current designs. More speciﬁcally, this means that we may need to shift our attention from the perception or judgment of self- and non-self-speciﬁc contents to those mechanisms that precede, e.g., enable and predispose those very contents. Neuronally, this entails that we may need to shift our attention from stimulus-induced activity as related to self- and non-self-speciﬁc stimuli to the resting state activity itself and its interaction with the former amounting to rest–stimulus interaction (Northoff et al., 2010). This shift in the methodological focus would be well compatible with the above described overlap between CMS during self-speciﬁc stimuli and the high resting state activity in the DMN. Hence, our focus may need to shift from stimulus-induced activity to the brain’s intrinsic activity, its resting state activity, and how the latter interacts with the former, e.g., rest–stimulus interaction (see also Northoff et al., 2010). How though can we include the resting state activity of the brain and its impact on stimulus-induced activity, e.g., rest– stimulus interaction (Northoff et al., 2010), as variables in our experimental design? One way is to manipulate the level of resting state activity itself by for instance opening or closing the eyes (Logothetis et al., 2009; Raichle, 2010) and see how it impacts the neural processing of self- and non-self-speciﬁc stimuli and thus their stimulus-induced activity. While this has

56

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

been done in predominantly the sensory domain (see Northoff et al. (2010) for a recent review), it remains to be shown in the case of the self. Experimentally, this implies a reversal in our assignment of dependent and independent variable. The self is no longer the independent variable but rather the dependent one while the brain’s resting state activity becomes the independent variables. Both rest–stimulus interaction and stimulus-induced activity are then dependent variables. While being fully aware of the methodological challenges, this though remains to be done (Fig. 1). 3. Anatomy of the self and subcortical–cortical integration 3.1. Neuroanatomy: radial-concentric organization and subcortical–cortical systems in the brain Many imaging studies on the self demonstrated involvement of cortical midline structures like the VMPFC, the PACC, the DMPFC and the PCC in the neural processing of self-speciﬁc stimuli (Northoff et al., 2006). The self-speciﬁcity of the neural activity in the CMS has been questioned on several grounds. At the same time the concept of cortical midline structures has been extended to include subcortical midline regions (Feinberg, 2009; Northoff & Panksepp, 2008; Panksepp & Northoff, 2009). Consecutively, an integrated subcortical–cortical midline system has been proposed as speciﬁc for the self. This raises the question for the anatomical and functional continuity between subcortical and cortical regions that may be crucial for the self. Nieuwenhuys proposed there is a medial–lateral organization in subcortical regions that are located concentrically or radially around the aqueduct, with progressive extension from medial to lateral locations (Nieuwenhuys, 1996; Nieuwenhuys, Veening, & van Domburg, 1988/89; Nieuwenhuys, Voogd, & Van Huijzen, 2007). Based on various distinct features (see below), he distinguished the subcortical regions into three distinct territories, core, median and lateral paracore, and lateral regions which, despite being closely interconnected, can be distinguished from each other. Core subcortical regions refer to those regions that are located in direct proximity to the aqueduct and may thus be described as paraventricular or periaqueductal. These regions include the PAG, the pontine central gray, the medial hypothalamus, the septum, the parabrachial nuclei and the dorsal vagal complex. While the subcortical median paracore regions are located directly adjacent to the core regions; subcortical median paracore regions include the series of raphe nuclei, the lateral hypothalamus, the bed nucleus of the stria terminalis. These are closely connected to the bilateral paracore regions that include the ventral tegmental area (VTA), the locus coeruleus, the substantia nigra, the nucleus reticularis. Based upon this organization, Feinberg proposed that these regions can be thought as of a series of concentric rings (Feinberg, 2009, this issue). The inner rings (core and paracore regions), can be distinguished from the outer (lateral-peripheral) rings with respective to their ﬁbers (myelinated or unmyelinated), biogenic amines (serotonin, noradrenaline/adrenaline, dopamine, histamne), circumventricular organs, gonado-steroid receptors, and coherent behavior (e.g., as induced by localized electrical stimulation of the brain) (see Nieuwenhuys (1996, pp. 560–567) and Feinberg (2009) for details). According to Niewenhuys and co-workers, the core and paracore regions functionally are characterized by their involvement in processing interoceptive stimuli and regulating the body’s homeostatic milieu, vegetative-autonomic functions, and a

Environment and Body: Stimuli

Brain: Intrinsic Neural Activity

Experimental Variables: Resting State Activity and Rest-Stimulus Interaction (Independent), Stimulus-induced activity and Degree of selfrelatedness (Dependent)

Process-based concept of self: Hubs and Self-NonSelf Continuum

Contents in Mental States: Bodily, Environmental

Experimental Variables: Degree of Self-referentiality (Independent), Stimulusinduced activity (Independent),

Content-based concept of self: Specific regions and Self-Non-Self Dichotomy

Fig. 1. Content- and process-based concepts of the self and the brain.

57

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

A

Lateral regions

CMS

B

TPJ

PCC

TP Paralimbic self

familiarity

MPFC

PACC

AI Midline

Lateral

other

Fig. 2. (A) The traditional medial–lateral twofold anatomical dichotomy. Red: CMS; Cyan: lateral regions. (B) Threefold anatomical distinction. Red: paralimbic; Blue: Midline; Cyan: lateral. Red square represent the regions activated under self condition in meta-analysis; Green triangle represent the regions activated under familiarity condition in meta-analysis; Blue dots represent the regions activated under other condition in meta-analysis. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

variety of speciﬁc emotional and motivational processes. These medial structures are distinguished in Feinberg’s scheme from the lateral or peripheral rings that are concerned with exteroceptive and sensorimotor stimuli. Based on MacLean’s and Nauta’s concept of the limbic system, Nieuwenhuys (1996) proposes the core–paracore system extends into the meencephalon and diencephalon and is closely connected to the hypothalamus and various regions in the forebrain including the amygdala, septum, the hippocampus, and parahippocampal gyrus. This led to the concept of the ‘greater, distributed or extended limbic system’ (de Olmos & Heimer, 1999; Heimer, 2003; Morgane, Galler, & Mokler, 2005; Morgane & Mokler, 2006). Feinberg (2009, this issue) proposed that this medial system further extends up the neural hierarchy into paralimbic regions that anatomically linked to emotional and motivational networks These paralimbic areas include the lower parts of the orbitofrontal cortex, the perigenual, supragenual anterior cingulate cortex (PACC, SACC), the posterior cingulate cortex (PCC), the retrosplenial cortex (RSC), the temporal pole and the insula. Extending the radial-concentric organization into the subcortical mesencephalic level and its extensions into the forebrain as the starting point, Feinberg (2009) argues that the annular ring-like organization is preserved at the level of the cortex, the inner ring represented by medially located selfrelated systems and the outer ring represented by the exterosensorimotor systems. Feinberg also assumes a middle ring on the cortical level that is interposed between the inner and outer rings and thus between paralimbic and lateral cortical regions. He calls this the integrative self-system and it includes regions like the medial orbitofrontal cortex, the ventromedial and dorsomedial prefrontal cortex (VMPFC, DMPFC) and the medial parietal cortex (MPC) which have recently been subsumed under the concept of cortical midline structures (CMS) (Northoff & Bermpohl, 2004; Northoff et al., 2006). Since it is sandwiched between inner and outer rings and their involvement in intero- and exteroceptive processing respectively, Feinberg assumes this middle ring to account for integrating and linking both kinds of stimuli, i.e., intero–exteroceptive integration.2 The CMS do grossly overlap with what especially in the imaging domain is often described as the default-mode network (DMN) that is supposed to be characterized by particularly high resting state activity, e.g., intrinsic activity (see above in chapter 1 of this part as well as Buckner et al., 2008; Raichle et al., 2001). How such high intrinsic or resting state activity in the CMS is related to intero–exteroceptive integration, as postulated by Feinberg, remains unclear though. Taken together, the traditional medial–lateral twofold anatomical dichotomy is here challenged by a threefold anatomical distinction between three different concentric rings that extend from subcortical to cortical regions. These three rings can be characterized as paralimbic, heteromodal/CMS midline and exterosensorimotor/lateral regions.

2 What though remains unclear whether such intero–exteroceptive integration on the cortical level corresponds to analogous processes on the level of the forebrain and the mesencephalon. On could for instance imagine that what is described as core system on mesencephalic level may extend into the paralimbic areas since both are located directly adjacent to the aqueduct/ventricle. While the median and lateral paracore regions on the mesencephalic level may correspond to the middle ring on the cortical level and thus cortical midline structures. Support comes here from the connectivity pattern. Cortical regions like the anterior cingulate (PACC, SACC, PCC), the caudal orbitofrontal cortex, the temporal poles and the insula are characterized by strong inputs from especially the subcortical core regions like the PAG (see Nieuwenhuys, 1996, p. 573). In contrast, the VMPFC and the DMPFC receive for instance strong input from especially the raphe nuclei as median paracore regions and the locus coeruleus as lateral paracore region (Morgane et al., 2005; Nieuwenhuys, 1996).

58

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

3.2. Correspondence between anatomical and conceptual distinctions We now have two different anatomical distinctions on the cortical level (see Fig. 2a and b). There is the traditional one between medial and lateral regions. Medial regions include the PACC, SACC; VMPFC, DMPFC, PCC, MPC, and precuneus (see Fig. 2a) which, within the context of the self, have been subsumed under the concept of cortical midline structures (CMS) (Northoff & Bermpohl, 2004; Northoff et al., 2006). The CMS is distinguished from more lateral regions like the lateral prefrontal cortex and the lateral parietal cortex. The initial assumption was that the CMS are speciﬁc for self-relatedness which though as discussed is controversial. The question is whether there is a corrrespondence between anatomical and conceptual determination of the self as outlined above. What can one do now to increase the match between anatomical and conceptual determination of the self? One can either modify the conceptual determination of the self as proposed above, and/or one can modify the anatomical determination of the self in terms of medial regions and the CMS as suggested by Niewenhuys, Mesulam and Feinberg. Rather than dividing cortical regions into medial and lateral ones, this model suggests a threefold distinction between paralimbic, medial heteromodal (CMS) and exterosensorimotor/lateral regions on the cortical level (see Fig. 2b). The medial regions and thus the CMS are no longer a homogenous anatomical entity but are split off instead into paralimbic regions (PACC, SACC, PCC) and the heteromodal (CMS) regions (VMPFC, DMPFC, Precuneus). Moreover, the insula which in the medial–lateral model is classiﬁed as lateral region is now considered part of the paralimbic system (see Feinberg, this issue). Hence, the same regions are classiﬁed and grouped in different ways in both anatomical models, the twofold medial–lateral model and the triadic paralimbic-heteromodal/CMS–exterosensorimotor/lateral model. 3.3. Meta-analysis of the self – methods We conducted a descriptive meta-analysis of all recent studies on the self and matched the respective anatomical locations of self, familiarity and non-self onto both anatomical models. The concept of self was deﬁned operationally as stimuli like speciﬁc words that were attributed a high degree of relevance to the self; familiar but not self-related stimuli such as famous faces, and other stimuli that were neither self-related nor familiar. Anatomically we were limited to focussing on cortical regions because data on subcortical regions and the self are sparsely available (Northoff et al., 2009; Schneider et al., 2008). Selection of studies: We included three kinds of stimulus-dependent conditions, i.e., single studies and their comparisons, in our meta-analysis. We included studies focusing on self-speciﬁc stimulus and compared them with non-self-speciﬁc stimuli, i.e., stimuli related to other personal familiar and non-personal familiar persons. This accounted for the self conditions. In addition we also included another set of studies that focused on personal familiarity by comparing personal familiar stimuli and non-personal familiar stimuli and self-speciﬁc stimuli. The third set of studies, the other condition, included the results from the comparison between other stimuli (non-self-speciﬁc and non-personal familiar stimuli) and self-speciﬁc and personal familiar stimuli. The inclusion of the three sets of studies allowed us to investigate the relationship between self, familiarity and other. All the studies were selected from the search result in Pubmed from 1999 to August of 2009. Both the fMRI and PET results were included in the current meta-analysis. For all the four conditions, the following inclusion criteria were applied: 1. Only data (brain activity coordinates) from adult healthy subjects were included while those from neurological or psychiatric patients were excluded. 2. Only studies measuring brain activity in the whole brain were included while studies based on regions of interest (ROI) analysis were excluded. 3. All studies reporting signiﬁcant activity changes in speciﬁc regions were included whereby only the coordinates of the peak voxel maxima were considered while we did not consider the volume of the activated clusters. 4. Signiﬁcant activity coordinates within the whole brain were included as distinguished from region-of-interested-based meta-analyses (see for instance Northoff et al. (2006) who focused only on the midline regions). 5. Signiﬁcant activation changes yielded by task (e.g., self- vs. non-self judgment)- and/or stimulus (e.g., self- vs. non-selfspeciﬁc) comparison (image subtraction method, parametric designs and brain imaging (fMRI, PET)-other signal (behavior, ERP) correlations) were included. In contrast, data about functional connectivity were not considered. 6. The coordinates reported in the space of the MNI template or the atlas of Talairach and Tournoux were included. See the following information for each condition: Self condition: We included 57 recent papers about self-speciﬁc processing. We used a rather broad and unspeciﬁc deﬁnition of self-related tasks describing all tasks where some material or content had to be related to the persons themselves, i.e., their own selves. We used the following keywords to ﬁnd the studies for the self condition: ‘‘fMRI” or ‘‘PET” with ‘‘self”, ‘‘self-related”, ‘‘self-relevant”, ‘‘own name”, ‘‘own face”, ‘‘autobiographical”, ‘‘First-person perspective” and ‘‘Agency” in the title or abstract of the studies. The tasks used in these papers included trait adjective judgment, retrieval of personality traits, face recognition, body recognition, personal thinking, name perception, autobiographical memory, own feeling, self-administered pain, person perspective tasks and agency tasks. The following contrasts were employed in the single studies: self vs. personal familiarity, self vs. control/baseline, self vs. public people, ﬁrst-person perspective vs. third-person perspective and self vs. other (agency task). The coordinates that showed signiﬁcantly

59

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

stronger brain activity comparing the self condition with other conditions (even all the condition showed deactivation compared with the baseline) in the single studies were considered. Finally, the coordinates of the brain regions showing correlation between self evaluation and BOLD signal were also included. Familiarity condition: According to the difference between personal familiar people and famous people (Sugiura et al., 2009), our familiarity condition included 23 recent papers that investigated the neural effects of personally familiar people, e.g. participants’ family, friends, classmates and relatives. We used the following keywords to ﬁnd the studies for the familiarity condition: ‘‘fMRI” or ‘‘PET” with ‘‘familiarity”, ‘‘familiar name”, ‘‘familiar face”, and ‘‘familiar voice” in the title or abstract of the studies. The tasks adopted in the single studies on familiarity included face recognition, body recognition, voice recognition, trait adjective judgment and name recognition. The following contrasts were employed in the single studies: personal familiarity vs. self, personal familiarity vs. stranger/baseline and personal familiarity vs. public people. The coordinates that showed signiﬁcantly stronger brain activity in personal familiarity condition when compared with the other condition in single studies were also included. Finally, the coordinates showing common brain activity for self and familiarity (Vanderwal, Hunyadi, Grupe, Connors, & Schultz, 2008) were also included in the familiarity condition. While the very same brain regions were not considered in the self condition (because we assumed that the common regions for self and familiarity represent the personal familiarity of self-speciﬁc stimuli). Other condition: The other condition included 23 recent papers that came from both the self condition and the familiarity condition. These studies employed trait adjective judgment about public people, agency, public people’s name recognition, public people’s face recognition, or retrieval of public people’s trait adjective, the third personal perspective tasks and the other agency tasks that were taken as control condition in the studies on self and familiarity. The other condition included the following contrasts in the single studies: public people vs. self, public people vs. stranger/baseline, public people vs. personal familiar people, third-person perspective vs. ﬁrst-person perspective and other vs. self (agency task). The coordinates that showed signiﬁcantly stronger brain activity comparing the other condition with the self and familiarity conditions in the single studies were included. General statistical analysis: We used Multilevel Kernel Density Analysis (MKDA) (Wager, Lindquist, Nichols, Kober, & Van Snellenberg, 2009) to process our meta-analysis, a voxel-wise coordinates based meta-analysis on the brain imaging studies. In MKDA, the coordinates are treated as the location of the activation; the coordinates from one contrast in one study make up a particular statistical contrast map (SCM). The main aim of the MKDA is to reconstruct a map of signiﬁcant regions for each statistical contrast map within each study, and analyze the consistency and speciﬁcity across all the studies in the neighborhood of each voxel. In the following, the detailed method used in the present study will be described. The coordinates (peak activation) in each single study were transferred in MKDA to a standard brain from the Montreal Neurologic Institute as distributed with SPM2 software (Wellcome Department of Imaging Neuroscience, London, UK) and the coordinates from the same contrast will make up one special SCM. In order to integrate the coordinates in space, the coordinate in each SCM were considered as one spherical kernel with radius = 10 mm. This means that the voxels around the coordinates in 10 mm were regarded as activated, the value of these voxels were threshold at a maximum of 1. This contributed to construct one indicator map for each SCM where the value 1 in the voxel represented a coordinate (reported in the single study) in the neighborhood. The indicator maps were then weighted by the number of the subjects and the kind of data analysis (random or ﬁxed). The current version of MKDA weights each SCM by the square root of the number of the subjects. Studies using ﬁxed effects analysis let to w down-weighting of the SCM by a factor of 0.75. We did not consider the z-scores of the single studies because (i) they are not provided by all studies and (ii) their inclusion has been shown to confound with the replicability

Table 1 Comparison between the two- and three-fold anatomical characterizations with regard to meta-analytic results from self, familiarity and other. Self

Familiarity

Other (no-self and no familiarity)

Paralimbic Anterior Posterior

PACC, Insula PCC

– PCC

– PCC, TP

Midline Anterior Posterior

MPFC –

MPFC –

– –

Lateral

–

–

TPJ

CMS Anterior Posterior

PACC, MPFC PCC

MPFC PCC

– PCC

Lateral regions Anterior Posterior

Insula –

– –

– TPJ, TP

PACC: perigenual anterior cingulate cortex, PCC: posterior cingulate cortex, MPFC: medial prefrontal cortex, TP: temporal pole, TPJ: temporoparietal junction.

60

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

of activation across studies hence making interpretation more difﬁcult (Kober et al., 2008; Wager et al., 2009). The weighted average of the indicator maps was subsequently compared with the maximum proportion of the activated comparison maps expected under the null hypothesis that there is no coherent spatial consistence across the SCMs. During the calculation, the random effects analysis was used. For the threshold, MKDA used the threshold derived from Monte Carlo Simulation of the global null hypothesis. The contiguous activated clusters of each SCM were identiﬁed, and were selected at random within a gray matter mask (smoothed to include an 8 mm border, derived from segmentation of the avg152T1.img template using SPM2). In the present study, we used 5000 Monte Carlo iterations (the results should be stabilize after 2000) (Wager et al., 2009). We conducted the meta-analysis for each condition separately, i.e., self, familiarity, and other, to reveal those brain regions speciﬁcally associated with each condition. 3.4. Meta-analysis of the self – results During the self condition, the meta-analysis yielded activated clusters in the PACC, the MPFC and PCC as well as in other regions including the left anterior insula and right inferior frontal gyrus (IFG). The familiarity condition revealed activated clusters in MPFC and PCC but neither in PACC nor the insula. While the other condition yielded activation clusters only in posterior midline regions like the PCC and other temporal regions like the bilateral temporoparietal junction (TPJ) and left temporal pole (TP). In a second step, we mapped these results onto both anatomical distinctions, the twofold medial–lateral one and the threefold paralimbic-midline-lateral one (see Fig. 2a and b and Table 1). This showed that both self and familiarity conditions were grouped together within the medial regions of the CMS without being distinguishable from each other. Moreover, the self condition also recruited a lateral region, the insula (see Fig. 2a and Table 1). Hence, the medial–lateral model can neither distinguish between self and familiarity grouping both in the same set of regions, the anterior CMS. Nor can it group the self coherently into the medial regions because it associates the self also with a lateral (as based on the traditional twofold model) region, the insula. Taken together, the medial–lateral anatomical determination fails to distinguish the self from both familiarity and lateral regions. We then considered whether the data would ﬁt the aforementioned paralimbic-heteromodal/CMS–exterosensorimotor/ lateral triadic division. Using this model, studies on the self recruited the PACC and the insula are considered paralimbic regions; these regions were speciﬁc for the self as distinguished from both familiarity and other (see Fig. 2b and Table 1). In addition, studies on the self also recruited medial prefrontal regions (MPFC) including both VMPFC and DMPFC; there they though overlapped with the familiarity condition. Hence, unlike the paralimbic regions, the anterior midline regions turned out to be non-speciﬁc for the self. All three, self, familiarity and other mapped onto posterior paralimbic regions like the PCC and the temporal pole. 3.5. Meta-analysis of the self – discussion The main results of our meta-analysis are: (i) no distinction of the self from familiarity in the medial–lateral model while both can be distinguished from each other in the threefold paralimbic-heteromodal/CMS–exterosensorimotor/lateral model; (ii) no speciﬁc association of the self with exclusively medial regions in the medial–lateral model because of recruitment of the insula which though in the threefold model is grouped among the paralimbic regions indicating self-speciﬁcity (see Feinberg, this issue); and (iii) anterior–posterior distinction with anterior regions being speciﬁc for self and/or familiarity as distinguished from other while posterior regions are recruited by all three. All three ﬁndings shall be discussed in more detail in the following. We found the self could not be distinguished from familiarity in the medial–lateral model while it was speciﬁcally associated with anterior paralimbic regions in the threefold model. The PACC was speciﬁc for the self condition; this region was neither recruited during familiarity nor during other. While in the medial–lateral model the PACC is grouped with other medial regions, the MPFC, it is distinguished from them in the threefold model as paralimbic region as distinguished from midline regions. The threefold anatomical model thus matches better with our conceptual (and psychological) distinction between self and familiarity. One may consecutively regard the threefold anatomical model (Feinberg, 2009) to be more empirically plausible with regard to the self than the twofold model. The threefold model considers the PACC to be different from medial prefrontal cortical regions like the VMPFC and DMPFC. As described above this is mainly based on cytoarchitectonic, neurochemical and connectional features. Our meta-analysis adds a psychological dimension to it by characterizing the PACC as speciﬁc for the self as distinguished from familiarity and other. Hence, the distinction of the anterior paralimbic regions from anterior midline regions in the threefold model is supported on psychological grounds. Interestingly, the very same region, the PACC, shows hyperactivity in the resting state in patients with depression, e.g., major depressive disorders, as it has been demonstrated in both human patients and animal models of depression (Alcaro, Panksepp, Witczak, Hayes, & Northoff, 2010). While other medial prefrontal regions, the midline regions, do not display such hyperactivity in depression. Most importantly, depressed patients do indeed show an increased preoccupation with their own self (Northoff, 2007), the increased self-focus which indeed has been associated with the PACC and especially its resting state activity (Grimm et al., 2009). This lends further empirical support, e.g., psychiatric support, to the threefold anatomical model and its distinction between paralimbic and midline regions.

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

61

Considering the psychological and psychiatric evidence in favor of the distinction between anterior paralimbic from midline regions, this raises the question of the underlying physiological mechanisms. The PACC shows most consistently negative BOLD responses (NBR), e.g., deactivation, during stimulation in fMRI (Northoff et al., 2007). The physiological basis of such NBR in PACC remains however unclear. First, there is some evidence linking PACC with neural inhibition and gaba-ergic modulation which though needs to be explored in further detail. The issue here is whether this is really speciﬁc for the PACC as distinguished from other regions or whether it is rather speciﬁc for NBR in general independent of speciﬁc regions like the PACC. Muthukumaraswamy and Singh (2009) for instance demonstrated the association of GABA with NBR in visual cortex (Muthukumaraswamy & Singh, 2009). Moreover, the speciﬁc relationship of NBR to the self as distinguished from familiarity needs to be demonstrated in order to link PACC and the self. Interestingly, depressed patients show reduced NBR in PACC (Grimm, Boesiger et al., 2009) which is also decoupled from gaba-ergic modulation (Walter et al., 2009). How that is related to the increased self-focus in depression remains unclear. Our second main ﬁnding concerns the relationship between the insula and the self. The insula is characterized as a lateral region in the twofold medial–lateral model and distinguishes the insula from the PACC as medial region. In th twofold model the self-stimuli recruits both a medial region (PACC) and a lateral region (insula). In the threefold anatomical model (Feinberg, 2009, this issue), the insula is grouped as paralimbic region together with the PACC, and we found that both paralimbic regions, PACC and insula, were speciﬁc for the self as distinguished from both familiarity and other. Hence, the threefold anatomical distinction seems to match better with the conceptual distinction between self and familiarity than twofold anatomical model. The speciﬁc involvement of the insula in self has recently been shown in two studies. The insula is closely involved in processing interoceptive stimuli from the body (Craig, 2002, 2004, 2009) which, following the threefold model, relays to the PACC. If the self is speciﬁcally associated with both insula and PACC, one would assume interoceptive processing to be crucially involved in the self and its distinction from familiarity. This supports Feinberg’s (2009, this issue) proposal that the paralimbic regions mediate an interoceptive- and thus bodily-based self as distinguished from an exteroceptive- and environment-based self. However, neither the PACC nor the insula are exclusively and speciﬁcally involved in processing interoceptive stimuli. Rather both regions have been demonstrated in processing exteroceptive stimuli as for instance emotional stimuli with the insula showing positive BOLD responses (PBR) rather than NBR as the PACC (Phan, Wager, Taylor, & Liberzon, 2002). Therefore, neither the insula nor the PACC can be regarded to be speciﬁc for interoceptive stimuli as distinguished from exteroceptive ones, However, these regions may be critically involved in processing self-related and emotional stimuli. The predominance of the above mentioned stimulus-induced NBR in the PACC indirectly indicates the level of resting state activity in that region. This level of resting state activity may be modulated by either exteroceptive or interoceptive stimuli presupposing what recently has been called rest–stimulus interaction (Northoff et al., 2010). However, while the interaction of the PACC resting state activity with exteroceptive stimuli has been demonstrated (Grimm et al., 2009), this remains to be demonstrated for interoceptive stimuli. Considering our meta-analytic ﬁndings here, one may assume that a speciﬁc interaction of the resting state activity in the PACC with intero- and exteroceptive stimuli may distinguish the self from familiarity and other. The self would then be generated by a speciﬁc rest–intero–extero interaction that may be mediated by a speciﬁc constellation of the neural activities in the insula and the PACC. Future studies could examine the functional connectivity and the constellation of signal changes, e.g., positive and negative BOLD responses (NBR, PBR) in the insula and the PACC. Our third main ﬁnding consisted in the distinction between anterior and posterior regions with only the former being involved in self and familiarity. In contrast to anterior paralimbic regions, posterior paralimbic regions are involved in all three familiarity, self and other. This suggests that anatomically, the threefold anatomical distinctions between paralimbic, heteromodal/CMS and exterosensorimotor/lateral regions holds with regard to the self may hold for the anterior regions where self from familiarity and other are dissociable. Therefore, the threefold anatomical distinction between paralimbic, heteromodal/CMS midline and exterosensorimotor/ lateral regions may need to be complemented by the anterior–posterior dimension. Our ﬁndings from the meta-analysis underline the importance of considering the anterior–posterior distinction within the threefold anatomical model. Future anatomical investigation may reveal whether there are anatomical features, cytoarchitectonic, neurochemical, or connectional (or others), that distinguish the anterior from the posterior regions especially within the paralimbic (and also the midline and lateral) regions. What does all this entail for the concept of the self? Characterizing the self by a speciﬁc rest–intero–extero interaction presupposes the self as a speciﬁc process. The rest–intero–extero interaction describes a speciﬁc process rather than a particular content. If the self does indeed correspond to the process of rest–intero–extero interaction, one may also assume a continuous relationship between self, familiarity and other. This means that there is a continuous transition from self over familiarity to other. Such a ‘‘more-or-less” distinction of the process-based concept of self should be distinguished from the ‘‘all-or-nothing” distinction between self and non-self as is presupposed in the content-based concept of the self where the content is either self-related or not. What is speciﬁc about the PACC and insula with regard to the self may thus not be so much their exclusive anatomical involvement in the self but rather the kind of balance between resting state activity and interoceptive and exteroceptive stimulus processing. There is thus not ‘anatomical speciﬁcity’ but rather ‘processing speciﬁcity’ that makes the PACC and insula special nodes or hubs” in the neural network underlying self and familiarity. Such ‘processing speciﬁcity’ may in part also derive from the intimate connections of the PACC and the insula with the interoself systems extending from hierarchically lower subcortical regions (Feinberg, 2009; this issue).

62

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

4. Conclusions We here investigated two problems central in the imaging of the self, conceptual–experimental and anatomical issues. Conceptually, we distinguished between content- and process-based views of the self which were also shown to require different experimental approaches. The content-based view deﬁnes the self by speciﬁc contents (bodily, mental or autobiographical) and searches for the neural correlates of these contents and their respective stimulus-induced activity. The process-based view, in contrast, focuses on the processes that enable and predispose the constitution of these contents which can be traced back to the relation between stimuli and organism. The process-based view focuses on resting state activity and its impact on the neural processing of self- and non-self-speciﬁc stimuli, e.g., rest–stimulus interaction rather than on stimulus-induced activity. Methodologically, this requires a shift from the self as an independent to a dependent variable experimental designs. Anatomically, we presented empirical evidence in favor of an anterior cortical paralimbic speciﬁcity for self-speciﬁc stimuli as distinguished from familiar and non-familiar non-self-speciﬁc stimuli. While this was the case for anterior paralimbic regions it could not be observed in posterior paralimbic regions. This supports a neuroanatomical dissociation between anterior and posterior midline, e.g., paralimbic regions, with regard to the self and its distinction from familiarity and other. Taken together, our ﬁndings support the view that the threefold anatomical distinction between paralimbic, heteromodal/CMS midline and exterosensorimotor/lateral regions is empirically more plausible with regard to the self than the traditional twofold distinction between medial and lateral regions. Moreover, the ﬁndings from our meta-analysis are more in line with a process- rather than a content-based concept of the self. However, both the empirical processes including the exact anatomical, physiological and functional characterization of the self as well as conceptual characterization of a process-based self need to be developed in further detail in the future. Acknowledgments The work on this paper was supported by grants from the CRC, the CIHR, and the CIHR-EJLB to GN. References Alcaro, A., Panksepp, J., Witczak, J., Hayes, D. J., & Northoff, G. (2010). Is subcortical–cortical midline activity in depression mediated by glutamate and GABA? A cross-species translational approach. Neuroscience and Biobehavioral Reviews, 34(4), 592–605. Boly, M., Phillips, C., Tshibanda, L., Vanhaudenhuyse, A., Schabus, M., Dang-Vu, T. T., et al (2008). Intrinsic brain activity in altered states of consciousness: How conscious is the default mode of brain function? Annals of the New York Academy of Sciences, 1129, 119–129. Buckner, R. L., Andrews-Hanna, J. R., & Schacter, D. L. (2008). The brain’s default network: Anatomy, function, and relevance to disease. Annals of the New York Academy of Sciences, 1124, 1–38. Christoff, K., Ream, J. M., Geddes, L. P. T., & Gabrieli, J. D. E. (2003). Evaluating self-generated information: Anterior prefrontal contributions to human cognition. Behavioral Neuroscience, 117(6), 1161–1168. Churchland, P. S. (2002). Self-representation in nervous systems. Science, 296(5566), 308–310. Craig, A. D. (2002). How do you feel? Interoception: The sense of the physiological condition of the body. Nature Reviews Neuroscience, 3(8), 655–666. Craig, A. D. (2004). Human feelings: Why are some more aware than others? Trends in Cognitive Sciences, 8(6), 239–241. Craig, A. D. (2009). How do you feel–now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10(1), 59–70. Dalgleish, T. (2004). The emotional brain. Nature Reviews Neuroscience, 5(7), 583–589. Damasio, A. (1999). The feeling of what happens: Body and emotion in the making of consciousness. Harcourt Brace. Damasio, A. (2003a). Feelings of emotion and the self. Annals of the New York Academy of Sciences, 1001, 253–261. Damasio, A. (2003b). Mental self: The person within. Nature, 423(6937), 227. D’Argembeau, A., Collette, F., Van der Linden, M., Laureys, S., Del Fiore, G., Degueldre, C., et al (2005). Self-referential reﬂective activity and its relationship with rest: A PET study. Neuroimage, 25(2), 616–624. 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. de Olmos, J. S., & Heimer, L. (1999). The concepts of the ventral striatopallidal system and extended amygdala. Annals of the New York Academy of Sciences, 877, 1–32. Feinberg, T. E. (2000). The nested hierarchy of consciousness: A neurobiological solution to the problem of mental unity. Neurocase, 6, 75–81. Feinberg, T. E. (2001a). Altered egos: How the brain creates the self? New York: Oxford University Press. Feinberg, T. E. (2001b). Why the mind is not a radically emergent feature of the brain? Journal of Consciousness Studies, 8, 123–145. Feinberg, T. E. (2009). From axons to identity: Neurological explorations of the nature of the self (1st ed.). W.W. Norton & Company (August 3, 2009). Gallagher, I. I. (2000). Philosophical conceptions of the self: Implications for cognitive science. Trends in Cognitive Sciences, 4(1), 14–21. Gallagher, H. L., & Frith, C. D. (2003). Functional imaging of ‘theory of mind’. Trends in Cognitive Sciences, 7(2), 77–83. Gillihan, S. J., & Farah, M. J. (2005). Is self special? A critical review of evidence from experimental psychology and cognitive neuroscience. Psychological Bulletin, 131(1), 76–97. Golland, Y., Bentin, S., Gelbard, H., Benjamini, Y., Heller, R., Nir, Y., et al (2007). Extrinsic and intrinsic systems in the posterior cortex of the human brain revealed during natural sensory stimulation. Cerebral Cortex, 17(4), 766–777. Grimm, S., Boesiger, P., Beck, J., Schuepbach, D., Bermpohl, F., Walter, M., et al (2009). Altered negative BOLD responses in the default-mode network during emotion processing in depressed subjects. Neuropsychopharmacology, 34(4), 932–943. 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(8), 2617–2627. 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. Proceedings of the National Academy of Science, USA, 98(7), 4259–4264. Heimer, L. (2003). A new anatomical framework for neuropsychiatric disorders and drug abuse. American Journal of Psychiatry, 160(10), 1726–1739. Keenan, J. P., Wheeler, M., Platek, S. M., Lardi, G., & Lassonde, M. (2003). Self-face processing in a callosotomy patient. European Journal of Neuroscience, 18(8), 2391–2395. Kelley, W. M., Macrae, C. N., Wyland, C. L., Caglar, S., Inati, S., & Heatherton, T. F. (2002). Finding the self? An event-related fMRI study. Journal of Cognitive Science, 14(5), 785–794.

G. Northoff et al. / Consciousness and Cognition 20 (2011) 52–63

63

Kircher, T., & David, A. S. (2003). The self and the schizophrenia. Oxford University Press. Kober, H., Barrett, L. F., Joseph, J., Bliss-Moreau, E., Lindquist, K., & Wager, T. D. (2008). Functional grouping and cortical–subcortical interactions in emotion: A meta-analysis of neuroimaging studies. Neuroimage, 42(2), 998–1031. Lambie, J. A., & Marcel, A. J. (2002). Consciousness and the varieties of emotion experience: A theoretical framework. Psychological Review, 109(2), 219–259. LeDoux, J. (2002). Synaptic self: How our brain become who we are? New York: Viking. Legrand, D. (2005). How the body shapes the mind? Trends in Cognitive Sciences, 9(9), 413–414. Legrand, D., & Ruby, P. (2009). What is self-speciﬁc? Theoretical investigation and critical review of neuroimaging results. Psychological Review, 116(1), 252–282. Logothetis, N. K., Murayama, Y., Augath, M., Steffen, T., Werner, J., & Oeltermann, A. (2009). How not to study spontaneous activity? Neuroimage, 45(4), 1080–1089. Marcel, A. J., & Lambie, J. A. (2004). How many selves in emotion experience? Reply to Dalgleish and Power (2004). Psychological Review, 111(3), 820–826. Metzinger, T., & Gallese, V. (2003). The emergence of a shared action ontology: Building blocks for a theory. Consciousness and Cognition, 12(4), 549–571. Mitchell, J. P., Banaji, M. R., & Macrae, C. N. (2005). The link between social cognition and self-referential thought in the medial prefrontal cortex. Journal of Cognitive Neuroscience, 17(8), 1306–1315. Morgane, P. J., Galler, J. R., & Mokler, D. J. (2005). A review of systems and networks of the limbic forebrain/limbic midbrain. Progress in Neurobiology, 75(2), 143–160. Morgane, P. J., & Mokler, D. J. (2006). The limbic brain: Continuing resolution. Neuroscience and Biobehavioral Reviews, 30(2), 119–125. Muthukumaraswamy, S. D., & Singh, K. D. (2009). Functional decoupling of BOLD and gamma-band amplitudes in human primary visual cortex. Human Brain Mapping, 30(7), 2000–2007. Nieuwenhuys, R. (1996). The greater limbic system, the emotional motor system and the brain. Progress in Brain Research, 107, 551–580. Nieuwenhuys, R., Veening, J. G., & van Domburg, P. (1988/89). Core and paracores; some new chemoarchitectural entities in the mammalian neuraxis. Acta Morphologica Neerlando-Scandinavica, 26(2–3), 131–163. Nieuwenhuys, R., Voogd, J., & Van Huijzen, C. (2007). The human central nervous system: A synopsis and Atlas (4th ed.). Berlin Heidelberg: Springer. Northoff, G. (2004). Philosophy of the brain. The brain problem. Amsterdam, New York: John Benjamins Publisher. Northoff, G. (2007). Psychopathology and pathophysiology of the self in depression – Neuropsychiatric hypothesis. Journal of Affective Disorders, 104(1–3), 1–14. Northoff, G., & Bermpohl, F. (2004). Cortical midline structures and the self. Trends in Cognitive Science, 8(3), 102–107. 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. Northoff, G., & Panksepp, J. (2008). The trans-species concept of self and the subcortical–cortical midline system. Trends in Cognitive Sciences, 12(7), 259–264. Northoff, G., Qin, P., & Nakao, T. (2010). Rest–stimulus interaction in the brain: A review. Trends in Neurosciences, 33(6), 277–284. Northoff, G., Schneider, F., Rotte, M., Matthiae, C., Tempelmann, C., Wiebking, C., et al (2009). Differential parametric modulation of self-relatedness and emotions in different brain regions. Human Brain Mapping, 30(2), 369–382. Northoff, G., Walter, M., Schulte, R. F., Beck, J., Dydak, U., Henning, A., et al (2007). GABA concentrations in the human anterior cingulate cortex predict negative BOLD responses in fMRI. Nature Neuroscience, 10(12), 1515–1517. Panksepp, J. (1998a). Affective neuroscience: The foundations of human and animal emotions. Oxford University Press. Panksepp, J. (1998b). The preiconscious substrates of consciousness: Affective states and the evolutionary origins of the SELF. Journal of Consciousness Studies, 5, 566–582. Panksepp, J. (2003). At the interface of the affective, behavioral, and cognitive neurosciences: Decoding the emotional feelings of the brain. Brain and Cognition, 52(1), 4–14. Panksepp, J. (2007). The neuroevolutionary and neuroaffective psychobiology of the prosocial brain. In R. I. M. Dunbar & L. Barrett (Eds.), The Oxford handbook of evolutionary psychology (pp. 145–162). Oxford, UK: Oxford University Press. Panksepp, J., & Northoff, G. (2009). The trans-species core SELF: The emergence of active cultural and neuro-ecological agents through self-related processing within subcortical–cortical midline networks. Consciousness and Cognition, 18(1), 193–215. Phan, K. L., Wager, T., Taylor, S. F., & Liberzon, I. (2002). Functional neuroanatomy of emotion: A meta-analysis of emotion activation studies in PET and fMRI. Neuroimage, 16(2), 331–348. Platek, S. M., Loughead, J. W., Gur, R. C., Busch, S., Ruparel, K., Phend, N., et al (2006). Neural substrates for functionally discriminating self-face from personally familiar faces. Human Brain Mapping, 27(2), 91–98. Qin, P., Di, H., Liu, Y., Yu, S., Gong, Q., Duncan, N., et al. (2010). Anterior cingulate activity and the self in disorders of consciousness. Human Brain Mapping, in press. Raichle, M. E. (2010). Two views of brain function. Trends in Cognitive Sciences, 14(4), 180–190. Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers, W. J., Gusnard, D. A., & Shulman, G. L. (2001). A default mode of brain function. Proceedings of the National Academy of Science, USA, 98(2), 676–682. Schneider, F., Bermpohl, F., Heinzel, A., Rotte, M., Walter, M., Tempelmann, C., et al (2008). The resting brain and our self: Self-relatedness modulates resting state neural activity in cortical midline structures. Neuroscience, 157(1), 120–131. Seger, C. A., Stone, M., & Keenan, J. P. (2004). Cortical activations during judgments about the self and an other person. Neuropsychologia, 42(9), 1168–1177. Stuss, D. T., Gallup, G. G., Jr., & Alexander, M. P. (2001). The frontal lobes are necessary for ‘theory of mind’. Brain, 124(Pt 2), 279–286. Sugiura, M., Sassa, Y., Watanabe, J., Akitsuki, Y., Maeda, Y., Matsue, Y., et al (2009). Anatomical segregation of representations of personally familiar and famous people in the temporal and parietal cortices. Journal of Cognitive Neuroscience, 21(10), 1855–1868. Turk, D. J., Heatherton, T. F., Kelley, W. M., Funnell, M. G., Gazzaniga, M. S., & Macrae, C. N. (2002). Mike or me? Self-recognition in a split-brain patient. Nature Neuroscience, 5(9), 841–842. Turk, D. J., Heatherton, T. F., Macrae, C. N., Kelley, W. M., & Gazzaniga, M. S. (2003). Out of contact, out of mind: The distributed nature of the self. Annals of the New York Academy of Sciences, 1001, 65–78. 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 Sciences, 11(4), 153–157. Vanderwal, T., Hunyadi, E., Grupe, D. W., Connors, C. M., & Schultz, R. T. (2008). Self, mother and abstract other: An fMRI study of reﬂective social processing. Neuroimage, 41(4), 1437–1446. Vogeley, K., & Fink, G. R. (2003). Neural correlates of the ﬁrst-person-perspective. Trends in Cognitive Sciences, 7(1), 38–42. Wager, T. D., Lindquist, M. A., Nichols, T. E., Kober, H., & Van Snellenberg, J. X. (2009). Evaluating the consistency and speciﬁcity of neuroimaging data using meta-analysis. Neuroimage, 45(Suppl. 1), S210–S221. Walter, M., Henning, A., Grimm, S., Schulte, R. F., Beck, J., Dydak, U., et al (2009). The relationship between aberrant neuronal activation in the pregenual anterior cingulate, altered glutamatergic metabolism, and anhedonia in major depression. Archives of General Psychiatry, 66(5), 478–486. Wicker, B., Ruby, P., Royet, J. P., & Fonlupt, P. (2003). A relation between rest and the self in the brain? Brain Research. Brain Research Reviews, 43(2), 224–230. Yaoi, K., Osaka, N., & Osaka, M. (2009). Is the self special in the dorsomedial prefrontal cortex? An fMRI study. Social Neuroscience, 1, 9. Zahavi, D. (2005). Subjectivity and selfhood: Investigating the First-person perspective. London: The MIT Press. Zhu, Y., Zhang, L., Fan, J., & Han, S. (2007). Neural basis of cultural inﬂuence on self-representation. Neuroimage, 34(3), 1310–1316.

Consciousness and Cognition 20 (2011) 64–74

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Through the looking glass: Self and others q Corrado Sinigaglia a,⇑, Giacomo Rizzolatti b,c,⇑ a b c

University of Milan, Department of Philosophy, via Festa del Perdono 7, I-20122 Milano, Italy University of Parma, Department of Neuroscience, via Volturno 39, I-43100 Parma, Italy IIT (Italian Institute of Technology) Brain Center for Social and Motor Cognition, Parma, Italy

a r t i c l e

i n f o

Article history: Available online 8 January 2011 Keywords: Mirror mechanism Action understanding Self and other Motor system

a b s t r a c t In the present article we discuss the relevance of the mirror mechanism for our sense of self and our sense of others. We argue that, by providing us with an understanding from the inside of actions, the mirror mechanism radically challenges the traditional view of the self and of the others. Indeed, this mechanism not only reveals the common ground on the basis of which we become aware of ourselves as selves distinct from other selves, but also sheds new light on the content of our self and other experience, showing that we primarily experience ourselves and the others in terms of our own and of their motor possibilities respectively. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction The discovery of mirror neurons has been one of the most intriguing and exciting ones in cognitive neuroscience over the last two decades. The deﬁning characteristic of these neurons is that they discharge both when an individual performs a given motor act and when an individual observes someone else performing a similar motor act. It has been argued that these neurons represent a fundamental neural mechanism that not only underlies action understanding (Rizzolatti, Fogassi, & Gallese, 2001) but also, and most importantly, solely allows us to understand the actions of others ‘‘from the inside’’, encoding them in terms of our own motor possibilities (Rizzolatti & Sinigaglia, 2010). Our purpose in the present article is twofold. The ﬁrst is to give a general account of the mirror mechanism. This mechanism has often been misunderstood, by being considered either as a surprising ‘‘trick’’ that magically explains the most diverse behaviors or as an ‘‘annoying’’ neural machinery that perturbs the well-established and comforting distinction between cognition and motion. We will not deal here with the speculations on the ‘‘magic’’ of mirroring. Rather, we will focus on the neural basis of the mirror mechanism in order to dispel some misunderstandings mainly generated by a quite superﬁcial reading of the neurophysiological literature on mirror neurons. Our second aim is to discuss the role of the mirror mechanism, and of the motor system in general, in shaping our sense of self and our sense of others. Generally, it is assumed that the distinction between our sense of self and our sense of others is nothing else but a sub-case of the distinction between the self and the surrounding world, being both based on completely separate neural representations. We will challenge this assumption by arguing that what the mirror mechanism tells us is that the self and the others are so strictly intertwined that, at least at the basic level, they are intimately rooted in a common motor ground. Our article will be subdivided into three main sections. In the ﬁrst one, we will examine the functional properties of the basic mirror mechanism in monkeys and humans. In the second section we will focus on the unique characteristics of mirrorbased action understanding. We will highlight that the mirror mechanism, by mapping the observed actions onto the motor q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap.

⇑ Corresponding authors.

E-mail addresses: [email protected] (C. Sinigaglia), [email protected] (G. Rizzolatti). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.11.012

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

65

repertoire of the observers, provides them with an understanding of those actions that is the same in nature as the ﬁrst-person action understanding they exploit when acting. The third section will investigate whether and to what extent the mirror mechanism may impact on the self and other distinction as well as on the content of our sense of self and of our sense of other, at least at the basic level. We will argue that such a mechanism not only provides out the common ground on the basis of which we become aware of ourselves as selves distinct from other selves, but also sheds new light on the primary content of our self- and other experience, showing that we primarily experience ourselves and the others in terms of our own and their motor possibilities, respectively. 2. The two problems of mirroring The mirror mechanism is a neurophysiological mechanism that transforms sensory information describing the actions of others into a motor format similar to that which the observers endogenously generate when they actually perform those actions. The mirror mechanism is present in many cortical areas and brain centers of birds, monkeys, and humans. The functions of these areas and centres vary to a great extent, ranging from song production, to emotional processes and the organization of goal-directed motor acts. Thus, as other basic mechanisms, the mirror mechanism subserves different functional roles according to the brain center where it is located, its function ranging from the recognition of the song of con-speciﬁcs in birds (Keller & Hahnloser, 2009; Prather, Peters, Nowicki, & Mooney, 2008) to empathy in humans (Gallese, Keysers, & Rizzolatti, 2004). In the present article we will deal exclusively with the neural circuits endowed with the mirror mechanism located in the parietal and frontal lobe of monkeys and humans and primarily devoted to action organization. Notably, we will focus on the circuits underlying the organization of hand motor acts. This circuit consists of two main nodes: area F5 of the frontal lobe and areas AIP/PFG of the inferior parietal lobule (IPL) (Fogassi et al., 2005; Rizzolatti & Luppino, 2001; Rozzi, Ferrari, Bonini, Rizzolatti, & Fogassi, 2008). 2.1. What do mirror neurons encode? The fundamental and most debated issue concerning the mirror mechanism located in the parieto-frontal circuits is its contribution to action understanding. This issue, which may appear unitary, is actually formed by two distinct questions. First: What do mirror neurons code when they discharge in response to the observation of others’ motor behavior? Second: What is the role of this coding in action understanding? The best way to assess what aspect of the motor behavior F5 and AIP/PFG mirror neurons encode is to establish what they encode when they discharge during voluntary motor behavior. In both these conditions the recorded electrical activity is the action potentials generated by the neurons, i.e. their output. It is common knowledge that the output of neurons, regardless of how they are triggered, conveys the same information to other neurons. Thus, once it is determined what aspect of the motor behavior mirror neurons encode, it is also established what they encode when they are triggered by an observed behavior. There is a large amount of evidence that F5 and IPL motor neurons mainly code the goal of the motor acts to be performed (Fogassi et al., 2005; Jeannerod, Arbib, Rizzolatti, & Sakata, 1995; Murata, Gallese, Luppino, Kaseda, & Sakata, 2000; Rizzolatti et al., 1988; Sakata, Taira, Murata, & Mine, 1995). Particularly compelling is the evidence provided by Umiltà et al. (2008). They trained monkeys to grasp objects using two different types of pliers, ‘normal pliers’, which require typical grasping movements of the hand (opening and then closing the ﬁngers), and ‘reverse pliers’, which require hand movements in the opposite order (closing and then opening the ﬁngers). The results showed that the discharge of the recorded neurons correlated in both conditions with the goal of the motor act, regardless of the fact that the hand movements performed to achieve it were exactly the opposite. F5 and IPL mirror neurons do not differ in their motor properties from the motor neurons of the same areas devoid of visual properties (di Pellegrino, Fadiga, Fogassi, Gallese, & Rizzolatti, 1992; Gallese, Fadiga, Fogassi, & Rizzolatti, 1996; Rizzolatti, Fadiga, Gallese, & Fogassi, 1996; Rochat et al., 2010). Thus, when mirror neurons ﬁre in response to action observation they send information to other centers about the goal of the observed actions, exactly as when they are engaged in action execution. Recently, a single neuron study investigated F5 mirror neuron responses to the observation of motor acts performed within (peripersonal space) or outside (extrapersonal space) the reach of the monkey (Caggiano, Fogassi, Rizzolatti, Thier, & Casile, 2009). The results showed that many F5 mirror neurons were differentially modulated by the spatial location of the observed motor act. Some neurons were selective for actions executed in the monkey’s peripersonal space, while others preferred extrapersonal space stimuli. These ﬁndings indicate that goal encoding of mirror neurons may provide the observer with information about where the observed action is performed. Several fMRI studies provided evidence supporting the early ﬁndings of mirror goal encoding in humans (Buccino et al., 2001; Decety, Chaminade, Grèzes, & Meltzoff, 2002; Grafton, Arbib, Fadiga, & Rizzolatti, 1996; Rizzolatti et al., 1996). Gazzola, Rizzolatti, Wicker, and Keysers (2007) instructed volunteers to observe video-clips where either a human or a robot arm grasped objects. In spite of the differences in shape and kinematics between the human and robot arms, the parietofrontal areas endowed with mirror properties were activated in both conditions. These results were recently extended by Peeters et al. (2009) who investigated the mirror activations in response to the observation of human hand, robot hand

66

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

and tool actions. They found the activation of a mirror network formed by intraparietal and ventral premotor cortex bilaterally. In addition, they reported that the observation of tool actions produced a speciﬁc activation of a rostral sector of the left anterior supramarginal gyrus. A parallel, comparative fMRI study carried out on monkeys showed an activation only of the grasping mirror network (Nelissen, Luppino, Vanduffel, Rizzolatti, & Orban, 2005). No tool-speciﬁc area was found. The issue of whether the human parieto-frontal mirror network codes action goals was also addressed by fMRI and TMS studies investigating the activation of motor areas during listening to action-related sounds (Galati et al., 2008; Gazzola, Aziz-Zadeh, & Keysers, 2006; Lewis, Brefczynski, Phinney, Janik, & DeYoe, 2005; see also Lewis, Talkington, Puce, Engel, & Frum, 2010). In particular, Lewis et al. (2005) reported that hearing and categorizing animal vocalizations, as opposed to the sound of hands manipulating tools, preferentially activated the middle portion of the superior temporal gyri bilaterally, while hearing and categorizing tool sounds activated the mirror network. 2.2. What are mirror neurons for? Why should the motor system resonate with the action performed by others, encoding the observed motor goal? The original answer to this question was that the primary function of the mirror neurons is to enable the observer to understand the actions of others (di Pellegrino et al., 1992; Gallese et al., 1996; Rizzolatti, Fadiga, et al., 1996). According to this view, the observation of others’ actions elicits, in the observer’s brain, a motor activation similar to that endogenously occurring during the planning and the execution of those actions. The similarity between these two activations allows the observer to understand directly others’ actions without the necessity of any inferential processing (Rizzolatti et al., 2001). Of course, claiming that mirror neurons are critical for understanding the motor acts done by others does not imply that these neurons magically bear such an understanding per se; rather, this means that their output triggers a complex network of neurons, some of which are involved in the execution of those motor acts. Thus, mirror-based action understanding is rooted in the activation of a complex sensori-motor network (Rizzolatti, Fogassi, & Gallese, 2009). A similar, albeit not identical, network is activated when an individual is thinking or performing that motor act (Jeannerod, 2001). This overlap allows one to immediately understand others by means of a direct matching of the observed motor acts with one’s own motor representation of those acts (Rizzolatti & Sinigaglia, 2008). Strong empirical evidence supporting this view came from two series of experiments in which the meaning of others’ motor acts could be encoded in the absence of visual information describing them. In the ﬁrst study, monkeys heard the sounds of a motor act (such as ripping a piece of paper) without seeing it (Kohler et al., 2002); in the second, monkeys knew that behind a screen there was an object and could see the experimenter’s hand disappearing behind the screen, but could not see the hand/object interaction which represented their triggering feature when the action was performed in full view (Umiltà et al., 2001). The results showed that F5 mirror neurons became active in both cases, thus indicating that their activation reﬂected the comprehension of other’s motor acts, rather than the sensory contingencies of motor act presentation. In order to fully appreciate the contribution of the mirror mechanism to action understanding, it is crucial to compare visual and motor encoding of motor acts. In fact, according to some authors the goal encoding is primarily a function of the cortex within the superior temporal sulcus (STS) (Csibra, 2007; see also Jacob, 2008, 2009). This proposal is based on the properties of STS neurons that, as described by Perrett and colleagues in a series of elegant studies in monkeys (Jellema & Perrett, 2005; Perrett et al., 1989), discharge during the observation of actions done by others. A similar role for STS was also proposed in humans, on the basis of fMRI experiments (for a review see Allison, Puce, and McCarthy (2000) and Puce and Perrett (2003)). There is no doubt that STS neurons play a fundamental role in describing the actions of others. However, it is rather unlikely that the STS exhausts by itself the process of goal encoding relegating the parieto-frontal mirror mechanism to a secondary role in this function. To be a really credible candidate for goal encoding, a cortical region should be characterized by the capacity to encode the goal-relatedness of an action with the greatest degree of generality. Now, the available evidence shows that this capacity characterizes the parieto-frontal mirror neurons but not the STS cells. In fact parietal and frontal mirror neurons encode the goal of the observed motor acts regardless of whether they are performed with the mouth, hand or even tools. No such neurons appear to exist in the STS. Most importantly, the very possibility of the existence of STS neurons encoding actions with the same great degree of generality as mirror neurons is implausible. If a STS neuron selectively encodes the visual features of a given hand action (e.g. grasping), it is very hard to imagine how the same neuron could selectively encode also the visual features of a mouth performing the same motor act. Of course, one could postulate an association process as that described for the temporal lobe (Miyhashita, 1988; Sakay & Miyhashita, 1991). However, also in this case, the association will concern spatio-temporally adjacent visual representations of bodily part movements and not visual representations of the same motor goal obtained with different effectors. In contrast, mirror neurons, in virtue of their motor nature, may be triggered by different visual stimuli (e.g. hand and mouth actions) encoding a common goal (e.g. grasping). Only the presence of a motor scaffold supplying the motor goal-relatedness of observed actions can allow this generalization that goes beyond that achievable by mere visual association. A recent study provides empirical evidence in favor of this point (Cattaneo, Sandrini, & Schwarzbach, 2010). The study employed a TMS adaptation paradigm (Silvanto, Muggleton, & Walsh, 2008). This paradigm is based on the observation that TMS disinhibits habituation, caused by repetitive stimulus presentation, in the stimulated cortical area. Cattaneo et al. (2010) presented participants with ‘‘adapting’’ movies showing hand and foot motor acts and asked them to respond to as quickly as

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

67

possible when the motor act shown in the test picture had the same goal as that in the adapting movies. TMS pulses were delivered over the ventral premotor cortex bilaterally, the left IPL and the left STS. The results showed that the delivery of TMS over both premotor and IPL cortices shortened the reaction times to adapted stimuli, regardless of the effector performing the observed motor act; in contrast, TMS stimulation of the STS induced a shortening of reaction times for adapted motor acts, but only when also the effector was the same. 3. Understanding actions from the inside 3.1. Mirror-based goal understanding What do we really mean by characterizing the function of the mirror mechanism in terms of action understanding? And what type of action understanding do the mirror neurons subserve? To answer these two questions it is necessary to make a distinction, often overlooked, between movement and action mirroring (Rizzolatti, Fadiga, Fogassi, & Gallese, 1999). In fact, not all types of mirroring concern motor goals. There are several mirroring processes (such as, for instance, motor contagion or motor mimicry) that run at a lower level than that of action mirroring. Both movement and action mirroring depend on the mirror mechanism. In both cases the visual information is transformed into a motor format. However, in movement mirroring, the visual information concerns simple movements; in contrast, action mirroring implies the encoding of the motor goal of the observed motor act. There is no evidence that movement mirroring exists in monkeys, while, in contrast, there is overwhelming evidence that it exists in humans. This evidence is based on TMS studies that showed that the observation of meaningless gestures performed in front of an individual elicits an increase of motor evoked potentials (MEPs) in those muscles that are active during the execution of the same gesture (Fadiga, Fogassi, Pavesi, & Rizzolatti, 1995; see also Rizzolatti and Sinigaglia (2008) for a review). By exploring the contribution of the mirror mechanism to action understanding we are dealing with an aspect of mirroring only, that related to high-level motor acts tuned to others’ motor goals. It also goes without saying that when we (and others) claim that the mirror mechanism plays a crucial role in understanding the behavior of others, this does not imply that there are no other mechanisms involved in action understanding. Some of these mechanisms are very basic, relying on the association between a given stimulus and its corresponding effect. For example, one can realize that a gesture might convey a threat, without necessarily transforming it into a motor format. A monkey can be scared when observing somebody throwing a stone towards it in a way that corresponds to its own motor repertoire or a way that the animal never did before (Wood, Glynn, Phillips, & Hauser, 2007; Wood & Hauser, 2008). What count here is the throwing effect rather than the precise gesture mirroring. On the other hand, there is a long tradition that accounts for action understanding by referring uniquely to the capability of individuals to ‘‘read’’ the mind of others, attributing a causal role to their mental states in representing and executing actions. The nature and the format of this ‘‘mindreading’’ are still a matter of controversy (Carruthers & Smith, 1996; Goldman, 2006; Hutto & Ratcliffe, 2007; Malle, Moses, & Baldwin, 2001). There is, however, a certain consensus that this capability is fully ﬂedged only in humans, and most likely absent in monkeys. A crucial point has to be stressed here. There is a fundamental difference between mirror-based action understanding and the understanding of others’ behavior relying either on a lower-order associative mechanism or on a higher-order metarepresentational capability. An example could be helpful to this regard. Imagine that a pianist is demonstrating a given chord or a given sequence of chords on a soundless piano. From time to time, the pianist deceives his students by performing ﬁnger movements that are similar to the chord movements from a motor point of view, but which are devoid of any musical meaning. Mary is an absolute beginner, while John is already a good pianist. The hoax of the teacher will be immediately understood by John, who surprised will ask the teacher what his strange ﬁnger movements are for, while Mary will not able to recognize the difference between the true and the fake chords. In other words, knowing how to play the piano provides a different kind of understanding of the observed movements. This knowledge not only cannot be grounded in a mere associative mechanism, but does not either necessarily imply any explicit mentalizing. Indeed, there is no reason to assume that Mary is less able than John in reading the mind of the pianist, by meta-representing him as having certain propositional attitudes such as a given belief and/or a given desire, in order to account for the fact that she cannot actually understand what the teacher is doing, i.e. whether he is playing true or fake chords. What counts here is the capability to understand the meaning of the observed movements on the basis of one’s own motor repertoire. We have called this kind of understanding of others’ actions as the understanding from the inside (Rizzolatti & Sinigaglia, 2010). This simple example tells us that the richer is our motor repertoire the sharper is our sensitivity to others’ actions, so that our ability to act shapes our experience allowing us to make sense of others’ behavior. This conclusion is in line with evidence coming from a large number of brain-imaging studies (Aglioti, Cesari, Romani, & Urgesi, 2008; Buccino et al., 2004; Calvo-Merino, Glaser, Grèzes, Passingham, & Haggard, 2005; Calvo-Merino, Grèzes, Glaser, Passingham, & Haggard, 2006; Cross, Hamilton, & Grafton, 2006; Haslinger et al., 2006). It has been shown, for example, that viewing videos of classical ballet or capoiera activates differently the mirror mechanism of participants, depending on whether they were experts in classical ballet or in capoeira (Calvo-Merino et al., 2005, 2006; see also Cross et al., 2006). Similar results have been also obtained in a series of experiments on other skilled actions such as piano (Haslinger et al., 2006) or basketball playing (Aglioti et al., 2008), demonstrating that the activation of the mirror mechanism during action observation depends on the observer’s

68

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

motor expertise. As this expertise develops, diversiﬁes and becomes increasingly sophisticated, the ability to understand the actions of others increases, diversiﬁes and becomes increasingly sophisticated. In other words, the more the motor goals are ﬁnely represented, the greater is the signiﬁcance acquired by details of the observed actions, which share the same motor intentional content with the actions the observer might execute. It is due to this sharing that action understanding can become extremely subtle – while still being immediate and without presupposing the meta-representational abilities which are alleged to be at the basis of the classical view of mind reading. 3.2. From motor goals to motor intentions The functional properties of the mirror mechanism for action discussed up to now indicate that its activation reﬂects what is going on hic et nunc. However, there is evidence that parietal and frontal mirror neurons are involved in encoding not only the observed motor acts but also the entire action of which the observed motor acts is part. Single cell recordings from IPL (area PFG) and from the ventral premotor cortex (area F5) of the monkey showed the existence of speciﬁc set of neurons named ‘‘action-constrained’’ neurons (Bonini et al., 2010; Fogassi et al., 2005). These neurons discharge in association with speciﬁc motor acts, but become maximally activated when the coded motor act is embedded into a speciﬁc motor action. Thus, for example, action-constrained grasping neurons strongly discharge when the monkey is grasping a piece of food for bringing it to the mouth, but not when the animal is grasping something for placing it into a container. Most interestingly, many of these action-constrained neurons have mirror properties (Bonini et al., 2010; Fogassi et al., 2005). These neurons selectively discharge when the monkey observes a motor act part of a speciﬁc action (e.g., grasping-for-eating but not grasping-for-placing). Their activation provides, therefore, information not only on the fact that an individual is grasping, but also and most importantly on why the individual is most likely doing it. In virtue of this mechanism the observer, besides recognizing the observed motor act, is also able to anticipate what is the motor intention underlying the whole action. In other words, the observer is able to understand the motor intention with which the agent is doing what he is doing. The mirror mechanism plays a role in intention understanding also in humans. Brain-imaging studies have shown that recognizing the motor intention behind a given motor act activates the right frontal and parietal nodes of the mirror network (Hamilton & Grafton, 2008; Iacoboni et al., 2005). More recently, by using a high-density electrical neuroimaging, the temporal dynamics of brain activations was investigated in individuals observing hand motor acts (i.e. grasping a mug) and attempting to understand the motor intention behind them (Ortigue, Sinigaglia, Rizzolatti, & Grafton, 2010). The motor acts were presented within or without a context. In either case, the results showed a similar brain activation pattern characterized by an early recruitment of the left posterior temporal and inferior parietal cortices followed by a signiﬁcant activity increase in the right temporo-parietal region. It has been suggested that the early strong left hemisphere activation was due to the recruitment of a lateralized mirror network mediating the understanding of the goal of object-directed motor acts. The successive later right hemisphere activation would indicate the involvement of this hemisphere in understanding the intention of others. In addition, an EMG study (Cattaneo et al., 2007) demonstrated, albeit indirectly, that motor intention encoding in humans is based on a motor chain organization similar to that found in monkeys. Typically developing (TD) children were instructed, in one condition, either to grasp a piece of food to eat it or to place it into a container, and in another, to observe an experimenter performing the same actions. The activity of the mouth-opening mylohyoideus (MH) muscle was recorded. The results showed that both the execution and observation of eating actions produced a marked increase of MH muscle activity as early as the reaching phase in grasping-for-eating, while no MH activity was found in the execution and observation of placing actions. In the same EMG study, children with autistic spectrum disorders (ASD) were also asked to execute or observe eating and placing actions. As in the case of TD children, no MH activity was observed in children with ASD during the execution and the observation of placing. During the execution of grasping for eating, MH activation was present, but occurred much later than in children with AS. Furthermore, there was no MH activation at all when children with ASD observed grasping for eating done by another individual. These ﬁndings indicate that children with ASD have a severe impairment in motor organization leading to a deﬁcit in chaining motor acts into intentional actions. Furthermore, their intentional motor chains are not active during action observation. In other words, the intentions of others do not ‘‘enter’’ into their mirror system. Intentions are not understood from the inside; they can be only captured from the outside, by means of inferences. This interpretation is supported by a behavioral study (Boria et al., 2009) showing that in order to understand other people’s intentions, ASD children tend to rely not on the observed motor behavior, but on the semantics of the object that is manipulated or on the context in which the observed motor act takes place. Guessing others’ intentions, however, on the basis of object semantics gives children with ASD a rigid and often unreliable way of understanding others. It may be that inferential processing based on additional contextual or social information present in the environment could help ASD children overcome the pitfalls of an object-based intention guessing mechanism. However, even with this additional inferential processing the comprehension of others could hardly reach the reliability as well as the ﬁrst-person action understanding based on one’s own motor expertise. As already discussed for action understanding, claiming that the mirror mechanism plays an important role in processing others’ intentions is not tantamount to state that mirror-based intention understanding covers all varieties of understanding

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

69

others’ intentions. Nor does it involve the assumption that every kind of intention understanding depends on (is related to) the activation of the mirror mechanism. This assumption is not only empirically groundless, but also ends up completely failing to grasp the originality of mirror-based intention understanding. What the mirror mechanism properties suggest is that the nature and the scope of mirror-based intention understanding are strictly tied to the nature and scope of the motor intention that makes a series of motor acts (e.g. reaching, grasping, bringing-to-the-mouth) parts of a whole action (reaching for grasping for bringing-to-the-mouth), regardless of whether such an action is executed by an individual or simply observed while being carried out by another individual (Gallese, 2007; Rizzolatti & Sinigaglia, 2007; Sinigaglia, 2009). 4. Mirroring self and others The second aim of our study was to examine whether the mirror mechanism is involved in shaping our sense of self and our sense of others, at least at a basic level. As for the function of the mirror mechanism, also in this case it would be useful to subdivide the issue into two separate problems that, although strictly intertwined, are different in nature. The ﬁrst problem, widely discussed in the literature, concerns the impact of the mirror mechanism on the distinction between the self and the others; the second problem, much less debated and yet not less important than the former, concerns the kind of sense of self and of sense of others that the mirror mechanism in particular, and the motor system in general, make possible. 4.1. The mirror roots of the self and other distinction As far as the ﬁrst problem is concerned, it seems almost obvious to assume, at least at ﬁrst glance, that the attribution of actions to the self or to the others should be based on separate neural representations. Two distinct neural networks should underlie our and others’ actions. However, it is just this kind of assumptions that the discovery of the mirror mechanism has radically undermined. Indeed, what the functional properties of the mirror mechanism tell us is that the self and the other are so strictly intertwined that, even at the basic level, self- and other-attribution processes are mutually related each other, being both intimately rooted in a common motor ground (Gallese et al., 2004; Gallese, Rochat, Cossu, & Sinigaglia, 2009; Rizzolatti et al., 2001; Sinigaglia, 2009). More precisely, the mirror mechanism clearly indicates that (i) in order to be attributed either to the self or to the others, actions should be represented as actual motor possibilities for the agent and (ii) the distinction between self and other should stem from their shared motor goals and motor intentions, because it is on the basis of this common motor ground that we are able to differentiate ourselves from the other selves. Let us take a closer look to the ﬁrst point. It is immediate to realize that mirror-based action observation is but one of the situations demonstrating that motor representation of action is the prerequisite of the self (and other) attribution (Jeannerod, 2001, 2003). Take, for instance, the case of motor imagery. Indeed, there is a large amount of evidence that motor imagery recruits cortical (e.g. dorsal and ventral premotor cortex, primary motor cortex) and subcortical (cerebellum, basal ganglia) areas typically involved in action execution (Jeannerod & Decety, 1995; Jeannerod & Frack, 1999; see also Jeannerod (2009) for a review). In addition, many temporal and spatial characteristics of executed actions have been demonstrated to be also present when those actions are simply imagined. For example, it has been shown that the time it takes to imagine to walk to a given place is the same as that which it takes to actually walk to that place (Decety, Jeannerod, & Prablanc, 1989). Similarly, the feasibility of an object-related action critically depends on the spatial orientation of the object with the respect to the agent when we are both executing and simply imagining that action (Pearson, 1994; Frak, Paulignan, & Jeannerod, 2001). These ﬁndings strongly suggest that action representations like motor images should be construed as being ‘‘in fact actions on their own right’’ (Jeannerod, 2003; see also Jeannerod, 2001). This is also true for the goal-related motor representation evoked by the observation of actions performed by others. To this regard, it is worth mentioning here that ﬁneness-of-grain of the content of the motor representations produced by the observation of an action performed by someone else might be different from those endogenously generated by imagining oneself performing the same action, being motor imagery closer to action execution than mirror-based action observation. In a recent TMS study (Cattaneo, Caruana, Jezzini, & Rizzolatti, 2009), motor evoked potentials (MEPs) were recorded from the right opponens pollicis (OP) muscle in participants either observing the experimenter using normal or reverse pliers to grasp objects or imagining themselves grasping objects with the same tool. The results showed that during the observation of the grasping action with normal and reverse pliers, the MEPs from OP were modulated by the action goal, regardless of whether its achievement required an opposite sequence of ﬁnger movements (extension/ﬂexion and ﬂexion/extension, respectively). On the contrary, during motor imagery, the MEPs amplitudes, regardless of the pliers used, reﬂected the muscular patterns involved in the execution of that action. Regardless of whether the action content of the mirror representations might be more goal-centerd than that of the corresponding motor images, mirror-based action observation and motor imagery are both rooted in one’s own motor repertoire, providing one with a comprehension of the observed/imagined actions from the inside, that is, as one’s own motor possibilities. Indeed, they recruit largely overlapping motor sources. Of course, this does not imply that motor imagery and mirror-based action observation have to be construed as the same phenomenon. Quite the contrary. Differently from the former, the latter does not require an endogenously deliberate action representation, being the mirror mechanism

70

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

automatically activated by the observation of others’ actions, provided that they belong to the observer’s motor repertoire. Nevertheless, like motor imagery, mirror-based action observation demonstrates that movements and motor acts can be considered part of an action even when they are not actually executed, being enough that they are represented as goal-related motor possibilities for an agent. What is so special about the mirror mechanism is that such goal-related motor possibilities are evoked when we are merely observing someone else acting. The actions of others are thus mapped onto our own motor possibilities: because of this mapping we can share a motor common ground with the others, and it is in virtue of such a sharing that we may represent a given motor action as our own or as being performed by someone else. On the other hand, mirror-based sharing of action can also account for the reason why in some circumstances the self/other distinction might partly fail. Indeed, several studies showed that participants erroneously self-attributed motor acts performed by others when embedded in situations where the origin of those motor acts was artiﬁcially made uncertain (see Jeannerod (2003, 2009) for a review). For instance, in a series of experiments (Daprati et al., 1997), participants were asked to execute simple ﬁnger movements without any direct visual control of their moving hand and to watch on a screen the image of a gloved hand which could be either their own or an alien hand executing the same or different movements. The results showed that participants made more errors when they were presented with an alien hand performing the same movements as their own, self-attributing others’ movements in about one-third of the trials. Similar ﬁndings have been obtained in another series of experiments (Van den Bos & Jeannerod, 2002) where both the participant’s hand and the experimenter’s hand were simultaneously presented, executing either the same movements, different movements or no movement at all. The overall pattern of results was that participants tended to self-attribute others’ movements more than to attribute their own movements to others, especially when the cues for discriminating between the two hands had been suppressed. On the basis of this kind of experiments as well as of a set of studies on action misattribution in schizophrenic patients (Frith, Blakemore, & Wolpert, 2002; Fourneret et al., 2002; Franck et al., 2001; see also Farrer and Franck (2007) for a review), it has been proposed that the self/other distinction could depend on a mechanism (who-mechanism) discriminating action representations endogenously generated from those externally evoked (Georgieff & Jeannerod, 1998; Jeannerod, 2003). Although the locution ‘‘who-mechanism’’ (or even ‘‘who-system’’) might give rise to some misunderstanding, suggesting the notion of putative brain centers univocally devoted to process self- or other-related information, the above reviewed ﬁndings clearly corroborate the critical role of the cortical motor system in encoding both our sense of self and our sense of others, at least at a basic level. 4.2. Being reﬂected by our own motor possibilities The functional properties of the mirror mechanism deeply impact not only on the self/other distinction but also and above all on the way in which we should think of our basic sense of self and of our basic sense of others. Note that, while the implications of the mirror mechanism for self- and/or other-attribution of action have been largely investigated, little research has been devoted to exploring whether and to what extent the mirror mechanism in particular and the motor system in general tell us something about our primary making sense of ourselves and of others. This is even more striking given that the two issues are strictly related, and a suitable answer to the ﬁrst one cannot be given leaving aside the understanding and the solution to the second one. To get going, it could help us to take into consideration another kind of action representation, that is, the action representation involved in the perception of visual affordances. As it is well known, the notion of affordance refers to the power of the environment to furnish the observer action possibilities (Gibson, 1979). An affordance is not a mere physical property, rather it indicates the action opportunities that a given object may offer to an organism that is able to use them. Thus, for instance, an object such as a mug affords several possibilities of motor acts: it can be grasped by its handle, by its body, by its upper part, and according how it has been grasped it can be brought to the mouth, moved away, thrown, and so on. Now, there is a large amount of neurophysiological and neuroimaging evidence that the sight of a graspable object such as a mug immediately retrieves the suitable set of grasping-related motor representations, even in the absence of any effective interaction and also any explicit intention to act (Craighero, Fadiga, Rizzolatti, & Umiltà, 1999). From a neurophysiological point of view, this can be accounted for by means of a mechanism transforming the objectual visual features into the corresponding set of motor acts. Single cell recordings from the ventral premotor cortex (area F5) and IPL (area AIP) of the monkey’s brain showed that this mechanism is instantiated by a special class of visuo-motor neurons (canonical neurons) that respond to the visual presentation of objects of different size and shape, even when the monkey is just ﬁxating them without being required to grasp them (Jeannerod et al., 1995; Murata et al., 1997; Raos, Umiltà, Fogassi, & Gallese, 2006; Rizzolatti et al., 1988; Umiltà, Brochier, Spinks, & Lemon, 2007). Similar results have been obtained also in humans. Brain-imaging studies demonstrated that the visual presentation of a graspable object automatically recruits the cortical motor system, even in the absence of any motor output (Buccino, Sato, Cattaneo, Roda, & Riggio, 2009; Chao & Martin, 2000; Grafton, Fadiga, Arbib, & Rizzolatti, 1997; Grèzes, Tucker, Armony, Ellis, & Passingham, 2003). Taken together, these ﬁndings demonstrate that the perception of an object may elicit a motor activation in the observer’s brain even in the absence of any overt motor behavior, thus indicating that the object is encoded in the same way when perceiving and acting upon it (Rizzolatti & Gallese, 1997). This suggests not only that object perception is strictly intertwined with action, but also that action constitutively shapes the content of perception, characterizing the perceived object in terms of the motor acts it may afford.

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

71

Note that the notion that object perception is, at the basic level, nothing else but a potential form of action and that as such it provides us with a primary way of being engaged with the surrounding world has important consequences for the account of the way in which we experience ourselves (Costantini & Sinigaglia, 2011). Indeed, in perceiving something as graspable, throwable or kickable we become aware of ourselves as of the selves that can grasp, throw or kick. This implies that we do not experience ourselves as a given entity (e.g. a physical body) and then realize that such an entity can grasp or kick the object in front of us. Rather, it means that in perceiving something as graspable or as kickable we are given to ourselves as being-able-to-grasp or as being-able-to-kick, that is, as a set of motor possibilities. In other words, our experience of the surrounding things cannot but be accompanied by the experience of ourselves as a ‘‘source’’ for action (Gallese & Sinigaglia, 2010). Like motor imagery, affordance perception also shows that there is no need for individuals to perform any movement in order to become aware of their motor possibilities as their own. Such an awareness does not rely, of course, on the various forms of self-reﬂection that are prerequisite of a full-ﬂedged sense of self, because it is basic enough to be non-reﬂective in nature (Bermúdez, 1998, 2002; Gallagher, 2003; Gallagher & Zahavi, 2008). When we are perceiving something as graspable, we are implicitly aware of ourselves as being-able-to-grasp, even if we do not direct our attention to it. There is no need for reﬂection here, i.e. for taking a step back from affordance perception in order to consider ourselves and our motor possibilities. Indeed, affordance perception carries with it a non-reﬂective awareness of ourselves and our own motor possibilities. However, if one really cannot help employing the notion of reﬂection in order to highlight the self-relativity that is constitutive of any form of self-awareness, starting from the basic ones, one may take it according to its ‘‘optical meaning’’, as it has been proposed by Heidegger (1988, p. 159). Here, reﬂection means ‘‘to show itself in a reﬂection from something’’ (Heidegger, 1988, p. 159). Thus, we could say that we primarily experience ourselves inasmuch as we are reﬂected from the surrounding things that provide us with a set of our own motor possibilities. In Heidegger’s words: ‘‘Each one of us is what he pursues and cares for. In everyday terms, we understand ourselves and our existence by means of the activities we pursue and the things we take care of’’ (Heidegger, 1988, p. 159). By means of our own ‘‘activities’’, however, we understand not only ourselves but also the others, at least inasmuch as their own ‘‘activities’’ – according to our terminology, their own motor possibilities – can be ‘‘reﬂected’’ from our own ones. Indeed, what the mirror mechanism tells us is that the very same motor possibilities that shape, at least at the basic level, our sense of self also shape our sense of other selves inasmuch their motor possibilities can be mapped onto our own ones. Of course, just like the selfness of the self, the otherness of the others as it is reﬂected from our own motor possibilities cannot be construed as covering all the distinct layers characterizing our full-ﬂedged sense of others. What we become aware of by resonating with the others is the range and the nature of their own source for action, that is, the range and the nature of their motor possibilities as reﬂected by our own motor goals and motor intentions. Two points deserve our attention here. First: because of its motor ground, reﬂecting others’ possibilities from our own ones may occur at different degrees of generality, ranging from the low-level forms of single movement resonance to the high-level modalities of being tuned to others’ motor goals and intentions. To the latter regard, we already mentioned that the richer and more diversiﬁed is our motor repertoire the sharper is our sensitivity to others’ actions, so that our capability to make sense of others turns out to be rooted in our capability to make sense of ourselves. It follows that, if more individuals share the same motor repertoire, the richer and more diversiﬁed such a motor repertoire is, the more these individuals will be able to be mutually reﬂected by their own motor possibilities, thus coming to a more and more ﬁne-grained understanding from the inside of each other. In other words, the more individuals share their own motor repertoire with each other, the more ﬁne-grained is the experience they make of action possibilities when these action possibilities are relative both to their own selves and to other selves. The second point concerns the difference between motor goals and motor intentions. The chain organization of the cortical motor system provides the mirror mechanism with the possibility to encode not only single motor goals per se (e.g. reaching, grasping, holding, etc.), but also motor goals as being intentionally related one to another, thus representing the motor intention with which they might be achieved (e.g. reaching for grasping for bringing-to-the-mouth or reaching for grasping for moving-away). The richness of our motor repertoire does not depend only on the ﬁneness-ofgrain of motor goals representation; rather, it essentially relies on our capability to represent from the inside more and more complex goal architectures, recruiting them both when we perform a given action and when we observe someone else performing it. This capability critically contributes to shaping our experience of ourselves and of other selves, providing us with a multilayered motor representation both of our own and of others’ action possibilities. We propose that sharing such a motor representation paves the way for the higher-level forms of self- and other-awareness generally thought to be at the core of our full-ﬂedged sense of self and sense of others, given that it enables us to make experience both of ourselves and of others not only as acting selves, that is as selves endowed with a set of goal-related movements, but also and above all as intending selves, that is as selves able to perform and to represent those movements with speciﬁc motor intentions. 5. Concluding remarks In the present article we ﬁrst addressed the issue of the mirror mechanism as providing us with an original and primary way to understand the actions of others, to then investigate its impact on our basic sense of self as well as on our basic sense of others.

72

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

In order to dispel some misunderstandings present in the literature, we scrutinized the functional properties of the mirror neurons, directly facing what we labeled the two real problems of mirroring, that is, what mirror neurons encode and what they are for. Both the motor format of mirror neuron encoding and the mirror neuron sensitivity to the motor goals and the motor intentions of actions, regardless of whether those actions are actually executed or simply observed, strongly support the notion that the mirror mechanism provides us with an understanding from the inside of the actions of others, making sense of their behavior on the basis of our own motor possibilities. The highlight of the nature and of the range of such action understanding allowed us also to assess the impact of the mirror mechanism on the self and other distinction as well as its role in shaping our basic sense of self and our basic sense of others. To this regard, we challenged the traditional view according to which the attribution of actions to the self or to the others should be based on separate neural representations. Far from relying on two radically distinct neural networks, the attribution of a given action to ourselves or to others appears to be rooted in our capability to represent that action as our own actual motor possibilities. This common motor ground also plays a critical role in shaping our basic sense of self and our basic sense of others, being both constitutively intertwined with one another. Indeed, what we become aware of by mirroring the others is the range and the nature of their own motor possibilities as reﬂected by our own motor goals and motor intentions. Of course, by claiming a primacy of the mirror mechanism in strictly binding our sense of self with our sense of others we do not mean that such mechanism allows us to account for all the varieties of our self and other experience. Nevertheless, we believe that understanding how the mirror mechanism in particular and the motor system in general contribute to our making sense both of ourselves and of others might pave the way to a deep reappraisal of different aspects of our full-ﬂedged sense of self and sense of others, forcing us to rethink most of them.

References Aglioti, S. M., Cesari, P., Romani, M., & Urgesi, C. (2008). Action anticipation and motor resonance in elite basketball players. Nature Neuroscience, 11(9), 109–116. Allison, T., Puce, A., & McCarthy, G. (2000). Social perception from visual cues: Role of the STS region. Trends in Cognitive Sciences, 4, 267–278. Bermúdez, J.-L. (1998). The paradox of self-consciousness. Cambridge, MA: MIT Press. Bermúdez, J.-L. (2002). Sources of self-consciousness. Proceedings of the Aristotelian Society, 102, 87–107. Bonini, L., Rozzi, S., Serventi, F. U., Simone, L., Ferrari, P. F., & Fogassi, L. (2010). Ventral premotor and inferior parietal cortices make distinct contribution to action organization and intention understanding. Cerebral Cortex, 20, 1372–1385. Boria, S., Fabbri-Destro, M., Cattaneo, L., Sparaci, L., Sinigaglia, C., Santelli, E., et al (2009). Intention understanding in autism. PloS ONE, 4(5), e5596. Buccino, G., Binkofski, F., Fink, G. R., Fadiga, L., Fogassi, 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., Lui, F., Canessa, N., Patteri, I., Lagravinese, G., Benuzzi, F., et al (2004). Neural circuits involved in the recognition of actions performed by nonconspeciﬁcs: An fMRI study. Journal of Cognitive Neuroscience, 16, 114–126. Buccino, G., Sato, M., Cattaneo, L., Roda, F., & Riggio, L. (2009). Broken affordances, broken objects: A TMS study. Neuropsychologia, 47, 3074–3078. Caggiano, V., Fogassi, L., Rizzolatti, G., Thier, P., & Casile, A. (2009). Mirror neurons differentially encode the peripersonal and extrapersonal space of monkeys. Science, 324, 403–406. Calvo-Merino, B., Glaser, D. E., Grèzes, J., Passingham, R. E., & Haggard, P. (2005). Action observation and acquired motor skills: An FMRI study with expert dancers. Cerebral Cortex, 15, 1243–1249. Carruthers, P., & Smith, P. K. (Eds.). (1996). Theories of theories of mind. Cambridge: Cambridge University Press. Calvo-Merino, B., Grèzes, J., Glaser, D. E., Passingham, R. E., & Haggard, P. (2006). Seeing or doing? Inﬂuence of visual and motor familiarity in action observation. Current Biology, 16, 1905–1910. Cattaneo, L., Caruana, F., Jezzini, H., & Rizzolatti, G. (2009). Representation of goal and movements without overt motor behavior in the human motor cortex: A TMS study. Journal of Neuroscience, 29, 11134–11138. Cattaneo, L., Fabbi-Destro, M., Boria, S., Pieraccini, C., Monti, A., Cossu, G., et al (2007). Impairment of actions chains in autism and its possible role in intention understanding. Proceedings of the National Academy of Sciences, 104, 17825–17830. Cattaneo, L., Sandrini, M., & Schwarzbach, J. (2010). State-dependent TMS reveals a hierarchical representation of observed acts in the temporal, parietal and premotor cortices. Cerebral Cortex, 20, 2252–2258. Chao, L. L., & Martin, A. (2000). Representation of manipulable man-made objects in the dorsal stream. NeuroImage, 12, 478–484. Costantini, M., & Sinigaglia, C. (2011). Grasping affordance. A window onto social cognition. In A. Seeman (Ed.), Joint Attention: New Developments. Cambridge MA: MIT Press. Craighero, L., Fadiga, L., Rizzolatti, G., & Umiltà, C. (1999). Action for perception: A motor-visual attentional effect. Journal of Experimental Psychology and Human Perceptual Performance, 25, 1673–1692. Cross, E. S., Hamilton, A. F., & Grafton, S. T. (2006). Building a motor simulation de novo: Observation of dance by dancers. NeuroImage, 31(3), 1257–1267. Csibra, G. (2007). Action mirroring and action understanding: An alternative account. In P. Haggard, Y. Rosetti, & M. Kawato (Eds.). Sensorimotor foundations of higher cognition. Attention and performance (Vol. XII, pp. 453–459). Oxford: Oxford University Press. Daprati, E., Franck, N., Georgieff, N., Proust, J., Pacherie, E., Daléry, J., et al (1997). Looking for the agent: An investigation into consciousness of action and self-consciousness in schizophrenic patients. Cognition, 65, 71–86. Decety, J., Chaminade, T., Grèzes, J., & Meltzoff, A. N. (2002). A PET exploration of the neural mechanism involved in reciprocal imitation. NeuroImage, 15, 265–272. Decety, J., Jeannerod, M., & Prablanc, C. (1989). The timing of mentally represented actions. Behavioural Brain Research, 34, 35–42. di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V., & Rizzolatti, G. (1992). Understanding motor events: A neurophysiological study. Experimental Brain Research, 91, 176–180. Fadiga, L., Fogassi, L., Pavesi, G., & Rizzolatti, G. (1995). Motor facilitation during action observation: A magnetic stimulation study. Journal of Neurophysiology, 73, 2608–2611. Farrer, C., & Franck, N. (2007). Self-monitoring in schizophrenia. Current Psychiatry Review, 3, 243–251. Fogassi, L., Ferrari, P. F., Gesierich, B., Rozzi, S., Chersi, F., & Rizzolatti, G. (2005). Parietal lobe: From action organization to intention understanding. Science, 302, 662–667. Fourneret, P., de Vignemont, F., Franck, N., Slachevsky, A., Dubois, B., & Jeannerod, M. (2002). Perception of self-generated action in schizophrenia. Cognitive Neuropsychiatry, 7, 139–156.

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

73

Frak, V. G., Paulignan, Y., & Jeannerod, M. (2001). Orientation of the opposition axis in mentally simulated grasping. Experimental Brain Research, 136, 120–127. Franck, N., Farrer, C., Georgieff, N., Marie-Cardine, M., Daléry, J., D’Amato, T., et al (2001). Defective recognition of one’s own actions in schizophrenic patients. American Journal of Psychiatry, 158, 454–459. Frith, C. D., Blakemore, S. J., & Wolpert, D. M. (2002). Abnormalities in the awareness and control of action. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 355, 1771–1788. Galati, G., Committeri, G., Spitoni, G., Aprile, T., Di Russo, F., Pitzalis, S., et al (2008). A selective representation of the meaning of actions in the auditory mirror system. NeuroImage, 40, 1274–1286. Gallagher, S. (2003). Bodily self-awareness and object perception. Theoria et Historia Scientiarum: International Journal for Interdisciplinary Studies, 7, 53–68. Gallagher, S., & Zahavi, D. (2008). The phenomenological mind. An introduction to philosophy of mind and cognitive science. London: Routledge. Gallese, V. (2007). Before and below theory of mind: Embodied simulation and the neural correlates of social cognition. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 362, 659–669. Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action recognition in the premotor cortex. Brain, 119, 593–609. Gallese, V., Keysers, C., & Rizzolatti, G. (2004). A unifying view of the basis of social cognition. Trends in Cognitive Sciences, 8, 396–403. Gallese, V., Rochat, M., Cossu, G., & Sinigaglia, C. (2009). Motor cognition and its role in the phylogeny and ontogeny of action understanding. Developmental Psychology, 45(1), 103–113. Gallese, V., & Sinigaglia, C. (2010). The bodily self as power for action. Neuropsychologia, 48, 746–755. Gazzola, V., Aziz-Zadeh, L., & Keysers, C. (2006). Empathy and the somatotopic auditory mirror system in humans. Current Biology, 16, 1824–1829. Gazzola, V., Rizzolatti, G., Wicker, B., & Keysers, C. (2007). The anthropomorphic brain: The mirror neuron system responds to human and robotic actions. NeuroImage, 35, 1674–1684. Georgieff, N., & Jeannerod, M. (1998). Beyond consciousness of external reality: A ‘‘Who’’ system for consciousness of action and self-consciousness. Consciousness and Cognition, 7, 465–477. Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton-Mifﬂin. Goldman, A. (2006). Simulating minds: The philosophy, psychology and neuroscience of mindreading. Oxford: Oxford University Press. Grafton, S. T., Arbib, M. A., Fadiga, L., & Rizzolatti, G. (1996). Localization of grasp representations in humans by positron emission tomography: Observation versus imagination. Experimental Brain Research, 12, 103–111. Grafton, S. T., Fadiga, L., Arbib, M. A., & Rizzolatti, G. (1997). Premotor cortex activation during observation and naming of familiar tools. Neuroimage, 6(4), 231–236. Grèzes, J., Tucker, M., Armony, J., Ellis, R., & Passingham, R. E. (2003). Objects automatically potentiate action: An fMRI study of implicit processing. European Journal of Neuroscience, 17, 2735–2740. Hamilton, A. F. C., & Grafton, S. T. (2008). Action outcomes are represented in human inferior frontoparietal cortex. Cerebral Cortex, 18, 1160–1168. Haslinger, B., Erhard, P., Altenmuller, E., Schroeder, U., Boecker, H., & Ceballos-Baumann, A. O. (2006). Transmodal sensorimotor networks during action observation in professional pianists. Journal of Cognitive Neuroscience, 17, 282–293. Heidegger, M. (1988). Basic problems of phenomenology. Bloomington: Indiana University Press (Engl. Transl. by A. Hofstadter). Hutto, D., & Ratcliffe, M. M. (Eds.). (2007). Folk psychology re-assessed. Dordrecht: Springer. Iacoboni, M., Molnar-Szakacs, I., Gallese, V., Buccino, G., Mazziotta, J., & Rizzolatti, G. (2005). Grasping the intentions of others with one’s owns mirror neuron system. PLOS Biology, 3, 529–535. Jacob, P. (2008). What do mirror neurons contribute to human social cognition? Mind & Language, 23, 190–223. Jacob, P. (2009). The tuning-fork model of human social cognition: A critique. Consciousness and Cognition, 18, 229–243. Jeannerod, M. (2001). Neural simulation of action: A unifying mechanism for motor cognition. NeuroImage, 14, 103–109. Jeannerod, M. (2003). The mechanism of self-recognition in humans. Behavioural Brain Research, 142, 1–15. Jeannerod, M. (2009). The sense of agency and its disturbances in schizophrenia: A reappraisal. Experimental Brain Research, 192, 527–532. Jeannerod, M., Arbib, M. A., Rizzolatti, G., & Sakata, H. (1995). Grasping objects: The cortical mechanisms of visuomotor transformation. Trends in Neuroscience, 18, 314–320. Jeannerod, M., & Decety, J. (1995). Mental motor imagery: A window into the representational stages of action. Current Opinion in Neurobiology, 5, 727–732. Jeannerod, M., & Frack, V. G. (1999). Mental simulation of action in human subjects. Current Opinion in Neurobiology, 9, 735–739. Jellema, T., & Perrett, D. I. (2005). Neural basis for the perception of goal directed actions. In A. Easton & N. J. Emery (Eds.), The cognitive neuroscience of social behavior (pp. 81–112). Hove/New York: Psychology Press. Keller, G. B., & Hahnloser, R. H. (2009). Neural processing of auditory feedback during vocal practice in a songbird. Nature, 457, 187–190. Kohler, E., Keysers, C., Umiltà, M. A., Fogassi, L., Gallese, V., & Rizzolatti, G. (2002). Hearing sounds, understanding actions: Action representation in mirror neurons. Science, 297, 846–848. Malle, B. F., Moses, J. L., & Baldwin, D. A. (Eds.). (2001). Intentions and intentionality: Foundations of social cognition. Cambridge, MA: MIT Press. Lewis, J. W., Brefczynski, J. A., Phinney, R. E., Janik, J. J., & DeYoe, E. A. (2005). Distinct cortical pathways for processing tool versus animal sounds. Journal of Neuroscience, 25, 5148–5158. Lewis, J.W., Talkington, W.J., Puce, A., Engel, L.R., Frum, C. (2010). Cortical networks representing object categories and high-level attributes of familiar realworld action sounds. Journal of Cognitive Neuroscience September 2. [Epub ahead of print]. Miyhashita, Y. (1988). Neuronal correlate of visual associative long-term memory in the primate temporal cortex. Nature, 335, 817–820. Murata, A., Fadiga, L., Fogassi, L., Gallese, V., Raos, V., & Rizzolatti, G. (1997). Object representation in the ventral premotor cortex (Area F5) of the monkey. Journal of Neurophysiology, 78, 2226–2230. Murata, A., Gallese, V., Luppino, G., Kaseda, M., & Sakata, H. (2000). Selectivity for the shape, size and orientation of objects in the hand-manipulation-related neurons in the anterior intraparietal (AIP) area of the macaque. Journal of Neurophysiology, 83, 2580–2601. Nelissen, K., Luppino, G., Vanduffel, W., Rizzolatti, G., & Orban, G. A. (2005). Observing others: Multiple action representation in the frontal lobe. Science, 310, 332–336. Ortigue, S., Sinigaglia, C., Rizzolatti, G., & Grafton, S. T. (2010). Understanding actions of others: The electrodynamics of the left and right hemispheres. A high-density EEG neuroimaging study. PloS ONE, 5(8), e12160. doi:10.1371. Pearson, L. M. (1994). Temporal and kinematic properties of motor behavior reﬂected in mentally simulated action. Journal of Experimental Psychology and Human Perceptual Performance, 20, 709–730. Peeters, R., Simone, L., Nelissen, K., Fabbri-Destro, M., Vanduffel, W., Rizzolatti, G., et al (2009). The representation of tool use in humans and monkeys: common and unique human features. Journal of Neuroscience, 29, 11523–11539. Perrett, D. I., Harries, M. H., Bevan, R., Thomas, S., Benson, P. J., Mistlin, A. J., et al (1989). Frameworks of analysis for the neural representation of animate objects and actions. Journal of Experimental Biology, 146, 87–113. Prather, J.F., Peters, S., Nowicki, S., & Mooney, R. (2008). Precise auditory-vocal mirroring in neurons for learned vocal communication. Nature i, 249–250. Puce, A., & Perrett, D. (2003). Electrophysiology and brain imaging of biological motion. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 358, 435–445. Raos, V., Umiltà, M. A., Fogassi, L., & Gallese, V. (2006). Functional properties of grasping-related neurons in the ventral premotor area F5 of the macaque monkey. Journal of Neurophysiology, 95, 709–729. Rizzolatti, G., Camarda, R., Fogassi, M., Gentilucci, M., Luppino, G., & Matelli, M. (1988). Functional organization of inferior area 6 in the macaque monkey: II. Area F5 and the control of distal movements. Experimental Brain Research, 71, 491–507. Rizzolatti, G., Fadiga, L., Fogassi, L., & Gallese, V. (1999). Resonance behaviors and mirror neurons. Archives Italiennes de Biologie, 137, 85–100.

74

C. Sinigaglia, G. Rizzolatti / Consciousness and Cognition 20 (2011) 64–74

Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996). Premotor cortex and the recognition of motor actions. Cognitive Brain Research, 3, 131–141. Rizzolatti, G., Fadiga, L., Matelli, M., Bettinardi, V., Paulesu, E., Perani, D., et al (1996). Localization of grasp representation in humans by PET: Observation versus execution. Experimental Brain Research, 11, 246–252. Rizzolatti, G., Fogassi, L., & Gallese, V. (2001). Neurophysiological mechanisms underlying the understanding and imitation of action’. Nature Reviews Neuroscience, 2, 661–670. Rizzolatti, G., Fogassi, L., & Gallese, V. (2009). The mirror neuron system: A motor-based mechanism for action and intention understanding. In M. Gazzaniga (Ed.), The cognitive neurosciences (pp. 625–640). Cambridge, MA: MIT Press. Rizzolatti, G., & Gallese, V. (1997). From action to meaning: a neurophysiological perspective. In J.-L. Petit (Ed.), Les neurosciences et la philosophie de l’action (pp. 217–229). Paris: Vrin. Rizzolatti, G., & Luppino, G. (2001). The cortical motor system. Neuron, 31, 889–901. Rizzolatti, G., & Sinigaglia, C. (2007). Mirror neurons and motor intentionality. Functional Neurology, 22, 205–210. Rizzolatti, G., & Sinigaglia, C. (2008). Mirrors in the brain. How our Minds share Actions and Emotions. Oxford: Oxford University Press. Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the parieto-frontal mirror circuit: Interpretations and misinterpretations. Nature Reviews Neuroscience, 11, 264–274. Rochat, M.J., Caruana, F., Jezzini, A., Escola, L., Intskirveli, I., Grammont, F., et al. (2010). Responses of mirror neurons in area F5 to hand and tool grasping observation. Experimental Brain Research June 26 [Epub ahead of print]. Rozzi, S., Ferrari, P. F., Bonini, L., Rizzolatti, G., & Fogassi, L. (2008). Functional organization of inferior parietal lobule convexity in the macaque monkey: Electrophysiological characterization of motor, sensory and mirror responses and their correlation with cytoarchitectonic areas. European Journal of Neuroscience, 28, 1569–1588. Sakata, H., Taira, M., Murata, A., & Mine, S. (1995). Neural mechanisms of visual guidance of hand action in the parietal cortex of the monkey. Cerebral Cortex, 5, 429–438. Sakay, K., & Miyhashita, Y. (1991). Neural organization for long-term memory of paired associates. Nature, 354, 152–155. Silvanto, J., Muggleton, N., & Walsh, V. (2008). State dependency in brain stimulation studies of perception and cognition. Trends in Cognitive Sciences, 12, 447–454. Sinigaglia, C. (2009). Mirror in action. Journal of Consciousness Studies, 16, 309–334. Umilta, M. A., Brochier, T., Spinks, R. L., & Lemon, R. N. (2007). Simultaneous recording of macaque premotor and primary motor cortex neuronal populations reveals different functional contributions to visuomotor grasp. Journal of Neurophysiology, 98(1), 488–501. Umiltà, M. A., Escola, L., Intskirveli, I., Grammont, F., Rochat, M., Caruana, F., et al (2008). How pliers become ﬁngers in the monkey motor system. Proceedings of the National Academy of Sciences, 105, 2209–2213. Umiltà, M. A., Kohler, E., Gallese, V., Fogassi, L., Fadiga, L., Keysers, C., et al (2001). ‘‘I know what you are doing’’: A neurophysiological study. Neuron, 32, 91–101. Van den Bos, E., & Jeannerod, M. (2002). Sense of body and sense of action both contribute to self-recognition. Cognition, 85, 177–187. Wood, J. N., Glynn, D. G., Phillips, B. C., & Hauser, M. D. (2007). The perception of rational, goal-directed action in non-human primates. Science, 317, 1402–1405. Wood, J. N., & Hauser, M. D. (2008). Action comprehension in non-human primates: Motor simulation or inferential reasoning? Trends in Cognitive Sciences, 12, 461–465.

Consciousness and Cognition 20 (2011) 75–81

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Neuropathologies of the self: Clinical and anatomical features q Todd E. Feinberg Albert Einstein College of Medicine, Yarmon Neurobehavior and Alzheimer’s Disease Center, Beth Israel Medical Center, First Avenue at 16th Street, New York, NY 10003, United States

a r t i c l e

i n f o

Article history: Received 2 September 2010 Available online 16 October 2010 Keywords: Asomatognosia Somatoparaphrenia Delusional misidentiﬁcation syndromes Ego functions Confabulation Neuropathologies of the self Psychological defense

a b s t r a c t The neuropathologies of the self (NPS) are disorders of the self and identity that occur in association with neuropathology and include perturbations of the bodily, relational, and narrative self. Right, especially medial-frontal and orbitofrontal lesions, are associated with these conditions. The ego disequilibrium theory proposes this brain pathology causes a disturbance of ego boundaries and functions and the emergence of developmentally immature styles of thought, ego functioning, and psychological defenses including denial, projection, splitting, and fantasy that the NPS patient has in common with the child. I hypothesize that during brain development between approximately ages 3 and 7 immature defensive functions and fantasies tend to be replaced by mature defenses and the inhibition of fantasy a process that depends upon maturational processes within the right hemisphere. I propose a four-tiered model of the NPS that emphasizes a multifactorial approach and includes both negative and positive, bottom up and top down, and neuropsychological and psychological factors. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction According to Sigmund Freud: Pathology has made us acquainted with a great number of states in which the boundary lines between the ego and the external world become uncertain or in which they are actually drawn incorrectly. There are cases in which parts of a person’s own body, even portions of his mental life – his perceptions, thoughts and feelings, appear alien to him and as not belonging to his ego; there are other cases in which he ascribes to the external world things that clearly originate in his own ego and that ought to be acknowledged by it. Thus even the feeling of our own ego is subject to disturbances and the boundaries of the ego are not constant (Freud, 1930). The neuropathologies of the self (NPS; Table 1; Feinberg, 2001, 2009a, 2010; Feinberg & Keenan, 2005) are a group of conditions in which a brain lesion causes a profound and speciﬁc alteration in the patient’s personal identity or personal relationships between the self and the world. Disorders of the bodily self impact the manner in which a person views the nature or limits of his or her physical being. Disturbances of the relational self affect the manner in which the individual interacts with objects and persons and affect the personal signiﬁcance of the self in relation to the world. Disturbances of the narrative self affect that way the individual describes personal past and present circumstances. This review provides a synopsis of a theory of these disorders. For more extended discussion of the theory, and analyses and case histories of patients with NPS, see Feinberg, 2001, 2009a,b, 2010; Feinberg, DeLuca, Giacino, Roane, & Solms, 2005. q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap. E-mail address: [email protected]

1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.017

76

T.E. Feinberg / Consciousness and Cognition 20 (2011) 75–81

Table 1 Some of the most widely reported neuropathologies of the self (NPS). The most delusional forms of these disorders are clinical end points on a multifactorial continuum that includes both negative and positive factors (see Fig. 1). A subset of these conditions, most speciﬁcally related to delusional misidentiﬁcation syndromes (DMS), have also been referred to or overlap with reduplicative paramnesia or delusions (Alexander et al., 1979; Pick, 1903; Weinstein & Burnham, 1991) delusional paramnesic misidentiﬁcation (e.g. Burgess et al., 1996), content-speciﬁc delusions (e.g. Malloy & Richardson, 1994), or monothematic delusions (e.g. Davies, Coltheart, Langdon, & Breen, 2001). The confabulations associated with NPS are most commonly of the fantastic (Bonhoeffer, 1901; Schnider, 2008) or personal (Feinberg, 1997, 2009b; Feinberg & Roane, 1997) subtypes (modiﬁed from Feinberg, 2010). Syndrome

Clinical features

Delusional anosognosia Somatoparaphrenia Delusional misidentiﬁcation syndrome (DMS) Capgras syndrome Frégoli syndrome DMS for the mirror image Phantom boarder syndrome Nurturing syndrome

Delusional denial of paralysis often accompanied by confabulation Delusional misidentiﬁcation of the arm often accompanied by confabulation Persistent delusional under or over-related related misidentiﬁcation of a person place or thing

Delusional companion syndrome

Under-related DMS Over-related DMS Under-related DMS for the mirror image The delusional belief that a person (s) is living in the residence or close by to the patient The delusional belief that a deceased loved one (typically a spouse) is alive and the patient interacts with him or her The patient adopts a soft toy such as a teddy bear and treats it like a close companion or child

NPS differ from purely cognitive conditions such as organic amnesia, language disorders, prosopagnosia, or visuospatial dysfunction, in that these conditions are neutral with reference to the individual’s relatedness to the self and the world in contrast to NPS which are speciﬁcally related and in many cases restricted to something of personal signiﬁcance to the individual (Feinberg & Roane, 1997). 1.1. Features of NPS The NPS are associated with delusions and delusional confabulations, however the confabulations in NPS differ from confabulations in general and purely amnestic confabulations in that in NPS the confabulations tend to be less ad hoc, more personally idiosyncratic, more inﬂuenced by the patient’s motivations, less tied to a speciﬁc domain of neurological impairment (multimodal) and more enduring and resistant to correction (delusional), Feinberg, 2010). The confabulations in NPS are related to or share particular features with fantastic, personal or motivated confabulations, delusional misidentiﬁcations, reduplicative paramnesia or reduplicative delusions, monothematic delusions, or content-speciﬁc delusions (see Table 1). The primary features of the confabulations in NPS that distinguish them from confabulations that are based solely upon amnesia or cognitive impairment is that the confabulations in NPS tend to be selective, multi-modal, emphasize material of personal signiﬁcance to the individual, motivated, and delusional (Feinberg, 2010). 1.2. Explaining the neuropathologies of the self: the ego disequilibrium theory There are numerous deﬁcit states – Jacksonian (1884) negative factors – that may accompany the conditions listed in the table, but the negative features associated with these conditions cannot fully explain these syndromes (see for example Goldstein, 1939). For instance, as discussed above the misbeliefs and misidentiﬁcations in DMS and NPS are generally multi-modal and therefore cannot be explained by dysfunction in a single sense modality, or a single perceptual class within one modality. In contrast to what occurs in prosopagnosia, in the Capgras syndrome the patient does not utilize extra-facial visual information such as the color or style of the hair, or the misidentiﬁed individual’s voice, to correct the error. And even though the misidentiﬁed individual repeatedly points out the error to the patient, the patient clings to the delusion. Capgras syndrome and cases of DMS in general also tend to be selective in that in the neurological cases typically only a single person or a few persons among many emotionally signiﬁcant persons in the patient’s life are misidentiﬁed. In contrast to neurobehavioral disorders such as aphasia or agnosia, the errors in the NPS are more delusional and may be wish-fulﬁlling (Feinberg, 1997; Feinberg & Roane, 1997, 2003; Feinberg et al., 2005). One hypothesis that attempts to satisfy these multiple constraints and unify the many different points of view on these conditions I call ego dysequilibrium theory (Feinberg, 2009a,b, 2010; Feinberg et al., 2005) which posits that the unique features of the adult neuropathologies of the self result from a neurologically derived alteration in the self boundaries that results in a regression to a developmentally earlier, hierarchically lower, or more primitive stage of psychological functioning that causes a recrudescence of the patterns of thought and psychological defense typical of these earlier periods. The primitive defensive patterns that are most common in the NPS include but are not limited to denial, projection, splitting, fantasy, and paranoia, functions that were dormant in the normal adult brain but are now activated by the neurological lesion. The ego disequilibrium theory accepts that there is a hierarchy of defensive functions. According to the hierarchical model of the psychological defenses proposed by Vaillant (1977, 1992, 1993), denial, delusional projection (including delusional paranoia) and distortion are the most primitive hierarchically lowest and most pathological defenses, followed by projection and fantasy that are considered immature defenses because they make their appearance later in child development. Along similar lines, Cramer (1991, 2006) proposes a hierarchy of defenses along a chronological time line in which psychological

T.E. Feinberg / Consciousness and Cognition 20 (2011) 75–81

77

defenses occur as a necessary and adaptive part of child development. Cramer ﬁnds the beginning roughly around age 3 the earliest defense to develop is psychological denial that remains the predominant defense until about age 7, followed by projection in which the individual deals with unacceptable and unwanted emotions or thoughts by attributing them to others. By about age 7 denial and projection are approximately equal and afterward identiﬁcation takes on an increasingly important role. Also during this developmental period, fantasy serves multiple adaptive and defensive functions. Cramer (1991, 2006) points out that denial through fantasy enables the child to cope with unpleasant realities and Taylor (1999) points out that the use of fantasy in the creation of imaginary companions serves a variety of adaptive functions for the child such as coping with emotional trauma, loneliness, and overcoming fears. These are the same functions that are relevant to the coping strategies and defenses of adults with neurological injury and dementia (Feinberg, 2009a, 2010). Finally, the psychological defense of splitting is another primitive defense that has relevance to the creation of imaginary companions in children and is also associated with the adult neuropathology cases (Berson, 1983). 1.3. A four-tiered model of the neuropathologies of the self The numerous negative and positive factors that could contribute to the NPS function at different levels of analysis. In Fig. 1, I propose a model of how a complex array of variables interacts to create any particular NPS. The model is roughly hierarchical with three levels of interacting factors and the ﬁnal syndrome in question on the top tier (see Feinberg (2010) for a more detailed analysis). 1.3.1. Level 1: cognitive deﬁcits On the ﬁrst and hierarchically lowest rung (level 1) are found some of the basic cognitive deﬁcits that – depending upon the particular syndrome – could play a role in the creation of the disorder. For example, in asomatognosia and somatopara-

Fig. 1. A hierarchical four-tiered model of representative factors contributing to the neuropathologies of the self. Speciﬁc cognitive deﬁcits may only be relevant to certain conditions, while self-related deﬁcits and positive features may be applied to all syndromes. Top down and bottom up factors contribute to the neuropathologies of the self. A temporal dimension emerges as syndromes evolve from the interaction of multiple lower level negative and higher level positive factors. 1Levine (1990) and Levine, Calvanio, and Rinn (1991). 2Heilman (1991) and Heilman, Barrett, and Adair (1998). 3With reference to anosognosia and asomatognosia see Geschwind (1965) and Gazzaniga (2000). 4For the neural representation of body ownership see Feinberg et al. (1990), Vallar and Ronchi (2009), andTsakiris (2009). 5Stuss (1991). 6Feinberg et al. (2005), Feinberg (2009a, 2009b), and Feinberg et al. (2010). 7Johnson (1991). 8 Ramachandran (1995). 9Coltheart (2007), Davies, Aimola Davies, and Coltheart (2005), and McKay, Langdon, and Coltheart (2005). 10With reference to Capgras syndrome see Alexander et al. (1979) and Ellis and Young (1990). 11With reference to DMS see Christodoulou (1977, 1986).

78

T.E. Feinberg / Consciousness and Cognition 20 (2011) 75–81

phrenia, sensory loss and neglect are important (Feinberg, Haber, & Leeds, 1990; Vallar & Ronchi, 2006) while in confabulation and delusional misidentiﬁcation, depending upon the particular syndrome in question, memory disorders, executive dysfunction, anatomical disconnection, visuoperceptual disorders, and spatial disorientation have been implicated as important lower level factors. 1.3.2. Level 2: self-related deﬁcits At the next hierarchical (level 2) are some of the speciﬁcally self-related deﬁcits that are important in the etiology of the NPS. Like level 1 factors these are negative deﬁcits, but these losses are speciﬁcally linked to self-related functions, such as the failure of self-monitoring that is known to occur as part of the dysexecutive syndrome after frontal damage (Stuss, 1991; Stuss & Benson, 1985; Stuss, Picton, & Alexander, 2001; Stuss, Rosenbaum, Malcolm, Christianna, & Keenan, 2005), a ‘‘reality monitoring defect” (Johnson, 1991; Johnson, Hayes, D’Esposito & Raye, 2000) an ‘‘anomaly detector” defect (Ramachandran, 1995) or a deﬁcit in a hypothetical ‘‘belief evaluation system” (Coltheart, 2005, 2007; Davies et al., 2005; McKay, Langdon, & Coltheart, 2005). We have suggested (Feinberg, 2009a,b; Feinberg & Keenan, 2005; Feinberg et al., 2005) that a critical negative feature at this level in the creation of the NPS is an alteration in the permeability of the ego boundaries that occurs as a result of bilateral or right frontal pathology. In some conditions such as the Capgras syndrome there is an under-relatedness on the part of the patient to signiﬁcant aspects of the self and personally signiﬁcant incoming information loses a feeling of relatedness (Ellis, 1998; Ellis & Young, 1990; Ellis, Young, Quayle, & de Pauw, 1997). In cases of anosognosia for hemiplegia this level of impairment represents unawareness (as opposed to delusional denial; level 4) of hemiplegia. With reference to the loss of a feeling of ownership of the arm as occurs in asomatognosia, the right temporoparietal regions are critical in establishing normal selfrelatedness to the contralateral body (Tsakiris, 2009) and damage to this region is an essential negative factor in the creation of both asomatognosia and somaotoparaphrenia. However, simple asomatognosia, the most basic loss of recognition of personal relatedness to the limb, is a level 2 impairment, in contrast to somatoparaphrenia (Feinberg, Venneri, Simone, Fan, & Northoff, 2010) a level four syndrome that is characterized by refractory delusional denial of ownership and delusional misidentiﬁcation of the limb that requires additional factors for its emergence. In contrast to the syndromes that are chieﬂy manifested by personal under-relatedness, there are other NPS, for example the Frégoli syndrome, where the patient inappropriately over-incorporates neutral aspects of the environment into the self, and thus the patient is over-related to selected aspects of the world (Feinberg, Eaton, Roane, & Giacino, 1999). The present theory covers both of these circumstances by suggesting there is an ego dysequilibrium that produces a two-way disturbance between the self and the environment speciﬁcally with regard to personal relatedness that could lead to disorders of both under and over-relatedness between the self and the world. 1.3.3. Level 3: defense, adaptation and motivation Simple unawareness of hemiplegia or the loss of relatedness to persons or parts of the body can be explained by the presence of some of the negative factors that contribute to these conditions. However, as noted above, there are also numerous positive features that are essential parts of these disorders that are best explained by the way the damaged brain adapts in a positive fashion to the various psychological deﬁcits and losses. At level 3 I propose that the emergence of primitive psychological defenses and the patient’s personal pattern of adaptation and motivation shape the next level of response. There are several reasons why brain dysfunction might activate the primitive defenses such as denial, projection, splitting, fantasy, and paranoia that I outlined above. Cramer (1991) for example points out that the mature defenses are more conscious and voluntary while the primitive defenses are in comparison largely unconscious and automatic. Thus, brain dysfunction that lowered the level of alertness or impaired attention and executive functions could also impact more voluntary and conscious defenses. Along the same lines, some authors propose that increasing cognitive skills based on brain maturation leads the progression from primitive to mature defenses (Cramer, 1991, 2006; Elkind, 1976; Laughlin, 1970; Lichtenberg & Slap, 1972; Wallerstein, 1985). When discussing level 2 deﬁcits, we discussed the disturbance in self boundaries. With reference to the NPS this represents a de-differentiation between inner and outer reality and the margins of the self, and the primitive defenses are the ones most likely to reﬂect this disturbance in ego boundaries (Cramer, 1991). The deployment of these primitive reﬂects the loss of ego boundaries and results in the tendency for the patient to externalize unwanted and disturbing aspects of reality. The defenses of psychological denial and projection both involve experiencing aspects of the self as only relevant to the external world (Breznitz, 1983). When fantasy is used as a psychological defense the needs and wishes of the individual take precedence over the constraints of external reality and in paranoia feelings of the self are transposed onto other persons. Therefore the distortion in ego boundaries that creates the perturbations of under- and over-relatedness between the self and world (a negative factor, level 2 in Fig. 1) would also promote the activation of the defenses that are characteristic of the immature ego (a positive factor, level 3). 1.3.4. The syndromes Finally, on level 4 the various negative and positive factors interact in complex and unpredictable ways to create the various NPS. For instance, during the development of somatoparaphrenia, on level 1 of the hierarchy a large right hemisphere lesion is likely, in most individuals, to create a number of generalized ‘‘cognitive deﬁcits” including proprioceptive defects,

T.E. Feinberg / Consciousness and Cognition 20 (2011) 75–81

79

hemisensory neglect, and anatomical disconnection. On the next level, a number of speciﬁcally self-related deﬁcits such as a loss of relatedness to the left side of the body, generalized impairments in self-monitoring, and a disturbance in ego boundaries focus the problem more speciﬁcally toward a disturbance that involves the self. As discussed there is an interaction between level two negative deﬁcits and level 3 positive adaptations. The disturbance in self-monitoring and the breakdown in ego boundaries and self-monitoring facilitates the emergence of primitive defenses and adaptive mechanisms (e.g. fantasy) that further shape the clinical presentation. If a sufﬁcient number of these variables are operative, and perhaps if they occur in the vulnerable individual, the full clinical syndrome of somatoparaphrenia in which the left arm is delusionally misidentiﬁed, or reduplicated, or fantasized about will emerge (Feinberg et al., 2010). The same pattern applies across a spectrum of NPS cases listed in the table. In a recent review of a representative sample of 21 reported cases of patients with imaginary others including cases of phantom border syndrome, DMS for the mirror image, nurturing syndrome and delusional companion syndrome wishful fantasy and paranoia were present in 9 cases (43%), delusional denial in 5 (24%), and splitting and projection in 3 (14%). In only 2 cases (9.5%) did the authors fail to comment on a defensive pattern (Feinberg, 2010). 2. Ego boundaries and neuroanatomy The NPS are associated with frontal pathology, and in particular right frontal pathology (Alexander, Stuss, & Benson, 1979; Burgess, Baxter, Martyn, & Alderman, 1996; Feinberg & Shapiro, 1989; Fleminger & Burns, 1993; Förstl, Burns, Jacoby, & Levy, 1991; Malloy, Cimino, & Westlake, 1992; Spangenberg, Wagner, & Bachman, 1998). Burgess et al. (1996) reviewed 41 reported cases of confabulation, paramnesias, reduplicative phenomena, and DMS. They found the highest percentage of cases had right frontal hemisphere (44%) or bilateral frontal (39%) lesions compared with only 9.7% who had left frontal lesions. Feinberg et al. (2005; see also Feinberg and Keenan (2005)) analyzed cases of DMS or delusional reduplication and found all 29 observations (100%) suffered right hemisphere damage, while only 15 (51.72%) suffered from left hemisphere damage, and in 28 out of 29 of the observations (96.6%), right frontal damage was present. In an investigation of the neuroanatomy of asomatognosia (confusion regarding ownership of the left arm) in right hemisphere stroke patients (Feinberg et al., 2010) we compared cases with simple asomatognosia (unelaborated unawareness of ownership of the limb) to cases with somatoparaphrenia (delusional denial of ownership of the limb, delusional misidentiﬁcation of the limb, confabulation). All patients with asomatognosia (simple asomatognosia and somatoparaphrenia) as well as control groups had signiﬁcant temporoparietal involvement, however the subgroup of somatoparaphrenia patients had the largest lesions overall and signiﬁcantly more frontal involvement when compared to patients with simple asomatognosia. Patients with asomatognosia had more medial frontal damage when compared to control groups but patients with somatoparaphrenia also had extensive orbitofrontal damage suggesting a further role for orbitofrontal damage in the development of somatoparaphrenia. These ﬁndings are in support of the role that the frontal (Stuss, 1991; Stuss et al., 2001, 2005) and medial frontal (Feinberg, 2009a; Feinberg et al., 2010; Northoff & Bermpohl, 2004; Northoff et al., 2006) regions play in many self-related functions. The medial and orbitomedial prefrontal regions are heteromodal association cortices and part of the integrative self system that serves to assimilate the interoself system with the external environment (see Feinberg, 2009a; also 2011). I propose that this region is intimately concerned with the integration of as well as the distinction between the self and world, and that damage to this system could contribute to a breakdown of ego boundaries and confusion between the internal representation of the self and external stimuli. The right frontal zones may be dominant for the regulation of these ego functions. 3. Conclusions: the right hemisphere and ego disequilibrium The neuropathologies of the self are a group of highly complex and multi-determined syndromes in which brain dysfunction causes a profound and speciﬁc disturbances of ego boundaries and ego functions (ego disequilibrium). This disturbance facilitates the emergence of primitive ego functions and primitive psychological defenses that are essential for full creation of the NPS. The neuropathology in the majority of these cases involves the right hemisphere with a predilection for frontal involvement especially within right medial heteromodal association cortices. This latter region is hypothesized to play a critical role in the establishment of ego boundaries and to mediate the relationship between self and world. According to this hypothesis, the preservation and activation of the largely verbal defenses, such as verbal denial, projection, splitting, and fantasy may be the result of the remaining, and presumably relatively intact left verbal hemisphere. The evidence suggests that given the fact these immature defenses and fantasies are preserved and even activated in the presence of right hemisphere damage, the emergence of the mature defenses in childhood, and their functioning in adulthood, critically depends up right hemisphere functions. The right hemisphere’s role in the mediation of the boundaries of the self (ego functions) could explain why right (especially medial frontal) damage could disrupt ego boundaries and promote the activation of the immature defenses. It might be possible that if the immature (and largely verbal defenses) are lateralized to the dominant hemisphere one could speculate that there may even be some parallel lateralization of the mature defenses to the non-dominant hemisphere, or the right hemisphere may play some additional critical role in regulating or suppressing the immature defenses.

80

T.E. Feinberg / Consciousness and Cognition 20 (2011) 75–81

References Alexander, M. P., Stuss, D. T., & Benson, D. F. (1979). Capgras syndrome: A reduplicative phenomenon. Neurology, 29, 334–339. Berson, R. J. (1983). Capgras’ syndrome. American Journal of Psychiatry, 140, 969–978. Bonhoeffer, K. (1901). Die akuten Geisteskrankheiten der Gewohnheitstrinker. Jena: Gustav Fischer. Breznitz, S. (1983). The denial of stress. New York: International Universities Press. Burgess, P. W., Baxter, D., Martyn, R., & Alderman, N. (1996). Delusional paramnesic misidentiﬁcation. In P. W. Halligan & J. C. Marshall (Eds.), Method in madness: Case studies in cognitive neuropsychiatry (pp. 51–78). East Sussex, UK: Psychology Press. Christodoulou, G. N. (1977). The syndrome of Capgras. British Journal of Psychiatry, 130, 556–564. Christodoulou, G. N. (1986). Role of depersonalization–derealization phenomena in the delusional misidentiﬁcation syndromes. Bibliotheca Psychiatrica, 164, 99–104. Coltheart, M. (2005). Delusional belief. Australian Journal of Psychology, 57, 72–76. Coltheart, M. (2007). The 33rd Bartlett Lecture: Cognitive neuropsychiatry and delusional belief. Quarterly Journal of Experimental Psychology, 60, 1041–1062. Cramer, P. (1991). The development of defense mechanisms: Theory, research, and assessment. New York: Springer-Verlag. Cramer, P. (2006). Protecting the self: Defense mechanisms in action. New York: The Guilford Press. Davies, M., Aimola Davies, A., & Coltheart, M. (2005). Anosognosia and the two-factor theory of delusion. Mind and Language, 20, 209–236. Davies, M., Coltheart, M., Langdon, R., & Breen, N. (2001). Monothematic delusions: Towards a two-factor account. Philosophy, Psychiatry & Psychology, 8, 133–158. Elkind, D. (1976). Cognitive development and psychopathology: Observations on egocentrism and ego defense. In E. Schopler & R. J. Reichler (Eds.), Psychopathology and child development: Research and treatment (pp. 167–183). New York: Plenum Press. Ellis, H. D. (1998). Cognitive neuropsychiatry and delusional misidentiﬁcation syndromes: An exemplary vindication of the new discipline. Cognitive Neuropsychiatry, 3, 81–90. Ellis, H. D., & Young, A. W. (1990). Accounting for delusional misidentiﬁcations. British Journal of Psychiatry, 157, 239–248. Ellis, H. D., Young, A. W., Quayle, A. H., & de Pauw, K. W. (1997). Reduced autonomic responses to faces in Capgras delusion. Proceedings of the Royal Society of London: Biological Sciences, B264, 1085–1092. Feinberg, T. E. (1997). Anosognosia and confabulation. In T. E. Feinberg & M. J. Farah (Eds.), Behavioral neurology and neuropsychology (1st ed., pp. 369–390). New York: McGraw-Hill. Feinberg, T. E. (2001). Altered egos: How the brain creates the self. New York: Oxford University Press. Feinberg, T. E. (2009a). From axons to identity: Neurological explorations of the nature of the self. New York: W.W. Norton. Feinberg, T. E. (2009b). Confabulation, the self, and ego functions: The ‘‘Ego Dysequilibrium Theory”. In W. Hirstein (Ed.), Confabulation: Views from neuroscience, psychiatry, psychology and philosophy (pp. 91–107). New York: Oxford. Feinberg, T. E. (2010). Neuropathologies of the self: A general theory. Neuropsychoanalysis, 12, 133–158. Feinberg, T. E. (2011). The nested neural hierarchy and the self. Conssciousness and Cognition, 20, 4–15. Feinberg, T. E., DeLuca, J., Giacino, J. T., Roane, D. M., & Solms, M. (2005). Right hemisphere pathology and the self: Delusional misidentiﬁcation and reduplication. In T. E. Feinberg & J. P. Keenan (Eds.), The lost self: Pathologies of the brain and identity (pp. 100–130). New York: Oxford. Feinberg, T. E., Eaton, L. A., Roane, D. M., & Giacino, J. T. (1999). Multiple Fregoli delusions after traumatic brain injury. Cortex, 35, 373–387. Feinberg, T. E., Haber, L. D., & Leeds, N. E. (1990). Verbal asomatognosia. Neurology, 40, 1391–1394. Feinberg, T. E., & Keenan, J. P. (2005). Where in the brain is the self? Consciousness and Cognition, 14, 661–678. Feinberg, T. E., & Roane, D. M. (1997). Anosognosia, completion and confabulation: The neutral-personal dichotomy. Neurocase, 3, 73–85. Feinberg, T. E., & Roane, D. M. (2003). Anosognosia. In T. E. Feinberg & M. J. Farah (Eds.), Behavioral neurology and neuropsychology (pp. 345–362). New York: McGraw-Hill. Feinberg, T. E., & Shapiro, R. M. (1989). Misidentiﬁcation-reduplication and the right hemisphere. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 2, 39–48. Feinberg, T. E., Venneri, A., Simone, A. M., Fan, Y., & Northoff, G. (2010). The neuroanatomy of asomatognosia and somatoparaphrenia. Journal of Neurology, Neurosurgery and Psychiatry, 81, 276–2781. Fleminger, S., & Burns, A. (1993). The delusional misidentiﬁcation syndromes in patients with and without evidence of organic cerebral disorder: A structured review of case reports. Biological Psychiatry, 33, 22–32. Förstl, H., Burns, A., Jacoby, R., & Levy, R. (1991). Neuroanatomical correlates of clinical misidentiﬁcation and misperception in senile dementia of the Alzheimer’s type. Journal of Clinical Psychiatry, 52, 268–271. Freud, S. (1930). Civilization and its discontents. Standard edition (Vol. 21, p. 66). London: Hogarth Press. Gazzaniga, M. S. (2000). Cerebral specialization and interhemispheric communication: Does the corpus callosum enable the human condition? Brain, 123, 1293–1326. Geschwind, N. (1965). Disconnexion syndromes in animals and man. Brain, 88, 585–644. Goldstein, K. (1939). The organism: A holistic approach to biology derived from pathological data in man. New York: American Book Company. Heilman, K. M. (1991). Anosognosia: Possible neuropsychological mechanisms. In G. P. Prigatano & G. L. Schacter (Eds.), Awareness of deﬁcit after brain injury: Clinical and theoretical issues (pp. 53–62). New York: Oxford University Press. Heilman, K. M., Barrett, A. M., & Adair, J. C. (1998). Possible mechanisms of anosognosia: A defect in self-awareness. Philosophical Transactions Royal Society of London B Biological Science, 353, 1903–1909. Jackson, J. H. (1884). Evolution and dissolution of the nervous system. Croonian Lectures delivered at the Royal College of Physicians. In J. Taylor (Ed.). Selected writings of John Hughlings Jackson (Vol. 2, pp. 45–75). New York: Basic Books. Johnson, M. K. (1991). Reality monitoring: Evidence from confabulation in organic brain disease patients. In G. P. Prigatano & G. L. Schacter (Eds.), Awareness of deﬁcit after brain injury: Clinical and theoretical issues (pp. 176–197). New York: Oxford University Press. Johnson, M. K., Hayes, S. M., D’Esposito, M., & Raye, C. L. (2000). Confabulation. In J. Grafman & F. Boller (Eds.), Handbook of neuropsychology (2nd ed., pp. 383–407). Amsterdam, Netherlands: Elsevier Science. Laughlin, H. P. (1970). The ego and its defenses. New York: Appleton-Century-Crofts. Lichtenberg, J. D., & Slap, J. W. (1972). On the defense mechanism: A survey and synthesis. Journal of the American Psychoanalytic Association, 20, 776–792. Levine, D. N. (1990). Unawareness of visual and sensorimotor defects: A hypothesis. Brain and Cognition, 13, 233–281. Levine, D. N., Calvanio, R., & Rinn, W. E. (1991). The pathogenesis of anosognosia for hemiplegia. Neurology, 41, 1770–1781. Malloy, P., Cimino, C., & Westlake, R. (1992). Differential diagnosis of primary and secondary delusions. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 5, 83–96. Malloy, P. F., & Richardson, E. D. (1994). The frontal lobes and content-speciﬁc delusions. The Journal of Neuropsychiatry and Clinical Neurosciences, 6, 455–466. McKay, R., Langdon, R., & Coltheart, M. (2005). ‘‘Sleights of mind”: Delusions, defences and self-deception. Cognitive Neuropsychiatry, 10, 305–326. Northoff, G., & Bermpohl, F. (2004). Cortical midline structures and the self. Trends in Cognitive Sciences, 8, 102–107. 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, 440–457. Pick, A. (1903). On reduplicative paramnesia. Brain, 26, 260–267. Ramachandran, V. S. (1995). Anosognosia in parietal lobe syndrome. Consciousness and Cognition, 4, 22–51. Schnider, A. (2008). The confabulating mind: How the brain creates reality. New York: Oxford.

T.E. Feinberg / Consciousness and Cognition 20 (2011) 75–81

81

Spangenberg, K. B., Wagner, M. T., & Bachman, D. L. (1998). Neuropsychological analysis of a case of abrupt onset mirror sign following a hypotensive crisis in a patient with vascular dementia. Neurocase, 4, 149–154. Stuss, D. T. (1991). Disturbances of self-awareness after frontal system damage. In G. P. Prigatano & G. L. Schacter (Eds.), Awareness of deﬁcit after brain injury: Clinical and theoretical issues (pp. 63–83). New York: Oxford University Press. Stuss, D., & Benson, D. F. (1985). The frontal lobes. New York: Raven Press. Stuss, D. T., Picton, T. W., & Alexander, M. P. (2001). Consciousness, self-awareness, and the frontal lobes. In S. P. Salloway, P. F. Malloy, & J. D. Duffy (Eds.), The frontal lobes and neuropsychiatric illness (pp. 101–109). Washington, DC: American Psychiatric Publishing. Stuss, D. T., Rosenbaum, R. S., Malcolm, S., Christianna, W., & Keenan, J. P. (2005). The frontal lobes and self-awareness. In T. E. Feinberg & J. P. Keenan (Eds.), The lost self: Pathologies of the brain and identity (pp. 50–64). New York: Oxford. Taylor, M. (1999). Imaginary companions and the children who create them. New York: Oxford University Press. Tsakiris, M. (2009). My body in the brain: A neurocognitive model of body-ownership. Neuropsychologia. doi:10.1016/j.neuropsychologia.2009.09.034 [Epub ahead of print]. Vaillant, G. E. (1977). Adaptation to life. Boston: Little, Brown. Vaillant, G. E. (1992). Ego mechanisms of defense: A guide for clinicians and researchers. Washington, DC: American Psychiatric Press. Vaillant, G. E. (1993). The wisdom of the ego. Cambridge, MA: Harvard University Press. Vallar, G., & Ronchi, R. (2006). Anosognosia for motor and sensory deﬁcits after unilateral brain damage: A review. Restorative Neurology and Neuroscience, 24, 247–257. Vallar, G., & Ronchi, R. (2009). Somatoparaphrenia: A body delusion. A review of the neuropsychological literature. Experimental Brain Research, 192, 533–551. Wallerstein, R. S. (1985). Defenses, defense mechanisms, and the structure of the mind. Journal of the American Psychoanalytic Association, 31S, 201–225. Weinstein, E. A., & Burnham, D. L. (1991). Reduplication and the syndrome of Capgras. Psychiatry, 54, 78–88.

Consciousness and Cognition 20 (2011) 82–93

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Embodiment, ownership and disownership q Frédérique de Vignemont CNRS – New York University, Department of Philosophy, 5 Washington Place, New York, 10003 NY, United States

a r t i c l e

i n f o

Article history: Received 2 September 2010 Available online 12 October 2010 Keywords: Embodiment Ownership Disownership Self body schema Body image Somatoparaphrenia Rubber hand Tool Alien hand

a b s t r a c t There are two main pathways to investigate the sense of body ownership, (i) through the study of the conditions of embodiment for an object to be experienced as one’s own and (ii) through the analysis of the deﬁcits in patients who experience a body part as alien. Here, I propose that E is embodied if some properties of E are processed in the same way as the properties of one’s body. However, one must distinguish among different types of embodiment, and only self-speciﬁc embodiment can lead to feelings of ownership. I address issues such as the functional role and the dynamics of embodiment, degrees and measures of ownership, and shared body representations between self and others. I then analyse the interaction between ownership and disownership. On the one hand, I show that there is no evidence that in the Rubber Hand Illusion, the rubber hand replaces the biological hand. On the other hand, I argue that the sense of disownership experienced by patients towards their body part cannot be reduced to the mere lack of ownership. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Our body may be the object we know the best, from which we constantly receive a ﬂow of information from vision, touch, proprioception, the vestibular and the interoceptive systems. Not only do we receive more information on our body than on any other objects, but we also have an internal access to it that we have to no other bodies. What makes our body so special may thus be that unlike other physical objects and other bodies, we perceive it from the inside. Only in our body, or at least in what we represent as our own body, do we feel bodily sensations. We also care for our body like we care for no other bodies, let alone other objects. Finally, it seems that it is only our own body that obeys our will with no intermediary. Yet, the relation between the body and the self is complex, giving rise to vivid philosophical debates (see Table 1). Some debates have been going on for centuries like questions about the ontological nature of the self and personal identity or about the epistemological certainty we have that this is our own body with no possible doubt. Other debates have emerged from recent scientiﬁc progress like issues about the moral and legal status of body parts in a time of research on biological materials or questions about the sensory and cognitive underpinnings of the sense of body ownership following a booming of studies in cognitive psychology and neuropsychology these last 10 years. Here, I shall set the agenda for the investigation of the sense of ownership of body parts.1 I shall focus on two main pathways, through the study of embodiment and through the study of disownership (see Table 2). However, the relation between embodiment, ownership and disownership is too often left unarticulated. In the paper, I shall bring out some of the main lines of force in this literature and lay the foundation for a proper theory of the sense of ownership. Two questions are of particular interest. First, are embodiment and ownership a matter of all-or-nothing or does it come in degrees or in various types? Second, what is the relationship between experiences of ownership and experiences of disownership? q

1

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap. E-mail address: [email protected] I shall not address the question of embodiment and ownership of the body as a whole. For such an account, see for instance Blanke and Metzinger (2009).

1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.004

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

83

Table 1 Sample of issues concerning the body and the self. Ontological questions

Am I a body or do I own a body? If I own a body, which body is mine? What grounds personal identity? Memory or the body?

Ethical and legal questions

Can one have property rights upon body parts? If so, who owns the human body? Does the status of the body vary whether it is the living body, separated tissues, human ﬂuids or the body after death? Can the body be commercialized like any object?

Psychological questions

Is there a positive phenomenology of ownership? What grounds the sense of ownership? What is the functional role of the sense of ownership? How is the sense of ownership related to (a) bodily sensations, (b) action, and (c) emotion? Can one feel ownership towards any object?

Epistemological questions

Can bodily self-ascriptions be immune to error through misidentiﬁcation relative to the subject? What guarantees bodily immunity to error?

Table 2 Artiﬁcial embodiment and pathological disembodiment. Embodiment

References

Allograft

Transplantation of a limb that belonged to another individual e.g., Dubernard et al. (2003), Farnè, Roy, Giraux, Dubernard, and Sirigu (2002) Prosthesis Appropriation of prosthetic limbs in amputees e.g., Lotze et al. (1999), Murray (2004) Virtual avatar Perception and control of a body (or a part of a body) through e.g., Cole et al. (2009) virtual reality apparatus Supernumerary limb Hallucinatory experience of additional limbs e.g., Khateb et al. (2009) Rubber Hand Illusion (RHI) Illusion of ownership of a fake hand induced by the sight of e.g., Botvinick and Cohen (1998), Tsakiris and Haggard brushing of a rubber hand at the same time as sensations of (2005) brushing of the person’s own hidden hand Full-body illusion Out-of-the body experience or body swapping induced by e.g., Ehrsson (2007), Lenggenhager et al. (2007), Petkova synchronous stroking of one’s body and a virtual body. and Ehrsson (2009) Tool Modiﬁcation of the representation of the peripersonal space e.g., Cardinali, Frassinetti et al. (2009), Cardinali, Brozzoli and the personal space after tool use (2009), Farne and Ladavas (2000), Maravita and Iriki (2004) Disembodiment Segmental exclusion syndrome Peripheral deafferentation Somatoparaphrenia Depersonalization

Body Integrity Identity Disorder (BIID)

Under-utilization of the limb after recovery following the lack of use of the limb caused by a traumatic or infectious affection Loss of tactile and proprioceptive information Denial of ownership of one’s body part General alteration of the relation to the self: anomalous bodily experiences, emotional numbing, sensation of alienation from surroundings and anomalous subjective recall Urge to be amputated of one’s own perfectly healthy limb(s)

e.g., Sacks (1984)

e.g., Cole (1995) e.g., Vallar and Ronchi (2009), Feinberg (2009) e.g., Sierra et al. (2005)

e.g., First (2005), Brang et al. (2008)

2. The sense of body ownership Before investigating the grounds of the sense of body ownership, one must distinguish between feeling of ownership and judgement of ownership (for a similar distinction within the sense of agency, see Bayne & Pacherie, 2007). Some philosophers, however, take the sense of ownership to be exclusively judgmental (Bermudez, in press). On the deﬂationary conception of ownership, there is no such thing as a feeling of body ownership, that is, a positive phenomenology of ‘myness’ that goes beyond the mere experience of bodily properties. We are aware that this body is our own in the sense that we know it, not in the sense that we feel it. This may be true for some parts of the body, like internal organs. However, as O’Shaughnessy (1980, p. 211) says: ‘‘While my gall-bladder is a part of me that I know exists, my leg is that and more”. What is the difference? I believe that my gall-bladder is a part of my body and that it belongs to me. But in the case of my leg, it is not just something that I believe. It is something that I immediately feel, though it may not be phenomenologically intense. My leg strikes me as my own in the same way that the color of the ocean strikes me as blue. My body is manifested to me in a

84

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

more primitive form that beliefs or judgments. It is manifested in the form of feelings of ownership (de Vignemont, 2007). There is something it is like to experience parts of my body as my own, some kind of non-conceptual intuitive awareness of ownership. The recent re-discovery of the Rubber Hand Illusion (hereafter RHI) has opened a new path for the study of the feeling of body ownership (Botvinick & Cohen, 1998). Brieﬂy, in the classical set-up, participants sit with their arm resting on a table, hidden behind a screen. They look at a rubber hand presented in front of them, and the experimenter simultaneously strokes both the participant’s hand and the fake hand. Participants report feeling as if they were touched on the rubber hand and as if the rubber hand were part of their body. Thanks to the RHI, the feeling of body ownership has become a scientiﬁc object of investigation. One should thus be able to determine the perceptual, motor and cognitive conditions for a body part to be experienced as one’s own. Yet, after more than 10 years of studies manipulating the RHI, are we closer to understanding the feeling of ownership? I would like to stress three difﬁculties. First, the sense of ownership is rated by questionnaires in RHI studies, but it is unclear whether participants report their feeling of ownership or their judgment of ownership. One may even challenge that they experience any ownership feelings towards the rubber hand. A second related worry concerns the scale of ownership used in questionnaires, which often goes from 3 (strongly disagree) to +3 (strongly agree). For instance, in a study with 131 subjects, ownership was rated only at +0.4 in synchronous condition and at 1.2 in asynchronous condition (Longo, Schüür, Kammers, Tsakiris, & Haggard, 2008). These results are difﬁcult to interpret. Do they reﬂect the vividness of the feeling of ownership experienced by the participants and/or the feeling of conﬁdence in their judgment of ownership? Furthermore, although considered as the experimental paradigm to study ownership, the RHI induces only a very weak ownership, so elusive that one may even question whether it induces an illusion of ownership at all. Conversely, in the asynchronous condition, one may wonder why participants do not strongly disagree that the rubber hand felt like their own hand. Finally, those results seem to indicate that the sense of ownership is a matter of degree, as if there were no strict boundary between what one experiences as one’s own and what one does not experience as one’s own. A third possible impediment to the understanding of ownership is the almost exclusive focus on the RHI. Indeed, it seems that it is only by confronting the RHI with other cases of embodiment that one could achieve a real understanding of the grounds of ownership. For instance, one needs to explain how it is possible that it takes less than two minutes to induce ownership of a rubber hand while it takes months for a hand transplant to be experienced as one’s own, or that one can experience sensations at the tip of a tool and modify the representation of the boundary of one’s body to include it, and still not feel ownership towards it. What is at stake here is to understand the relation between ownership and embodiment. Recent evidence of embodiment of external objects has added complexity to an already quite intricate picture. It is now largely accepted that the representation of one’s body can stretch to include allograft, prostheses, rubber hands, virtual avatars and tools (see Table 2). What is included can be in ﬂesh and blood, in rubber, in metal, or even virtual. It may be anatomically shaped or not. It may be controlled, or internally felt, or both. Despite their differences, the allograft, the prosthesis, the rubber hand, the virtual avatar and the tool are said to be ‘embodied’, though the notion of embodiment is too often left undeﬁned, and its relation to ownership is rarely explicitly spelled out. The relation between embodiment and ownership has been conceived in at least three different ways. First, ownership and embodiment are conceived as synonymous. But this view seems to be incompatible with the embodiment of tools with no associated ownership. Second, ownership and embodiment are conceived in opposition: ownership consists in selfattribution of the body whereas embodiment consists in self-localization (Lopez, Halje, & Blanke, 2008). But this view uses a narrow deﬁnition of the notion of embodiment, restricting it to spatial aspects. Third, ownership is conceived as a subcomponent of embodiment: the sense of embodiment consists in what it is like to have a body, and includes the sense of ownership among other experiences (Longo et al., 2008). I am more sympathetic with the latter conception, which uses a notion of embodiment wide enough to capture the peculiar relation to the body of all the various described cases, from allograft to tool. However, one should not confuse embodiment and the sense of embodiment. Embodiment corresponds to a speciﬁc type of information processing, whereas the sense of embodiment corresponds to the associated phenomenology, which includes feelings of body ownership. More particularly, I propose the following deﬁnition of embodiment: Embodiment: E is embodied if and only if some properties of E are processed in the same way as the properties of one’s body. In more philosophical terms, embodiment corresponds to a speciﬁc mode of presentation of the property of an object, which results from a speciﬁc way the property is processed. For instance, the location of the rubber hand can be presented either like the location of any object in the world (external mode of presentation) or like the location of a part of my body (embodied mode of presentation). And in the same way that I do not necessarily know that Hesperus refers to the same planet as Phosphorus, I may not be aware that the rubber hand presented under the external mode refers to the same object as the rubber hand presented under the embodied mode. Now, an object can be presented under the embodied mode, and yet not be experienced as one’s own. The notion of embodiment and the notion of the sense of ownership are at different levels of analysis. The properties of E can be processed in the same way as the properties of my body without E being experienced as my own body part. This seems to be the case for tool embodiment. But can I have a sense of ownership of E without E being embodied? I can judge that E is mine on the basis of further grounds than embodiment. But I cannot feel that E is mine without E being embodied. Embodiment is a necessary condition for the feeling of ownership. However, as we shall see, the notion of embodiment is not homogeneous. The agenda for any theory of ownership is thus twofold. First, one needs to analyse the notion of embodiment. Second, one needs to contrast embodiments that lead to ownership feelings and those that do not lead to ownership feelings.

85

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93 Table 3 Measures of embodiment. Spatial

If E is taken into account by the representation of the body space, by replacing a missing body part, by adding a body part, or by stretching an existing body part, If one is able to localize bodily sensations in E, If the location of E within the external frame is processed in the same way as the location of a part of one’s body, If the space surrounding E is processed as peripersonal space,

Motor

If one feels that E directly obeys one’s will, If one feels that a part of one’s body is moving when E is moving, If E is taken into account as an effector by the motor system in action planning,

Affective

If E is protected from hazardous situations, If one reacts in the same way when E is threatened or hurt and when a part of one’s body is threatened or hurt,

Then E is embodied

3. The measures of embodiment The deﬁnition of embodiment in terms of the types of processing that characterize the representation of one’s body does not inform us about the speciﬁc way the properties of one’s body are processed. One possible strategy to address this question is to analyse the various attempts to operationalize the notion of embodiment in clinical and experimental studies. Unfortunately, most studies on amputees merely rate the patients’ degree of satisfaction with the graft or the prosthesis or the frequency of use of the prosthesis. The only objective measure comes from brain imaging studies that show brain plasticity, especially of the primary somatosensory and motor areas, which integrate the graft/prosthesis in the neural representation of the body (Giraux, Sirigu, Schneider, & Dubernard, 2001; Lotze et al., 1999; Maruishi et al., 2004). More interesting are the studies with the RHI, with tools and with virtual reality, which have proposed several implicit measures of embodiment that can be grouped into three categories: spatial measures, motor measures and affective measures (see Table 3). 3.1. Spatial measures of embodiment The spatial measures are the most preponderant in the literature. The embodied object E can be considered within three different spatial frames of reference: the bodily frame, the external frame and the peripersonal frame. The bodily frame is the frame of the body space, as deﬁned by its boundaries and by its internal segmentation into body parts. Body geometry (i.e. the size of body parts) is especially important when acting. To reach successfully for an object, one needs to know the length of one’s arm. If E is taken into account by the representation of the body space, by replacing a missing body part, by adding a new body part, or by stretching an existing body part, then E can be said to be embodied. For instance, it was shown that prosthesis users overestimated the length of their residual limb as the result of prosthesis use (McDonnell, Scott, Dickison, Theriault, & Wood, 1989). Similarly, a recent study by Cardinali, Frassinetti and coll. (2009) showed an increase of the represented length of the arm after tool use. As Butler (1872, p. 267) noted in Erewhon more than a century ago, ‘‘Observe a man digging with a spade; his right fore-arm has become artiﬁcially lengthened, and his hand has become a joint”. Furthermore, the representation of the body space is also the representation of the space where one can feel bodily sensations. Hence, to be able to localize bodily sensations in E shows that E is taken into account by the representation of the body space and that E is embodied. And indeed this has been found with tools (e.g., Yamamoto, Moizumi, & Kitazawa, 2005), rubber hands (e.g., Botvinick & Cohen, 1998; Durgin, Evans, Dunphy, Klostermann, & Simmons, 2007), virtual hands (Ramachandran & Rogers-Ramachandran, 1996), and prostheses (Murray, 2004). As reported by an amputee, ‘‘I can actually ‘feel’ some things that come into contact with the prosthesis, without having to see them” (in Murray, 2004, p. 970). The external frame is the frame of the external world within which one navigates. If the location of E within the external frame is processed in the same way as the location of a part of one’s body, then E is embodied. For instance, after a few minutes of remote control of a robot arm through a virtual reality apparatus, one participant was afraid to drop the object held by the robot hand for fear that it might fall on his foot, as if his own foot were below the robot arm (Cole, Sacks, & Waterman, 2000). This spatial response of embodiment is reﬂected by proprioceptive drift (i.e. mislocalization of one’s touched hand or one’s body towards the rubber hand or the virtual body) in the RHI and in full-body illusions (e.g., Botvinick & Cohen, 1998; Lenggenhager, Tadi, Metzinger, & Blanke, 2007). The proprioceptive drift in the RHI, however, was found only when participants gave perceptual responses, but not when they gave motor responses (e.g. moving one’s touched hand or reaching for it with the contralateral hand, cf. Kammers, de Vignemont, Verhagen, & Dijkerman, 2009). The peripersonal frame is the frame of the space immediately surrounding a speciﬁc part of one’s body (630–50 cm). When a threatening object enters the spatial margin of safety around their body, animals engage in a range of protective behaviors (Cooke & Graziano, 2003). The representation of such a narrow sector of space is grounded on the activity of multisensory neurons whose activity decreases as the distance between visual stimuli and tactile stimuli increases. In humans, it was found that the location of visual stimuli interfered with tactile performance, but only when the visual distractor was close to a part of the body (i.e. cross-modal congruency effect, cf. Spence & Walton, 2005). Hence, if the space surrounding

86

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

E is processed as peripersonal space, then E is embodied. A number of studies have shown that the peripersonal space is increased by tool use (e.g., Cardinali, Brozzoli, & Farnè, 2009; Farnè & Làdavas, 2000; Maravita & Iriki, 2004). In addition, in RHI and full-body illusion studies, participants showed cross-modal congruency effects: visual stimuli close to the rubber hand (or the virtual body) affected tactile perception (Aspell, Lenggenhager, & Blanke, 2009; Zopf, Savage, & Williams, 2010). 3.2. Motor measures of embodiment Motor measures often constitute key measures of embodiment in virtual reality studies and in tool studies. However, they are less used to evaluate the RHI, and when they are, it is often with little success (Longo et al., 2008; Schütz-Bosbach, Mancini, Aglioti, & Haggard, 2006). Embodiment has consequences both on action awareness and action planning. At the subjective level, one feels that one’s body directly obeys one’s will. There is no need to be actually moving to experience the sense of bodily obedience. The body is experienced as directly available to carry out actions. Although related like two sides of the same coin, the sense of bodily obedience is distinct from the sense of control. Unlike the sense of control, which is primarily about movements, the sense of bodily obedience is primarily about the effectors that perform the movements. Hence, if one feels that E directly obeys one’s will, then E is embodied. A further dimension of action awareness is the experience that one’s body is moving. If one feels that a part of one’s body is moving when E is moving, then E is embodied. Both the sense of bodily obedience and the sensation of movements can be found in amputees when they see a virtual arm moving (Cole, Crowle, Austwick, & Slater, 2009; Ramachandran & Rogers-Ramachandran, 1996). This is consistent with the restoration of motor cortex activity in amputees with phantom limbs during observation of virtual hand movements (Giraux & Sirigu, 2003). Similarly, it was found that hemiplegic patients with anosognosia vividly felt having generated a movement of their paralyzed arm after seeing a prosthetic arm moving (Fotopoulou et al., 2008, p. 3438): ‘‘Sometimes I had the impression that even if I had closed my eyes I would feel it move. I mean I felt the movement in my hand. I didn’t just see it”. Finally, healthy individuals can feel that they control the movements of a virtual hand if they match their own bodily movements (Short & Ward, 2009) or their motor imagery (Perez-Marcos, Slater, & Sanchez-Vives, 2009), no matter the visual resemblance and the spatial proximity between the real hand and the virtual hand.2 At the level of motor planning, the motor system must take into account the properties of the effector such as its size, its strength, its location, its posture, etc. Hence, if the motor system takes the properties of E into account as properties of the effector in action planning, then E is embodied. For instance, Cardinali, Frassinetti and coll. (2009) showed that using a long mechanical tool to grasp objects altered the kinematics of subsequent free-hand grasping movements and of other nontrained movements like pointing. The motor responses are more contrasted in RHI studies than in tool studies. On the one hand, participants report feeling no control over the rubber hand (Longo et al., 2008). Furthermore, the location of the rubber hand is not taken into account by the motor system (Kammers, de Vignemont et al., 2009). On the other hand a further study showed that the size of the grip aperture of the rubber hand affected subsequent grasping movements (Kammers, Kootker, Hogendoorn, & Dijkerman, 2010). Finally, there is some debate whether one can induce motor facilitation effect in RHI studies. Motor imagery and action observation induce corresponding motor activity in subjects (i.e. motor facilitation effect). In line with these results, it was found an increased tendency of muscular activity when participants saw a virtual arm slowly rotating after synchronous stimulation of the biological arm and the virtual arm (Slater, Perez-Marcos, Ehrsson, & Sanchez-Vives, 2008). However, in Schütz-Bosbach and coll. (2006), the reverse effect was found: the observation of the rubber hand in movement induced a motor facilitation effect only after asynchronous stimulation. It is unclear, however, whether motor facilitation effect should be accepted as a measure of embodiment. 3.3. Affective measures of embodiment The affective measures evaluate behavioral and physiological responses when the body is put in dangerous situations that can potentially harm it. By self-preservation, one avoids such situations as much as one can (e.g., one does not use one’s hand to stoke a campﬁre or to stir a pot of boiling soup) and one reacts vividly when it happens, showing an increase of heart rhythm and of skin conductance reaction (SCR), as well as brain activity of the pain network. Hence, if E is protected from hazardous situations and one reacts to threats to or injuries of E in the same way as one reacts when a part of one’s body is threatened or hurt, then E is embodied. One can immediately see that the embodiment of tools fails to meet the affective criteria (Povinelli, Reaux, & Frey, 2010). Roughly speaking, one uses spoons to stir the pot of boiling soup. By contrast, when participants see the rubber hand being hit by a hammer or penetrated by a needle, they show a strong SCR (Ehrsson, Wiech, Weiskopf, Dolan, & Passingham, 2007). Similarly, when participants see their virtual avatar in a stressful virtual height situation, they show an affective reaction similar to the one that occurs when one experiences vertigo (Meehan, Razzaque, Insko, Whitton, & Brooks, 2005). To conclude, there is a wide variety of measures of embodiment based on what is known about the speciﬁc ways one represents the space of one’s body and one’s body in space, one experiences one’s body in relation to action and one reacts to harmful situations for one’s body. The problem, however, is that the responses are not always correlated. That does not 2 It is worth noting that in all the studies, subjects always had the intention to perform the movement they saw performed by the prosthetic or virtual hand. One may wonder what would happen if they did not intend to move. Would they still feel that they are moving?

87

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93 Table 4 Differences between rubber hand embodiment and tool embodiment.

Mechanism Spatiality Anatomical shape Affective response for harmful situations Sense of bodily obedience Sense of body ownership

Rubber hand embodiment

Tool embodiment

Passive process of visual capture of touch Incorporation Yes Yes No Yes

Active process of motor learning Extension No No Yes No

show that the object is not embodied. Those responses are indeed sufﬁcient conditions for embodiment, but not necessary conditions. But the lack of correlation questions the homogeneity of the notion of embodiment. 4. The manifold of embodiment There are many differences between the artiﬁcial embodiment of allograft, prostheses, rubber hands, virtual avatars and tools, and at various levels. To start with, some artiﬁcial embodiments occur in amputees with missing limbs (e.g., graft, prosthesis), while others occur in healthy individuals with a fully functioning body (e.g., RHI; virtual avatar). What is embodied can be the whole body (e.g., virtual avatar, full-body illusion) or merely a body part (e.g., rubber hand, prosthesis, graft). And whereas in most situations the external objects have to be bodily shaped to be embodied, though not necessarily visually resembling the subject’s biological limb, this is not true for tools. As for the dynamics, embodiment may take from no more than a couple of minutes (e.g., RHI) to a couple of months (e.g., graft). This huge variability goes along with differences at the level of the associated phenomenology. It starts from ‘‘I feel that it is an extension of my body” to ‘‘I feel that it is part of my body”, from ‘‘I feel that it is mine” to ‘‘I feel that it is me”. Although a recent study showed similar enhancement of visual detection when visual stimuli were projected on a rubber hand and on a gardening tool after training (Kao & Goodale, 2009), one of the most important contrasts may be between the embodiment of rubber hands in RHI and the embodiment of tools after tool use (see Table 4) (for more detail, see de Vignemont & Farnè, 2010). The most striking difference may be the lack of ownership feeling for tools. As Botvinick (2004, p. 783) noted, ‘‘the feeling of ownership that we have for our bodies clearly does not extend to, for example, the fork we use at dinner”. In the face of these differences, the alternative is the following. Either one claims that tools are not really embodied (Povinelli et al., 2010) or one claims that there is more than one type of embodiment. I shall adopt the latter solution and propose a taxonomy of the manifold of embodiment to capture the diversity of phenomena at stake. 4.1. A few distinctions First, we need to contrast embodiment with what I call full embodiment: E is fully embodied if and only if all its properties are processed in the same way as the properties of one’s body. I suspect that only one’s own biological limbs under normal circumstances ﬁt this deﬁnition. On the contrary, for being embodied, there is no need for all the properties of E to be taken into account by the corresponding type of bodily processing. It sufﬁces that some are. And indeed the lack of correlation between the various measures shows that an object can be embodied for some tasks, but not for others. One may believe that differences among the various types of embodiment result exclusively from differences between the mechanisms that induce embodiment. On this view, the differences between rubber hand embodiment and tool embodiment are the direct consequences of the differences between multimodal sensory integration and action. However, recent versions of the RHI involve action (e.g. participants actively stroked a toothbrush with their unseen biological hand while seeing the rubber hand performing the same movement, synchronously or not) (Kammers, Longo, Tsakiris, Dijkerman, & Haggard, 2009; Newport, Pearce, & Preston, 2010; Tsakiris, Prabhu, & Haggard, 2006). Yet, in the motor version of the RHI, the rubber hand is not taken into account to a greater extent by the motor system (Newport et al., 2010). The difference with tool embodiment has not been erased. Rather than analysing the various types of embodiment in terms of triggering mechanisms, I suggest appealing to their functional role(s). The key question then is why E is embodied. For what reasons and in what contexts should the brain process E in the same way as one’s body? In line with the Perception–Action model of vision, I suggest distinguishing between perceptual embodiment and motor embodiment. An object is perceptually embodied if it is processed in the same way as a part of one’s body for perceptual tasks. An object is motorically embodied if it is processed in the same way as a part of one’s body for motor tasks. The distinction between perceptual and motor embodiment is linked to the speciﬁc type of body representations within which E is integrated. Although controversial, it is classically assumed that there are at least two types of body representations (de Vignemont, 2010; Gallagher, 2005; Paillard, 1999). The body schema is an umbrella term that refers to sensorimotor representations of bodily properties, both when the body is the goal and when it is the means of the action. Action-oriented body representations are required at several stages in action planning and control. By contrast, the body image groups all the other representations about the body that are not used for action, whether they are perceptual, conceptual or emotional. One may then suggest that perceptual embodiment consists in representing the object within the body image, whereas motor embodiment consists in representing the object within the body schema. Perceptual embodiment and motor embodiment are always associated when it comes to one’s biological body, but not for objects like tools,

88

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

prostheses or rubber hands. For instance, the location of the rubber hand is perceptually embodied, but not motorically (Kammers, de Vignemont et al., 2009). In addition to functional distinction, one may differentiate distinct types of body representations on the basis of their respective dynamics (O’Shaughnessy, 1980). Short-term body representations consist in constantly updated on line representations of bodily properties, such as posture. Long-term body representations consist in relatively stable representations of long-term bodily properties such as the spatial organization of the body parts and their respective size. The body image is often conceived as a long-term body representation (e.g., body structural description, cf. Schwoebel & Coslett, 2005), whereas it is classically assumed that the body schema is a highly dynamic body representation, built up at time t, stored in working memory, and erased at time t + 1 by the next one. However, this view is misleading. The functional and the temporal distinctions are orthogonal. On the one hand, the body image is exploited for any perceptual task, including for instance judgments about short-term bodily properties (e.g., hand location). On the other hand, for any action, one needs to know long-term bodily properties (e.g., what effectors one has, their size, the degree of freedom of the joints). The size of one’s limbs can be temporarily altered for the sake of one action, if for instance one uses a tool. However, it is quite costly to compute the size of body parts each time one plans to move. Rather, it seems more likely that one exploits a default long-term body representation of one’s stable bodily parameters, which is indeﬁnitely reusable. We may think of it in terms of plastic bands: we can stretch them as much as we want but they always come back to their default size. Hence, in addition to motor embodiment and perceptual embodiment, one should distinguish between short-term embodiment (i.e., short-term properties of E processed in the same way as properties of one’s body) and long-term embodiment (i.e., long-term properties of E processed in the same way as properties of one’s body). This distinction might explain why the location of the rubber hand is perceptually embodied, but not its rubbery texture. The embodiment of the rubber hand would be only short-term. On the basis of this rough taxonomy,3 one may start offering an account of the sense of ownership. In particular, I shall discuss a recent hypothesis, which I shall call the Body model hypothesis, which posits integration within a long-term body representation at the core of the sense of ownership (Tsakiris, 2010; Tsakiris & Haggard, 2005). 4.2. The body model hypothesis To account for the fact that one does not feel ownership towards tools, although one does towards rubber hands, it has been suggested that tools are not ‘incorporated’, that is, are not integrated within the ‘body model’ (de Preester and Tsakiris, 2009). Tools are represented as extrinsic extensions of the body, and not as intrinsic parts of the body. According to the proponents of the Body model hypothesis, the body model is supposed to determine what can or cannot be experienced as one’s own. Although the description of the body model is often left incomplete, a minimal requirement seems to be that it represents a template of the anatomical structure of the human body. For instance, according to the Body model hypothesis, only objects that are anatomically shaped can be incorporated. I shall focus here on a different constraint, namely the two-hand constraint: one cannot have more than two hands, and thus one cannot feel ownership towards more than two hands (for discussion of further constraints, see de Vignemont & Farnè, 2010). It has the immediate consequence that one should not be able to feel ownership towards rubber hands except if one is amputated of one hand or if one does not feel ownership towards one biological hand. The proponents of the Body model hypothesis thus conclude that the RHI is not only an ownership illusion, but also a disownership illusion. On this view, one would no longer feel ownership towards the hand that has been stimulated in synchrony with the rubber hand (Tsakiris, 2010). The fate of the biological hand during the RHI has been experimentally investigated, but with controversial results. In questionnaires, participants disagree when asked if they felt as if their biological hand had disappeared, but they also disagree when asked if they felt as if they had three hands (Longo et al., 2008). At the physiological and behavioral level, it was found a decrease in skin temperature of the biological hand following the RHI, as well as a slowing down of tactile processes (Folegatti, de Vignemont, Pavani, Rossetti, & Farnè, 2009; Moseley et al., 2008). These results have been interpreted by Moseley and coll. as evidence that the biological hand is replaced by its artiﬁcial counterpart. However, a similar slowing down of tactile processes was found following prismatic displacement (Folegatti et al., 2009).4 It should thus be related to visuo-proprioceptive conﬂict rather than to disownership. Not only is there no evidence of disownership of the biological hand, but also two recent studies showed the possibility of simultaneously feeling ownership towards multiple rubber hands (Ehrsson, 2009; Newport et al., 2010). The two-hand constraint seems thus to be invalidated. I propose an alternative interpretation of the RHI on the model of supernumerary limbs. Some patients experience the presence of phantom hands or legs, in addition of their own biological hands and legs. Hence, one can feel ownership towards supernumerary limbs at no cost for the ownership of biological limbs. It can be explained in two ways. Either there are some degrees of liberty relative to the anatomical template of the human body, and at one level, one can represent one’s body with three or even four hands. Or patients with supernumerary limbs have two distinct body models, one representing the extra limbs, and the other representing the biological limbs. There is then more than one body model, each respecting the 3 The functional and the temporal distinctions do not exhaust the variety of embodiment. For instance, the location of the rubber hand is not motorically embodied, while its posture is (see Kammers et al., 2009a, 2010). Yet, both tasks seem to depend on short-term motor embodiment. 4 Like the RHI, prismatic displacement involves a cross-modal mismatch between the seen and felt position of the hand. However, unlike the RHI, there is no fake hand involved, only one’s own hand seen at a different location than where it is felt thanks to prismatic goggles.

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

89

two-hand constraint. In any case, it seems that we can be quite generous with our feelings of ownership. We are not limited to two hands, two legs, but we can extend to three or even four of each. To conclude, the results I described do not inﬁrm the Body model hypothesis, but rather invite its proponents to reﬁne their characterization of the body model and of the constraints that it lays upon ownership. In particular, they need to explain (i) the difference between tool and rubber hand if both are processed as supplementary body parts and (ii) the reason why the body model cannot integrate tools, while other types of body representations can. Indeed, it was found that after tool use, subjects localized touches delivered on the elbow and middle ﬁngertip of their arm as if they were farther apart (Cardinali, Frassinetti et al., 2009). This result at a perceptual task can be taken as evidence that tools can be integrated not only within the body schema, but also within the body image. Why not then within the body model? 4.3. Self-speciﬁc embodiment Instead of appealing to a putative body model, one may exploit a further distinction among the different types of embodiment to study the sense of ownership. Some processes do not make a difference between processing properties of one’s body or properties of other bodies, like probably visual processing of eye color. They constitute what I call neutral bodily information processing, no matter whose body it is. Other processes are self-speciﬁc, that is, they exclusively process the properties of one’s body, or they process them differently than the properties of other people’s bodies. In other words, these processes are not shared between self and others. They are the only ones that can ground the feeling of ownership. It is not sufﬁcient for an object to be processed like a body among others for the object to be experienced as one’s own. It must be processed like only one’s own body is. Self-speciﬁc processes thus indicate that E is not only a part of a human body, but also a part of one’s own body, and of no other bodies. The underlying assumption is that what is speciﬁc to the processes and the experiences of one’s own body – whether it is the speciﬁc way one perceives it, one controls it, and/or one affectively reacts to what happens to it – indicates which body one experiences as one’s own. One must thus distinguish between neutral embodiment and self-speciﬁc embodiment. Only self-speciﬁc embodiment is a necessary condition of the feeling of ownership. And only self-speciﬁc embodiment can make the difference between what is merely embodied, and what is embodied and experienced as one’s own. Interestingly, however, it is controversial whether the measures that have been qualiﬁed as implicit measures of ownership are self-speciﬁc. For instance, it is now largely accepted that affective response to pain is shared between self and others. Strong affective responses and cortical activity of the pain network have indeed been found not only when one’s body is injured and when one sees a rubber hand’s being injured in synchronous condition, but also when one knows that another individual is being injured (e.g., Singer et al., 2004). Similarly, it has been suggested that measures of the peripersonal space could constitute measures of ownership (e.g., Zopf et al., 2010), but cross-modal congruency effects were found after the mere observation of visual stimuli close to a picture of a hand with no ownership illusion involved (Igarashi, Kimura, Spence, & Ichihara, 2008). Furthermore, if measures of the peripersonal space were measures of ownership (and not merely of embodiment), it remains to be explained why one does not feel ownership towards tools though they displace the peripersonal space. One may reply that it is not the responses that are self-speciﬁc, but their strength. For instance, it is not as if either one reacted to threat or not, but to what extent one reacts, and in particular in comparison between synchronous and asynchronous conditions. On this view, the strength of the response accounts for the intensity of the ownership feeling. Although interesting, this view faces serious difﬁculties. First, it was found that the more empathetic participants were, the stronger their affective response (Singer et al., 2004). It thus seems that far from being linked to the self, the strength of the affective response is correlated with the propensity to care for others. Second, it may happen that one reacts only weakly to one’s body being injured. If body ownership was decided on the basis of the intensity of the response, then one should not feel ownership towards one’s body in this case. This conclusion, however, sounds unlikely. The difference between feeling pain in my body and feeling your pain, so to speak, is not a mere matter of degree (de Vignemont & Jacob, submitted for publication). More generally, it remains to be empirically shown that all the so-called measures of ownership are really more intense for one’s body and that the responses are correlated with the sense of ownership. In particular, this seems to be challenged by dissociations between the proprioceptive drift and the sense of ownership (Holmes, Snijders, & Spence, 2006; Kammers, Verhagen et al., 2009). But if self-speciﬁcity cannot be always reduced to a mere matter of response intensity, how to account for degrees of ownership? I suggest that the answer can be found in the literature on multimodal integration, which also comes in degrees. In particular, Welch and Warren (1980) propose that integration between various sensory inputs requires what they call the ‘assumption of unity’, that is, the assumption that the inputs carry information about the same object. Furthermore, they claim that the extent of the integration depends on the reliability of the assumption. The more reliable the assumption that the two sensory inputs are about the same object, the more uniﬁed and integrated the two percepts. Similarly, I propose that the feeling of ownership is based on the assumption of ownership (i.e. E belongs to one’s own body), and the stronger the reliability of the assumption that E belongs to one’s own body, the greater the feeling of ownership. The reliability of the assumption of ownership is a function of the number of processes of E that are self-speciﬁc relative to the weighting assigned to these processes. For instance, it is likely that the reliability of the assumption of ownership for the rubber hand is weaker than the reliability of the assumption of ownership for the biological hand stimulated in synchrony. If so, the rubber hand could not take over the biological hand.

90

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

To conclude, in order to ascertain the grounds of the feeling of ownership, one needs to determine the types of processing of bodily properties that are self-speciﬁc. This can be done only by comparing various cases of embodiment together as well as by analysing similarities and differences between the representation of one’s body and the representation of other people’s bodies. I shall now explore a different path to investigate the sense of ownership, switching from ownership of fake body parts to disownership of one’s body parts. What it is like to experience our body as alien is indeed more vivid and intense than what it is like to experience our body as our own. The study of pathological feelings of disownership can thus shed light on the phenomenal content of feelings of ownership. And arguably, what is missing in patients experiencing disownership may constitute the grounds of the feeling of ownership. Yet, very few attempts have been made to link the two phenomena (e.g. de Vignemont, 2007). It might be partly because there is almost no experimental way to induce disownership in healthy subjects, and thus to operationalize the notion of disownership (if one rules out the RHI as a real disownership illusion).5 It might be also because there is some uncertainty about the relation between the lack of ownership feelings and disownership feelings (Bermudez, in press). 5. Ownership and disownership Although the sense of body ownership may appear as a given, various pathological conditions reveal the possibility of feeling disownership towards one’s body (see Table 2). For example, patients suffering from the psychiatric disorder of depersonalization experience a general alteration of their relation to the self, such that they often feel as if their body did not belong to them or as if it had disappeared, leading them to compulsively touch their body and pour hot water on it to reassure themselves of their bodily existence (Sierra, Baker, Medford, & David, 2005). Similarly, following brain lesion or epileptic seizure, patients with somatoparaphrenia (also sometimes called asomatognosia or alien hand sign) deny ownership of one of their limbs and often attribute it to another individual (Vallar & Ronchi, 2009). However, it is with Body integrity identity disorder (BIID) that the sense of disownership leads to the most tragic consequences. Patients with BIID feel the overwhelming desire to be amputated of one(s) of their perfectly healthy limbs, partly because it feels alien (Brang, McGeoch, & Ramachandran, 2008; First, 2005). Can one provide a uniﬁed account of these experiences of disembodiment and disownership? And if so, what are the implications for a theory the sense of ownership? 5.1. The manifold of disembodiment There are important differences between the various disorders. The etiology varies, from peripheral motor and sensory deﬁcits to more central deﬁcits, from neuropsychological syndromes to psychiatric conditions. In most cases, the experience of disembodiment affects a single limb, but it can sometimes affect several body parts (e.g., depersonalisation, BIID), or almost the whole body (e.g., deafferentation). More importantly, the experience of disembodiment includes a variety of distinct feelings: (i) Feeling of unfamiliarity: Some properties of the body feel abnormal so that patients barely recognize their body. The awareness of abnormality can result from a mismatch between past and present bodily experiences. It can also result from a lack of consistency within the overall experience of the body as a whole. For instance, a patient with somatoparaphrenia compared the temperature of his two hands and said about his ‘alien’ hand: ‘‘Feel, it’s warmer than mine” (Bisiach, Rusconi, & Vallar, 1991, p. 1030). (ii) Feeling of unreality: The patients acknowledge the presence of the limb, but they do not experience it as a living body part. Rather, the limb feels like a dead body part or a fake body part made of chalk or rubber. Their ‘real’ limb, their living body part, feels absent or missing. (iii) Feeling of uselessness: The body part is left out as lazy or worthless, like a sack of coal (Feinberg, 2009). This feeling often reﬂects the actual paralysis of the limb, except in psychiatric conditions. (iv) Feeling of disownership: The body part feels alien. It does not feel like a part of their body, although the patients can acknowledge that it is contiguous with the rest of their body. They experience that the body part does not belong to them. A last difference among the patients is that in addition to feelings of disownership, some patients are delusional and make judgments of disownership. One should thus distinguish between three scenarios of disownership. 1. Patients experience their limb as alien, but they still believe that it belongs to them (e.g., depersonalisation, deafferentation). 2. Patients experience their limb as alien, and they believe that it does not belong to them, and possibly attribute it to someone else (e.g. somatoparaphrenia). 3. Patients experience their limb as their own, but despite that, they judge that it does not belong to them (e.g., possibly BIID, although it is unclear). 5 Local anaesthesia can sometimes induce disownership in healthy subjects. But in Paqueron and coll (2003), only 5 out of 36 subjects denied the ownership of their limb (Paqueron et al. 2003).

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

91

Disownership can thus occur either at the phenomenological level (feeling of disownership) or at the doxastic level (judgment of disownership) or at both. In (1) and (3), patients do not take their own bodily experiences at face value, because they are overridden by other factors. Among the disorders I described, somatoparaphrenia may be the most fascinating because the feeling of disownership is often accompanied by delusional beliefs, which are ﬁrmly sustained despite incontrovertible and obvious evidence to the contrary: ‘‘Feinberg: Suppose I told you this was your hand? Mirna: I wouldn’t believe you” (Feinberg, DeLuca, Giacino, Roane, & Solms, 2005, p. 104). The content of the delusions is of two types. On the one hand, the patients entertain disownership delusion. They believe that this body part does not belong to them. On the other hand, they entertain confabulatory delusion. Either they believe that the body part belongs to another individual or they personify it. One cannot account for those delusions solely in terms of disownership feelings. The abnormal feeling of disownership may account for the speciﬁc content of the disownership delusion (i.e. this is not my hand), but it cannot account for the feeling of conﬁdence associated with it. And it seems unlikely that it can account for the content of the confabulation either. Why does the patient attribute the hand to her niece or to a dead husband? According to the most inﬂuential theory of delusion, the two-factor model, one should distinguish between the factors that trigger an initial implausible thought (and thus contribute in explaining the thematic content of a particular delusion) and the factors that explain the uncritical adoption of an implausible thought as a delusional belief (Langdon & Coltheart, 2000). The two-factor model appeals to sensory, motor and reasoning deﬁcits to account for delusions. However, one should consider as well the role that affective factors may play in the etiology of somatoparaphrenic delusions. Indeed, patients rarely feel indifferent towards their ‘alien’ limb, and their affective attitude varies from friendliness to hate (i.e. misoplegia). Hence, any account of somatoparaphrenia must be threefold. It must explain (i) the feeling of disownership, (ii) the delusion of disownership, and (iii) the confabulatory delusion. Here I shall limit myself to investigating the feeling of disownership. In particular, do I feel that this hand does not belong to me because it feels alien and/or because it does not feel like mine? 5.2. From lack of ownership to disownership What is the relation between feelings of disownership and feelings of ownership, or the lack of them? There are three possibilities: (i) the Independent Account theory; (ii) the Uniﬁed Account theory; (iii) The Discovery theory. Let us ﬁrst consider the Independent Account theory. On this view, feelings of disownership are grounded on some features of bodily experiences indicating alienation, and those features are unrelated to the absence of feeling of ownership. If this is true, then one needs two distinct accounts, one of disownership feelings and one of ownership feelings. Cases of bodily alienation would be interesting in themselves, but they would have no relevance for the study of ownership. However, if disownership feelings and ownership feelings were fully independent, they should be able to coexist. There should be individuals who report that their body feels alien while simultaneously feeling as their own. Instead, there seems to be a systematic negative correlation between the two types of phenomenology. Alternatively, one may defend a uniﬁed account of ownership and disownership feelings. According to the Uniﬁed Account theory, disownership feelings result from the disruption of self-speciﬁc embodiment, which constitutes the necessary grounds of ownership feelings. On this view, if I do not feel ownership towards a body part that I used to feel as mine, then I must feel disownership. This prediction seems to go in line with the negative correlation between the two types of phenomenology I just highlighted. However, it may be too extreme. Arguably, there are situations where one feels nor ownership nor disownership towards a body part. There may be no distinctive attributive phenomenology, in one direction or in the other. Instead, according to the Discovery theory that I defend, disownership feelings do not directly result from the disruption of self-speciﬁc embodiment, but from the detection of the disruption (or from the awareness of the abnormal absence of feelings of ownership). As said earlier, the phenomenology of ownership is very weak and elusive. Its absence can thus remain unnoticed. If so, there is no feeling of disownership. One must become aware of the lack of ownership feelings or the disruption of self-speciﬁc embodiment to experience disownership. This theory is close to the Discovery theory of anosognosia (Levine, Calvanio, & Rinn, 1991; Ramachandran, 1995). Patients with anosognosia for hemiplegia are paralyzed, and yet, they deny being paralyzed and claim being able to move, and even being actually moving. According to Levine and coll. (1991), sensorimotor deﬁcits are not phenomenologically salient and need to be discovered by perceiving discordant information. One needs to monitor one’s performance to detect anomalies. In other words, one assumes that one’s body is healthy, unless one is provided with evidence to the contrary. The default hypothesis is that one’s body is not paralyzed. Similarly, when facing what you take for a mirror, you automatically assume that you are seeing the reﬂection of your own body. This is your default hypothesis, but if you notice discrepancies between what you do and what you see, you then realize that it is not your body that you see. Like in mirror recognition, we have a default hypothesis about the boundaries of our own body, probably encoded in longterm self-speciﬁc body representations. In other words, we do not compute the reliability of the assumption of ownership of our ‘default body’ all the time. It is only if we detect discrepancies that we question which body is our own. There may be anomalies within the processing of a part of our body (e.g., long-term embodiment of the left hand with lack of short-term embodiment of the left hand). There may be mismatch between information about one body part and information about the contralateral body part (e.g., the left hand feels different from the right hand). What matters is that we detect the presence of

92

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

conﬂictual information, which leads us to challenge and lower the reliability of the assumption of ownership of this body part. Then we may feel disownership towards the body part. 6. Conclusion This paper did no deliver the key to the sense of ownership. Rather, it did the spadework to clarify the conceptual landscape of ownership, embodiment and disownership. In particular, I highlighted the conceptual knots that need to be untied before building up any theory of ownership. I addressed issues such as the functional role and the dynamics of embodiment, degrees and measures of ownership, shared body representation between self and others, and disownership delusions. I proposed that an object is embodied if and only if some of its properties are processed in the same way as the properties of one’s body. Furthermore, I argued that the feeling of ownership is grounded in self-speciﬁc embodiment and that its intensity depends on the reliability of the ownership assumption. Finally, I argued in favor of a metacognitive theory of disownership according to which the lack of ownership does not sufﬁce to induce feelings of disownership. Acknowledgment This work was supported by an ANR Grant 07-1-191653. References Aspell, J. E., Lenggenhager, B., & Blanke, O. (2009). Keeping in touch with one’s self: Multisensory mechanisms of self-consciousness. PLoS ONE, 4(8), e6488. Bayne, T., & Pacherie, E. (2007). Narrators and comparators: The architecture of agentive self-awareness. Synthese, 159, 475–491. Bermudez, J.L. (in press). Bodily awareness and self-consciousness. In S. Gallagher (Ed.), Oxford handbook of the self. Oxford University Press. Bisiach, E., Rusconi, M. L., & Vallar, G. (1991). Remission of somatoparaphrenic delusion through vestibular stimulation. Neuropsychologia, 29(10), 1029–1031. Blanke, O., & Metzinger, T. (2009). Full-body illusions and minimal phenomenal selfhood. Trends in Cognitive Sciences, 13(1), 7–13. Botvinick, M. (2004). Neuroscience probing the neural basis of body ownership. Science, 305, 782–783. Botvinick, M., & Cohen, J. (1998). Rubber hands ‘feel’ touch that eyes see. Nature, 391, 756. Brang, D., McGeoch, P. D., & Ramachandran, V. S. (2008). Apotemnophilia: A neurological disorder. NeuroReport, 19(13), 1305–1306. Butler, S. (1872). Erewhon. London: Penguin. Cardinali, L., Brozzoli, C., & Farnè, A. (2009b). Peripersonal space and body schema: Two labels for the same concept? Brain Topography, 21(3–4), 252–260. Cardinali, L., Frassinetti, F., Brozzoli, C., Urquizar, C., Roy, A. C., & Farnè, A. (2009a). Tool-use induces morphological updating of the body schema. Current Biology, 19(12), R478–R479. Cole, J. (1995). Pride and a daily marathon. Cambridge, MA: MIT Press. Cole, J., Crowle, S., Austwick, G., & Slater, D. H. (2009). Exploratory ﬁndings with virtual reality for phantom limb pain; from stump motion to agency and analgesia. Disability and Rehabilitation, 31(10), 846–854. Cole, J., Sacks, O., & Waterman, I. (2000). On the immunity principle: A view from a robot. Trends in Cognitive Sciences, 4, 167. Cooke, D. F., & Graziano, M. S. A. (2003). Defensive movements evoked by air puff in monkeys. Journal of Neurophysiology, 90, 3317–3329. de Preester, H., & Tsakiris, M. (2009). Body-extension versus body incorporation: Is there a need for a body-model? Phenomenology and the Cognitive Sciences, 8, 307–319. de Vignemont, F. (2007). Habeas Corpus: The sense of ownership of one’s own body. Mind and Language, 22(4), 427–449. de Vignemont, F. (2010). Body schema and body image-pros and cons. Neuropsychologia, 48, 669–680. de Vignemont, F. and Jacob, P. (submitted for publication). What it is like to feel another’s pain. Mind. de Vignemont, F., & Farnè, A. (2010). Widening the body to rubber hands and tools: What’s the difference? Revue de Neuropsychologie, Neurosciences Cognitives et Cliniques, 2(3), 1–9. Dubernard, J. M., Petruzzo, P., Lanzetta, M., Parmentier, H., Martin, X., Dawahra, M., et al (2003). Functional results of the ﬁrst human double-hand transplantation. Annals of Surgery, 238(1), 128–136. Durgin, F. H., Evans, L., Dunphy, N., Klostermann, S., & Simmons, K. (2007). Rubber hands feel the touch of light. Psychological Science, 18(2), 152–157. Ehrsson, H. H. (2007). The experimental induction of out-of-body experiences. Science, 317(5841), 1048. Ehrsson, H. H. (2009). How many arms make a pair? Perceptual illusion of having an additional limb. Perception, 38, 310–312. Ehrsson, H. H., Wiech, K., Weiskopf, N., Dolan, R. J., & Passingham, R. E. (2007). Threatening a rubber hand that you feel is yours elicits a cortical anxiety response. Proceedings of the National Academy of Sciences of the United States of America, 104(23), 9828–9833. Farnè, A., & Làdavas, E. (2000). Dynamic size-change of hand peripersonal space following tool use. NeuroReport, 11, 1645–1649. Farnè, A., Roy, A. C., Giraux, P., Dubernard, J. M., & Sirigu, A. (2002). Face or hand, not both: Perceptual correlates of reafferentation in a former amputee. Current Biology, 612(15), 1342–1346. Feinberg, T. E. (2009). From axons to identity: Neurological explorations of the nature of the self. New York: WW Norton. Feinberg, T. E., DeLuca, J., Giacino, J. T., Roane, D. M., & Solms, M. (2005). Right hemisphere pathology and the self: Delusional misidentiﬁcation and reduplication. In T. E. Feinberg & J. P. Keenan (Eds.), The lost self: Pathologies of the brain and identity. New York: Oxford University Press. First, M. B. (2005). Desire for amputation of a limb: paraphilia, psychosis, or a new type of identity disorder. Psychological Medicine, 35(6), 919–928. Folegatti, A., de Vignemont, F., Pavani, F., Rossetti, Y., & Farnè, A. (2009). Losing one’s hand: Visual-proprioceptive conﬂict affects touch perception. PLoS ONE, 4(9), e6920. Fotopoulou, A., Tsakiris, M., Haggard, P., Vagopoulou, A., Rudd, A., & Kopelman, M. (2008). The role of motor intention in motor awareness: An experimental study on anosognosia for hemiplegia. Brain, 131(Pt 12), 3432–3442. Gallagher, S. (2005). How the body shapes the mind. New York: Oxford University Press. Giraux, P., & Sirigu, A. (2003). Illusory movements of the paralyzed limb restore motor cortex activity. Neuroimage, 20(Suppl. 1), S107–S111. Giraux, P., Sirigu, A., Schneider, F., & Dubernard, J. M. (2001). Cortical reorganization in motor cortex after graft of both hands. Nature Neuroscience, 4(7), 691–692. Holmes, N. P., Snijders, H. J., & Spence, C. (2006). Reaching with alien limbs: Visual exposure to prosthetic hands in a mirror biases proprioception without accompanying illusions of ownership. Perception & Psychophysics, 68(4), 685–701. Igarashi, Y., Kimura, Y., Spence, C., & Ichihara, S. (2008). The selective effect of the image of a hand on visuotactile interactions as assessed by performance on the crossmodal congruency task. Experimental Brain Research, 184(1), 31–38. Kammers, M. P., de Vignemont, F., Verhagen, L., & Dijkerman, H. C. (2009a). The rubber hand illusion in action. Neuropsychologia, 47, 204–211.

F. de Vignemont / Consciousness and Cognition 20 (2011) 82–93

93

Kammers, M. P., Kootker, J. A., Hogendoorn, H., & Dijkerman, H. C. (2010). How many motoric body representations can we grasp? Experimental Brain Research, 202(1), 203–212. Kammers, M. P., Longo, M. R., Tsakiris, M., Dijkerman, H. C., & Haggard, P. (2009b). Speciﬁcity and coherence of body representations. Perception, 38, 1804–1820. Kammers, M. P., Verhagen, L., Dijkerman, H. C., Hogendoorn, H., de Vignemont, F., & Schutter, D. J. (2009c). Is this hand for real? Attenuation of the rubber hand illusion by transcranial magnetic stimulation over the inferior parietal lobule. Journal of Cognitive Neuroscience, 21(7), 1311–1320. Kao, K. L., & Goodale, M. A. (2009). Enhanced detection of visual targets on the hand and familiar tools. Neuropsychologia, 47, 2454–2463. 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. Langdon, R., & Coltheart, M. (2000). The cognitive neuropsychology of delusions. Mind and Language, 15, 183–216. Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317(5841), 109.6–109.9. Levine, D. N., Calvanio, R., & Rinn, W. E. (1991). The pathogenesis of anosognosia for hemiplegia. Neurology, 41(11), 1770–1781. Longo, M. R., Schüür, F., Kammers, M. P., Tsakiris, M., & Haggard, P. (2008). What is embodiment? A psychometric approach. Cognition, 107, 978–998. Lopez, C., Halje, P., & Blanke, O. (2008). Body ownership and embodiment: Vestibular and multisensory mechanisms. Neurophysiologie Clinique, 38(3), 149–161. Lotze, M., Grodd, W., Birbaumer, N., Erb, M., Huse, E., & Flor, H. (1999). Does use of a myoelectric prosthesis prevent cortical reorganization and phantom limb pain? Nature Neuroscience, 2(6), 501–502. Maravita, A., & Iriki, A. (2004). Tools for the body (schema). Trends in Cognitive Science, 8, 79–85. Maruishi, M., Tanaka, Y., Muranaka, H., Tsuji, T., Ozawa, Y., et al (2004). Brain activation during manipulation of the myoelectric prosthetic hand: A functional magnetic resonance imaging study. Neuroimage, 21(4), 1604–1611. McDonnell, P. M., Scott, R. N., Dickison, J., Theriault, R. A., & Wood, B. (1989). Do artiﬁcial limbs become part of the user? New evidence. Journal of Rehabilitation Research and Development, 26(2), 17–24. Meehan, M., Razzaque, S., Insko, B., Whitton, M., & Brooks, F. P. Jr., (2005). Review of four studies on the use of physiological reaction as a measure of presence in stressful virtual environments. Applied Psychophysiology Biofeedback, 30(3), 239–258. Moseley, G. L., Olthof, N., Venema, A., Don, S., Wijers, M., Gallace, A., et al (2008). Psychologically induced cooling of a speciﬁc body part caused by the illusory ownership of an artiﬁcial counterpart. PNAS, 105, 13169–13173. Murray, C. D. (2004). An interpretative phenomenological analysis of the embodiment of artiﬁcial limbs. Disability and Rehabilitation, 26, 963–973. Newport, R., Pearce, R., & Preston, C. (2010). Fake hands in action: Embodiment and control of supernumerary limbs. Experimental Brain Research, 204, 385–395. O’Shaughnessy, B. (1980). The will: Dual aspect theory. Cambridge: Cambridge University Press. Paillard, J. (1999). Body schema and body image—A double dissociation in deafferented patients. In G. N. Gantchev, S. Mori, & J. Massion (Eds.), Motor control. Today and tomorrow. Sophia: Academic Publishing House. Paqueron, X., Leguen, M., Rosenthal, D., Coriat, P., Willer, J. C., & Danziger, N. (2003). The phenomenology of body image distortions induced by regional anesthesia. Brain, 126, 702–712. Perez-Marcos, D., Slater, M., & Sanchez-Vives, M. V. (2009). Inducing a virtual hand ownership illusion through a brain–computer interface. NeuroReport, 20, 22(6), 589–594. Petkova, V. I., & Ehrsson, H. H. (2009). When right feels left: referral of touch and ownership between the hands. PLoS ONE, 4, e6933. Povinelli, D. J., Reaux, J. E., & Frey, S. H. (2010). Chimpanzees’ context-dependent tool use provides evidence for separable representations of hand and tool even during active use within peripersonal space. Neuropsychologia, 48, 243–247. Ramachandran, V. S. (1995). Anosognosia in parietal lobe syndrome. Consciousness and Cognition, 4, 22–51. Ramachandran, V. S., & Rogers-Ramachandran, D. (1996). Synaesthesia in phantom limbs induced with mirrors. Proceedings of the Royal Society of London, 263, 377–386. Sacks, O. (1984). A leg to stand on. London: Picador. Schütz-Bosbach, S., Mancini, B., Aglioti, S. M., & Haggard, P. (2006). Self and other in the human motor system. Current Biology, 16(18), 1830–1834. Schwoebel, J., & Coslett, H. B. (2005). Evidence for multiple, distinct representations of the human body. Journal of Cognitive Neuroscience, 17, 543–553. Short, F., & Ward, R. (2009). Virtual limbs and body space: Critical features for the distinction between body space and near-body space. Journal of Experimental Psychology: Human Perception and Performance, 35(4), 102–103. Sierra, M., Baker, D., Medford, N., & David, A. S. (2005). Unpacking the depersonalization syndrome: An exploratory factor analysis on the Cambridge Depersonalization Scale. Psychological Medicine, 35, 1–10. Singer, T., Seymour, B., O ’ Doherty, J., Kaube, H., Dolan, R. J., & Frith, C. D. (2004). Empathy for pain involves the affective but not sensory components of pain. Science, 303(5661), 1157–1162. Slater, M., Perez-Marcos, D., Ehrsson, H. H., & Sanchez-Vives, M. V. (2008). Towards a digital body: The virtual arm illusion. Frontiers in Human Neuroscience, 2, 6. Spence, C., & Walton, M. (2005). On the inability to ignore touch when responding to vision in the crossmodal congruency task. Acta Psychologica (Amst), 118(1–2), 47–70. Tsakiris, M. (2010). My body in the brain: A neurocognitive model of body-ownership. Neuropsychologia, 48, 703–712. Tsakiris, M., & Haggard, P. (2005). The rubber hand illusion revisited: Visuotactile integration and self-attribution. Journal of Experimental Psychology: Human Perception and Performance, 31, 80–91. Tsakiris, M., Prabhu, G., & Haggard, P. (2006). Having a body versus moving your body: How agency structures body-ownership. Consciousness and Cognition, 15, 423–432. Vallar, G., & Ronchi, R. (2009). Somatoparaphrenia: A body delusion. A review of the neuropsychological literature. Experimental Brain Research, 192(3), 533–551. Welch, R. B., & Warren, D. H. (1980). Immediate perceptual response to intersensory discrepancy. Psychological Bulletin, 88, 638–667. Yamamoto, S., Moizumi, S., & Kitazawa, S. (2005). Referral of tactile sensation to the tips of L-shaped sticks. Journal of Neurophysiology, 93, 2856–2863. Zopf, R., Savage, G., & Williams, M. A. (2010). Crossmodal congruency measures of lateral distance effects on the rubber hand illusion. Neuropsychologia, 48(3), 713–725.

Consciousness and Cognition 20 (2011) 94–98

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Brain connectivity and the self: The case of cerebral disconnection q Lucina Q. Uddin ⇑ Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA 94304, USA Asian University for Women, Chittagong, Bangladesh

a r t i c l e

i n f o

Article history: Available online 27 September 2010 Keywords: Split-brain Commissurotomy Corpus callosum Self-recognition Self-awareness Brain network Brain connectivity Agency Alien hand Anarchic hand

a b s t r a c t Over the past several years, the study of self-related cognition has garnered increasing interest amongst psychologists and cognitive neuroscientists. Concomitantly, lesion and neuroimaging studies have demonstrated the importance of intact cortico-cortical and cortico-subcortical connections for supporting high-level cognitive functions. Commissurotomy or ‘‘split-brain” patients provide unique insights into the role of the cerebral commissures in maintaining an individual’s sense of self, as well as into the unique selfrepresentation capabilities of each cerebral hemisphere. Here we review empirical work examining the integrity of self-related processes in patients with various disconnection syndromes, focusing on studies of self-recognition, ownership, and agency. Taken together, this body of work suggests that an intact corpus callosum enabling interhemispheric transfer is necessary for some, but not all types of self-representations. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction The self is a popular topic in psychology and neuroscience. However, the term ‘‘self” is often used to discuss multiple different cognitive phenomena, and can thus be difﬁcult to deﬁne (Klein & Gangi, 2010). While the ﬁeld has yet to agree upon a precise deﬁnition of the self, we can perhaps all concede that the self is a function of brain states and processes. One useful distinction is that between physical aspects of the self and psychological aspects of the self (Gillihan & Farah, 2005). Physical aspects of the self are often examined in studies of self-face recognition, agency, and perspective taking, whereas psychological aspects of the self are often operationalized with studies examining autobiographical memory or other types of selfknowledge (Uddin, in press). This conceptual distinction bears out in neuroimaging work, which has demonstrated that physical or embodied self-related processes and psychological or evaluative self-related processes rely on distinct large-scale brain networks (Lieberman, 2007; Uddin, Iacoboni, Lange, & Keenan, 2007). Researchers have been trying in recent years to pinpoint exactly which brain states and processes underlie which forms of self-representation, and how these different representations interact, with varying degrees of success. Like other higher cognitive functions, self-representation in the brain likely relies on the integrity of and communication between large-scale brain networks (Bressler & Menon, 2010). Commissurotomy, or surgical sectioning of the major cerebral commissures, was at one point the most effective treatment for minimizing the spread of epileptic seizures (Bogen & Vogel, 1976; Gazzaniga, Bogen, & Sperry, 1962) (Fig. 1). Commissurotomy and partial callosotomy procedures have been conducted in several cases of intractable epilepsy, though pharmacological treatments have been more widely adopted in recent years. Careful behavioral testing of the patients who have undergone

q

This article is part of a special issue of this journal on Brain and self: Bridging the gap.

⇑ Address: 780 Welch Road, Suite 201, Stanford, CA 94304, USA. Fax: +1 650 736 7200. E-mail address: [email protected] 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.009

L.Q. Uddin / Consciousness and Cognition 20 (2011) 94–98

95

Fig. 1. Structural MRI scan of a commissurotomy patient showing absence of the major cerebral commissures.

‘‘split-brain” surgery has shed light on the specialized roles of the two cerebral hemispheres, and in particular have allowed for testing theories of positive hemispheric competence (Zaidel & Clarke, 1990). Split-brain patients continue to present a unique opportunity for testing the functions of each hemisphere independent from the other, and for helping us to understand the role the corpus callosum plays in cognition (see (Gazzaniga, 2005; Zaidel, Iacoboni, Zaidel, & Bogen, 2003 for reviews). Callosal abnormalities can also result from congenital conditions, such as agenesis of the corpus callosum, or from acquired brain damage, tumors, or stroke. In this review we will summarize what is known with regard to self-representation and self-related cognition in split-brain patients and other patients with brain lesions affecting the cerebral commissures. We will focus primarily on studies of visual self-face recognition and agency, or the feeling of ownership of actions. One of the most widely used methods for assessing self-awareness is the test of self-recognition. Preyer was one of the ﬁrst to use mirrors to assess the development of self-concept, noting that ‘‘. . .the behavior of the child toward his image in the glass shows unmistakably the gradual growth of consciousness of self out of a condition in which objective and subjective changes are not yet distinguished from each other” (Preyer, 1889). Later, Amsterdam demonstrated that infants around 2 years of age begin to show evidence of self-recognition in front of a mirror (Amsterdam, 1972), which is the same period during which they begin to use personal pronouns. The ability to mirror-self-recognize has only been reliably demonstrated in higher primates, leading Gallup, who conducted much of the pioneering work, to suggest that the capacity is indicative of an underlying self-concept (Gallup, 1977). Self-recognition studies using functional magnetic resonance imaging have revealed that in the intact adult brain, a fronto-parietal network in the right hemisphere is involved in recognizing the selfface (Platek et al., 2006; Sugiura et al., 2005; Uddin, Kaplan, Molnar-Szakacs, Zaidel, & Iacoboni, 2005a). Another form of self-representation is captured in the concept of the sense of agency, deﬁned as the feeling of being the one generating an action (Farrer & Frith, 2002). This is related to the concept of ownership, or the feeling that a limb belongs to one’s body. Some split-brain patients present with an unusual condition called ‘‘alien” or ‘‘anarchic hand” syndrome, so named because the patient will display uncontrolled behavior by one extremity, intermanual conﬂict, or denial of limb ownership (Aboitiz et al., 2003). In this condition the sense of agency and sometimes ownership in the patient is disrupted, and the limb is perceived to be moving involuntarily or to not belong to the patient. We will begin by reviewing studies of self-face recognition in split-brain patients, then continue with a summary of studies of agency and ownership conducted in these and other patients with callosal disruptions. We will conclude with a discussion of what the reviewed studies reveal with regards to these types of self-representation and how they are implemented in the brain. 2. The self in ‘‘split-brain patients: self-face recognition The earliest study to examine responses to self-face stimuli in the split-brain was conducted by Preilowski, who used a scleral contact lens system to allow for prolonged exposure and ocular scanning of complex visual stimuli by one hemisphere at a time (Zaidel, 1975). He found in two split-brain patients that both showed the largest skin resistance response (an index of arousal) when their own faces were presented to the right hemisphere. Amazingly, the skin resistance changes were observed even when the stimuli could not be verbally identiﬁed. This led the author to ‘‘assume the existence of conscious awareness in the ‘mute, minor’ as well as the ‘speaking, major’ cortical half” (Preilowski, 1977). Sperry and colleagues, who have for many years espoused the view that the disconnected hemispheres are separately conscious in parallel at a moderately high and approximately equal level (Sperry, 1968) also tested self-recognition in the split-brain. They tested two commissurotomy patients’ reactions to key personal and affect-laden stimuli such as pictures of the self, relatives, pets, and belongings, as well as pictures of public, historical, and religious ﬁgures and people in the entertainment industry. The results of this work indicated that the concept of self and general social awareness were present and developed in both the right and left (language-dominant) hemispheres of these patients; responses to stimuli presented to each independent hemisphere were essentially comparable (Sperry, Zaidel, & Zaidel, 1979). Of particular interest, when

96

L.Q. Uddin / Consciousness and Cognition 20 (2011) 94–98

the subjects were asked to identify a photo of themselves inserted among similar photos in an array of items, they had no difﬁculty doing so with either hemisphere. This recognition was also accompanied by appropriate emotional reactions. The authors concluded that the combination of correct identiﬁcations and appropriate cognitive associations performed by the independent hemispheres provided evidence that human subjectivity and sense of self was similar in the two hemispheres. Several years later, self-recognition in the split-brain was again assessed, this time with a different paradigm. Morphed face images (between the patient and a familiar colleague, Fig. 2) were presented in a lateralized fashion to each hemisphere, and the patient was asked to indicate by button press whether the image presented was himself. In another condition, the patient was asked to judge whether the image presented was the familiar other. The study found that the right and left hemispheres of this particular split-brain patient showed opposite biases; while the right hemisphere showed a bias for recognizing the morphed face as the familiar other, the left hemisphere was biased in favor of recognizing the self. The authors concluded that while both hemispheres were capable of self-recognition, the left hemisphere had an enhanced ability to recognize the self (Turk et al., 2002). However, another interpretation of this ﬁnding is that the left hemisphere has a more vague self-representation, and thus is biased to recognize an image as self no matter how low a percentage of the self-face is actually present in the image. The right hemisphere, on the other hand, may have stricter criteria for including an image in the category of self. Subsequent studies used alternative strategies to test behavioral hemispheric competence. Keenan and colleagues also tested a callosotomy patient using morphed self-face stimuli. In this study, it was presumed that responses with the left hand reﬂected right hemisphere processing, and responses with the right hand reﬂected left hemisphere processing. They report that the patient made both greater true-positive and fewer false-positive responses when responding with the left hand to the question ‘‘does this image contain portions of your own face?” (Keenan, Wheeler, Platek, Lardi, & Lassonde, 2003). These ﬁndings suggest that the right hemisphere is more sensitive to self-recognition. However, it is not universally accepted that response hand is a reliable indicator of hemispheric performance (Weems & Zaidel, 2005). Another study published shortly after used the morphed face stimuli paradigm with lateralized presentation to test selfrecognition in a complete commissurotomy patient. This study attempted to redress the methodological concerns of the previous studies in three ways: (1) by distinguishing between response bias and sensitivity, (2) by using lateralized stimuli to ensure input and processing by one hemisphere at a time, and (3) by analyzing only data from ‘‘pure hemispheric” responses, or responses where the response hand was congruent to the visual ﬁeld of the stimulus. Using signal detection methods to detect sensitivity to detect a target (the self-face) independent of bias, the authors found, similar to Turk et al., that the left hemisphere had a ‘‘self” bias. However, both hemispheres were signiﬁcantly above chance in their sensitivity to detection of self-faces, whereas only the right hemisphere was sensitive to detection of a familiar other. The authors concluded that the representation of the self that allows for self-face recognition is thus not restricted to a particular hemisphere, but is available to each cerebral hemisphere independently (Uddin, Rayman, & Zaidel, 2005b). One complication when comparing across different studies of split-brain patients is that the patients often have undergone varying degrees of surgical disconnection. Turk et al. tested patient JW at age 48, who underwent a two-stage callosal surgery with sparing of the anterior commissure at the age of 25 (Turk et al., 2002). Keenan and colleagues tested ML, a 26year-old man who also underwent callosotomy, at the age of 22 (Keenan et al., 2003). Uddin and colleagues, on the other hand, tested a complete commissurotomy patient, NG, at age 70. She had undergone single stage midline section of the anterior commissure, corpus callosum, hippocampal commissure, and massa intermedia at age 29 (Uddin et al., 2005b). In addition to heterogeneity with respect to surgical histories, these patients had differing epileptic symptom severities pre- and post-operation. To complicate matters further, the experimental designs, behavioral response collection methods, and statistical analyses were not uniform across these three studies, making them difﬁcult to compare to one another. However, all report that both cerebral hemispheres of the patients were able to complete the self-identiﬁcation task, suggesting that this type of visual self-representation is available independently to each hemisphere. 3. The self in ‘‘split-brain patients: ownership and agency Some patients with lesions affecting the corpus callosum present with a curious behavior that has come to be termed ‘‘alien hand syndrome” (AHS), translated from the French ‘le signe de la main etrangere’ (Brion & Jedynak, 1972). First noted

Fig. 2. Example of stimuli used in self-recognition studies. Studies of self-face recognition have used morphed face stimuli, wherein a photograph of an individual is combined to varying degrees with a photograph of another individual (adapted from Uddin et al., 2005b).

L.Q. Uddin / Consciousness and Cognition 20 (2011) 94–98

97

in a discussion of apraxias by Goldstein (1908), the term was used to describe patients who denied ownership of one of their hands, or whose one hand involuntarily performed a movement, often in opposition to the other hand (Akelaitis, 1945). In 1992, Feinberg proposed a distinction between a frontal AHS, associated with reﬂexive grasping and compulsive manipulation of tools, and a callosal AHS characterized mainly by intermanual conﬂict. He reviewed evidence that frontal AHS results from damage to supplementary motor area, anterior cingulate gyrus, and medial prefrontal cortex, whereas callosal AHS results from anterior corpus callosal lesions (Feinberg et al., 1992). Aboitiz and colleagues further reﬁned the subtypes of AHS into four categories: (1) diagnostic dyspraxia and intermanual conﬂict, in which one hand performs actions contrary or opposite to the actions executed by the other hand, (2) alien hand sign, in which the patient reports a subjective feeling that the hand does not belong to him/her, (3) anarchic or wayward hand, in which the affected hand performs goal-directed movements that the patient does not perceive as controlled by his/her own will, and (4) supernumerary hands, in which the patient reports the subjective feeling of having an extra extremity (Aboitiz et al., 2003). Of these syndromes, a subset can result from callosal damage only while the others result from cortical damage in addition to callosal damage. Intermanual conﬂict is typically seen with posterior or mid-body corpus callosum lesions, or complete commissurotomy patients in the early post-operative period, and rarely persists in the long term. Alien hand sign can be caused by posterior callosal lesions, but typically also involves damage to parietal cortical areas. Anarchic hand typically results from frontal lobe or anterior corpus callosum lesions, and is distinct from alien hand in that the patient does not lose the feeling that the unwilled hand is his/her own (Aboitiz et al., 2003). There has been one reported case of hemispatial neglect as a result of subacute infarction of the entire corpus callosum, though this is rare (Muangpaisan, Srisajjakul, & Chiewvit, 2005). The cases of AHS resulting from corpus callosum lesions illustrate the role of the corpus callosum in maintaining an integrated sense of self with regards to body awareness and planning of actions. It appears that communication between the two hemispheres enabled by the anterior corpus callosum is necessary for awareness of initiation of goal-directed movements. Posterior corpus callosum integrity seems necessary for maintaining a sense of limb ownership, as this region interconnects parietal areas involved in self-body representation. Thus, unlike with self-face recognition, interhemispheric communication and coordination appears to be necessary for feeling ownership of ones own actions and body parts. 4. Summary and conclusions Studies of individuals with complete or partial corpus callosum lesions have revealed a complex story of representations of the self in the brain. Self-face recognition, on the one hand, appears to be available to both hemispheres in isolation, and does not rely on interhemispheric communication. Subjective feelings of limb ownership and agency, on the other hand, appear to require the posterior and anterior corpus callosum, respectively, in addition to intact fronto-parietal cortical functioning. It is worth noting that in his ﬁrst studies of individuals who had undergone a split-brain procedure, Bogen observed that his patients ‘‘show a remarkable absence of functional impairment in nearly all ordinary behavior” (Bogen, Fisher, & Vogel, 1965). But what accounts for this remarkable adaptation? It has been assumed that some information can be transferred between the hemispheres through subcortical pathways when cortical commissures are no longer available (Funnell, Corballis, & Gazzaniga, 2000). It has recently been shown in a complete commissurotomy patient that residual functional connectivity can be detected between the disconnected hemispheres by examination of low-frequency BOLD MRI signal. The authors suggested that this evidence of continued interhemispheric interaction in the absence of cortical commissures is evidence for subcortical coordination of cortical networks (Uddin et al., 2008). Directions for future work include examination of other types of self-related cognition in the disconnected hemispheres. While physical aspects of the self, including self-face recognition and agency have been the focus of quite a few studies, less attention has been directed at investigating psychological aspects of the self in split-brain patients and individuals with lesions affecting the corpus callosum. Studies of autobiographical memory and self-knowledge would in particular shed light on the role of the corpus callosum in the maintenance of the psychological self. Split-brain patients continue to challenge the notion of a uniﬁed self, as envisioned by Sherrington (1941), and suggest instead that different aspects of the multifaceted entity we call the self are subserved by distributed neural systems. Acknowledgments This work was supported by Award Number K01MH092288 from the National Institute of Mental Health. The content is solely the responsibility of the author and does not necessarily represent the ofﬁcial views of the NIMH or the NIH. References Aboitiz, F., Carrasco, X., Schroter, C., Zaidel, D., Zaidel, E., & Lavados, M. (2003). The alien hand syndrome: Classiﬁcation of forms reported and discussion of a new condition. Neurological Sciences, 24(4), 252–257. Akelaitis, A. I. (1945). Studies on the corpus callosum. IV. Diagnostic dyspraxia in epileptics following partial and complete section of the corpus callosum. American Journal of Psychiatry, 101, 594–599. Amsterdam, B. (1972). Mirror self-image reactions before age two. Developmental Psychobiology, 5(4), 297–305. Bogen, J. E., Fisher, E. D., & Vogel, P. J. (1965). Cerebral commissurotomy: A second case report. Journal of the American Medical Association, 194, 1328–1329.

98

L.Q. Uddin / Consciousness and Cognition 20 (2011) 94–98

Bogen, J. E., & Vogel, P. J. (1976). Neurological status in the long term following complete cerebral commissurotomy. In B. Schott (Ed.), Les syndromes de disconnexion calleuse chez l’homme (pp. 227–251). Lyon: Hopital Neurologique. Bressler, S. L., & Menon, V. (2010). Large-scale brain networks in cognition: Emerging methods and principles. Trends in Cognitive Sciences, 14(6), 277–290. Brion, S., & Jedynak, C. P. (1972). Disorders of interhemispheric transfer (callosal disconnection). 3 cases of tumor of the corpus callosum. The strange hand sign. Revista De Neurologia (Paris), 126(4), 257–266. Farrer, C., & Frith, C. D. (2002). Experiencing oneself vs another person as being the cause of an action: The neural correlates of the experience of agency. Neuroimage, 15(3), 596–603. Feinberg, T. E., Schindler, R. J., Flanagan, N. G., & Haber, L. D. (1992). Two alien hand syndromes. Neurology, 42(1), 19–24. Funnell, M. G., Corballis, P. M., & Gazzaniga, M. S. (2000). Insights into the functional speciﬁcity of the human corpus callosum. Brain, 123(Pt 5), 920–926. Gallup, G. G. Jr., (1977). Self-recognition in primates: A comparative approach to the bidirectional properties of consciousness. American Psychologist, 32(5), 329–338. Gazzaniga, M. S. (2005). Forty-ﬁve years of split-brain research and still going strong. Nature Reviews Neuroscience, 6(8), 653–659. Gazzaniga, M. S., Bogen, J. E., & Sperry, R. W. (1962). Some functional effects of sectioning the cerebral commissures in man. Proceedings of the National Academy of Sciences of the United States of America, 48, 1765–1769. Gillihan, S. J., & Farah, M. J. (2005). Is self special? A critical review of evidence from experimental psychology and cognitive neuroscience. Psychological Bulletin, 131(1), 76–97. Goldstein, K. (1908). Zur Lehre der motorischen Apraxie. Journal of Psychology and Neurology, 11, 169–187. Keenan, J. P., Wheeler, M., Platek, S. M., Lardi, G., & Lassonde, M. (2003). Self-face processing in a callosotomy patient. European Journal of Neuroscience, 18(8), 2391–2395. 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), 1–15. Lieberman, M. D. (2007). Social cognitive neuroscience: A review of core processes. Annual Review of Psychology, 58, 259–289. Muangpaisan, W., Srisajjakul, S., & Chiewvit, P. (2005). The alien hand syndrome: Report of a case and review of the literature. Journal of the Medical Association of Thailand, 88(10), 1447–1452. Platek, S. M., Loughead, J. W., Gur, R. C., Busch, S., Ruparel, K., Phend, N., et al (2006). Neural substrates for functionally discriminating self-face from personally familiar faces. Human Brain Mapping, 27(2), 91–98. Preilowski, B. (1977). Self-recognition as a test of consciousness in left and right hemisphere of ‘‘split-brain” patients. Activitas Nervosa Superior (Praha), 19(Suppl. 2), 343–344. Preyer, W. (1889). The mind of the child part II: The development of the intellect. New York: Appleton. Sherrington, C. (1941). Man on his nature. New York: Macmillan. Sperry, R. W. (1968). Hemisphere deconnection and unity in conscious awareness. American Psychologist, 23(10), 723–733. Sperry, R. W., Zaidel, E., & Zaidel, D. (1979). Self recognition and social awareness in the deconnected minor hemisphere. Neuropsychologia, 17(2), 153–166. Sugiura, M., Watanabe, J., Maeda, Y., Matsue, Y., Fukuda, H., & Kawashima, R. (2005). Cortical mechanisms of visual self-recognition. Neuroimage, 24(1), 143–149. Turk, D. J., Heatherton, T. F., Kelley, W. M., Funnell, M. G., Gazzaniga, M. S., & Macrae, C. N. (2002). Mike or me? Self-recognition in a split-brain patient. Nature Neuroscience, 5(9), 841–842. Uddin, L. Q. (in press). The self in autism: An emerging view from neuroimaging. Neurocase. 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 Sciences, 11(4), 153–157. Uddin, L. Q., Kaplan, J. T., Molnar-Szakacs, I., Zaidel, E., & Iacoboni, M. (2005a). Self-face recognition activates a frontoparietal ‘‘mirror” network in the right hemisphere: An event-related fMRI study. Neuroimage, 25(3), 926–935. Uddin, L. Q., Rayman, J., & Zaidel, E. (2005b). Split-brain reveals separate but equal self-recognition in the two cerebral hemispheres. Consciousness and Cognition, 14(3), 633–640. Uddin, L. Q., Mooshagian, E., Zaidel, E., Scheres, A., Margulies, D. S., Kelly, A. M., et al (2008). Residual functional connectivity in the split-brain revealed with resting-state functional MRI. NeuroReport, 19(7), 703–709. Weems, S. A., & Zaidel, E. (2005). The effect of response mode on lateralized lexical decision performance. Neuropsychologia, 43(3), 386–395. Zaidel, E. (1975). A technique for presenting lateralized visual input with prolonged exposure. Vision Research, 15(2), 283–289. Zaidel, E., & Clarke, J. M. (1990). Hemispheric independence: A paradigm case for cognitive neuroscience. In A. B. Scheibel & A. F. Wechsler (Eds.), Neurobiology of higher cognitive function. New York: Guilford Press. Zaidel, E., Iacoboni, M., Zaidel, D., & Bogen, J. (2003). The callosal syndromes. In K. M. Heilman & E. Valenstein (Eds.), Clinical neuropsychology (pp. 347–403). New York: Oxford University Press.

Consciousness and Cognition 20 (2011) 99–108

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Depersonalization: A selective impairment of self-awareness q Mauricio Sierra ⇑, Anthony S. David Depersonalization Disorder Unit, Institute of Psychiatry, King’s College, London, United Kingdom

a r t i c l e

i n f o

Article history: Available online 17 November 2010 Keywords: Dissociation Depersonalization Derealization Disembodiment Agency feelings Emotional numbing Anterior insula

a b s t r a c t Depersonalization is characterised by a profound disruption of self-awareness mainly characterised by feelings of disembodiment and subjective emotional numbing. It has been proposed that depersonalization is caused by a fronto-limbic (particularly anterior insula) suppressive mechanism – presumably mediated via attention – which manifests subjectively as emotional numbing, and disables the process by which perception and cognition normally become emotionally coloured, giving rise to a subjective feeling of ‘unreality’. Our functional neuroimaging and psychophysiological studies support the above model and indicate that, compared with normal and clinical controls, DPD patients show increased prefrontal activation as well as reduced activation in insula/limbic-related areas to aversive, arousing emotional stimuli. Although a putative inhibitory mechanism on emotional processing might account for the emotional numbing and characteristic perceptual detachment, it is likely, as suggested by some studies, that parietal mechanisms underpin feelings of disembodiment and lack of agency feelings. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction Depersonalization is a fascinating and intriguing phenomenon, which challenges commonly held assumptions regarding the nature of self. The condition manifests as a pervasive disruption of self-awareness at its most basic, preverbal level (i.e. what it feels like to be an entity, to exist), unlike dissociative conditions such as psychogenic amnesia, or dissociative identity disorder, which typically impair identity at levels involving autobiographical memory, self-narratives, and personality. ‘The person affected with depersonalization complains spontaneously that his or her mental activity, body, and surroundings are changed in their quality, so as to be unreal, remote, or automatized. Among the varied phenomena of the syndrome, patients complain most frequently of loss of emotions and feelings of estrangement or detachment from their thinking, their body, or the real world. In spite of the dramatic nature of the experience, the patient is aware of the unreality of the change. The sensorium is normal and the capacity for emotional expression intact’ (World Health Organization, 1992). Although ‘feelings of unreality’ is still commonly used as a short-hand to describe the phenomenon in clinical practice, most patients stress the ineffable nature of the experience and make use of metaphors which usually take two forms. A ﬁrst kind makes reference to a sense of being cut-off, alienated from oneself and surroundings. For example, patients would often talk about ‘being in a bubble’, or being ‘separated from the world by an invisible barrier such as a pane of glass, a fog, or a veil’ (Sierra, 2009). A second group of metaphors emphasise instead a qualitative change in the state of consciousness, and the feeling as if in ‘a dream’. . .‘stoned’, ‘not awake’ or an indescribable ‘muzzy feeling’, etc. This ineffable aspect of q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap.

⇑ Corresponding author.

E-mail address: [email protected] (M. Sierra). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.10.018

100

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

depersonalization sets it apart from other ‘neurotic’ conditions such as ‘hypochondriasis’, or ‘conversion disorders’, where vivid, detailed and often dramatic descriptions are commonplace. ‘‘What has really been changed or diminished with the onset of depersonalization cannot be expressed in speech. Even educated people (as in some cases in the literature) have given no clearer description, they only used metaphors. Now here is, I think, the point to which the interest of the psychopathologist should be directed. Where normal speech proves unable to deal with an event in consciousness, one may assume that something important is there. Perhaps an underlying brain anomaly makes itself perceptible in this way. Psychopathologists have not bothered very much about this remarkable fact’’ (Mayer-Gross, 1935, p. 106). Another commonly observed feature in patients’ accounts of their experience is the frequent use of the expression ‘as if’ to qualify their descriptions (e.g. ‘I have the feeling as if I am not really here, and as if these were not my hands’ etc.). Such expressions have been traditionally interpreted as evidence of the non-delusional (i.e. nonpsychotic) nature of depersonalization. However, the use of ‘as if’ expressions is more likely to be intended as a critique regarding the adequacy of the description used, rather than a critique of the reality of the experience itself. Thus, while it is true that patients remain painfully aware of the anomalous nature of their experience, they remain convinced that a fundamental, albeit ineffable change has taken place in them. Another conceptual problem with the use of ‘unreality feelings’ as a general descriptor of depersonalization is that the term introduces a negative deﬁnition which has poor explanatory value as it alludes to something missing from normal experience without clarifying its nature (Radovic & Radovic, 2002; Sierra & Berrios, 2001). Historically, there has been disagreement as to the nature of this putative ‘missing experience’, and different writers proposed that depersonalization stemmed from either perceptual, emotional, memory or body image related impairments. Underlying all these hypotheses is the notion that the phenomenological complexity of depersonalization could be reduced to the impairment of a single mental function. An alternative view, that depersonalization could be best conceptualised as a syndrome rather than a symptom, became well established in the ﬁrst half of the 20th century (Shorvon, 1946; Sierra & Berrios, 1997). The following description by Schilder (1928), illustrates this: ‘‘To the depersonalized individual the world appears strange, peculiar, foreign, dream like. Objects appear at times strangely diminished in size, at times ﬂat. Sounds appear to come from a distance. The tactile characteristics of objects likewise seem strangely altered, but the patients complain not only of the changes in their perceptivity but their imagery appears to be altered. Patients characterise their imagery as pale, colourless and some complain that they have altogether lost the power of imagination. The emotions likewise undergo marked alteration. Patients complain that that they are capable of experiencing neither pain or pleasure; love and hate have perished with them. They experience a fundamental change in their personality, and the climax is reached with their complaints that they have become strangers to themselves. It is as though they were dead, lifeless, mere automatons. The objective examination of such patients reveals not only an intact sensory apparatus, but also an intact emotional apparatus. All these patients exhibit natural affective reactions in their facial expressions, attitudes, etc.; so that it is impossible to assume that they are incapable of emotional response’’. In the above description Schilder describes four main and distinct experiential components; namely: (1) an experience of feeling cut-of or alienated from surroundings (i.e. derealization); (2) difﬁculties remembering or imagining things; (3) inability to feel emotions; and (4) a feeling of disembodiment, described as a feeling of being dead, or automaton-like. Interestingly, such four symptom-domains would seem to broadly correspond with those very mental functions historically deemed relevant to the genesis of depersonalization (Sierra & Berrios, 1997). Further evidence supporting the view that depersonalization is characterised by several distinct symptoms was marshalled by a study, which compared 200 historical cases of chronic depersonalization published in the neuropsychiatric literature since the late 19th century, with 45 current patients with depersonalization disorder (DPD). The study revealed the presence of ﬁve symptoms which showed little variation between the historical and modern clinical samples (Sierra & Berrios, 2001): (1) complaints of changes in body experience; (2) automaton-like feelings (i.e. loss of feelings of agency); (3) emotional numbing; (4) changes in the subjective experience of imagery and autobiographical recollections; and (5) complaints of changes in visual perception of surroundings. In spite of its apparent symptom diversity, it might still be the case that depersonalization could result from a single, pervasive experience of detachment equally affecting all aspects of experience. When described separately with regard to emotions, body experiencing, etc., this pervading detachment experience might give rise to the illusion of multiple symptoms. However, the fact that not all symptoms are always present; that some seem more stable than others, or show differential intensity (Sierra & Berrios, 2001), suggests that at least some of these symptoms belong to different experiential domains, with potentially distinct underlying mechanisms (Sierra & Berrios, 1998; Sierra, Lopera, Lambert, Phillips, & David, 2002). Furthermore, two recent exploratory factor analysis studies using the Cambridge Depersonalization Scale (CDS), support the view that, rather than being a one-dimensional construct, ‘depersonalization’ represents the expression of several distinct underlying dimensions (Sierra & Berrios, 1999; Sierra, Baker, Medford, & David, 2005; Simeon et al., 2008). The ﬁrst study was carried out on 145 DPD patients, most of whom had long-standing, constant depersonalization feelings (Sierra et al., 2005). Four well differentiated factors were found and were labelled as follows: (1) Anomalous body experience. (2) Emotional numbing. (3) Anomalous subjective recall. (4) Alienation from surroundings (i.e. derealization).

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

101

Moreover, the fact that an oblique rotation (a statistical factoring model which assumes correlation among factors), yielded a better solution than an orthogonal rotation (a model which assumes independent factors), suggests that the different components of depersonalization represent a cohesive clinical entity rather than the mere coexistence of unrelated phenomena. Recently, Simeon et al. (2008) used the CDS to carry out a conﬁrmatory factor analysis on 450 affected subjects and obtained a strikingly similar factorial solution. Four of their ﬁve factors clearly overlapped with those found by Sierra et al. (2005). In summary, converging evidence from both historical and contemporary phenomenological analysis of depersonalization, suggests that rather than being a unitary experience (i.e. feelings of unreality), the condition is likely to represent a clinical composite of several distinct symptoms: (1) feelings of disembodiment, (2) emotional numbing, (3) anomalous subjective recall, and (4) derealization (i.e. a feeling of alienation from surroundings). According to this syndromal view, ‘Depersonalization’ is a generic term encompassing all the above symptoms including ‘derealization’. This represents a departure from the prevalent assumption, which considers depersonalization and derealization are distinct independent conditions. What follows describes each of the constituent symptoms of depersonalization in some detail. 1.1. Disembodiment feelings Patients with depersonalization complain of a variety of related changes in body experience, which can be conceptualised generically as ‘disembodiment’. These are (1) Lack of body ownership feelings. (2) Feelings of loss of agency, which refer to the feeling that actions happens automatically without the intervention of a willing self. (3) Experience of disembodiment, which can range from a non-speciﬁc feeling of not being in the body, and heightened self-observation, to out of-body experiences, and autoscopic hallucinations. The latter two, however, are rare in depersonalization (Gabbard, Twemlow, & Jones, 1982). (4) Somatosensory distortions usually affecting the size of body parts or feeling very light have not been found to be characteristic of depersonalization and may be useful in the differential diagnosis with conditions such as schizophrenia, epilepsy or migraine, where somatosensory distortions are said to be frequent (Priebe & Rohricht, 2001; Rohricht & Priebe, 2002; Watanabe, Takahashi, & Tonoike, 2003). Interestingly, these profound subjective distortions in body image do not seem accompanied by objective changes in body schema (i.e. implicit regulation of posture and movement in relation to surrounding space (Cappon & Banks, 1965). 1.2. Emotional numbing Most patients with depersonalization report different degrees of attenuated emotional experience such as loss of affection, pleasure, fear or disgust. Some patients describe an absolute inability to experience emotional states, others describe a more subtle impairment characterised by an inability to experience emotional feelings which normally colour perception and mental activity. It has been suggested that the latter impairment may be causally related to descriptions of things looking ‘unreal’ (Sierra & Berrios, 1998). Indeed, the narratives of patients often suggest that this might be case: ‘‘[as I hear music] there is no response in me. Music usually moves me, but now it might as well be someone mincing potatoes ... I seem to be walking about in a world I recognise but don’t feel. I saw Big Ben alight last night, normally a moving sight to me, but it might have been an alarm clock for all I felt ... My husband and I have always been happy together but now he sits here and might be a complete stranger. I know he is my husband only by his appearance – he might be anybody for all I feel towards him’’ (Bockner, 1949). Such statements would seem to suggest that what seems more affected in depersonalization is the ability to imbue perceived objects or concrete situations with emotional feeling, rather than a general inability to experience emotional states (Sierra & Berrios, 1998). A related complaint is that of an inability to experience empathy and compassionate feelings. Lawrence et al. (2007) compared 16 DPD patients with 48 healthy controls along a series of tests designed to provide a measures of two types of empathy: cognitive and affective. In short, while cognitive empathy reﬂects the capacity to understand another person’s emotional state, ‘affective empathy’ reﬂects the ability to experience a congruous emotional response. The main ﬁndings of this study was that while patients with depersonalization showed an intact performance on cognitive empathy, there was evidence of a disruption in implicit physiological concomitants of affective empathy. Comparable ﬁndings emerged from a study looking at the emotional responses to emotive pictures of patients with DPD as compared with normal controls and anxiety disorder patients. Although patients with depersonalization did not experience any difﬁculties when rating the unpleasantness of pictures on a scale, they showed attenuated autonomic responses to arousing pictures and rated them as subjectively less arousing (Sierra et al., 2002). Just as it seems to be the case with anomalies of body experience, subjective complaints of emotional numbing are usually accompanied by a normal array of emotional motor expression. Such dissociation is important in the differential diagnosis given that in other conditions in which emotional numbing can be seen, such as in schizophrenia, depression or PTSD, subjective complaints are accompanied by impoverished emotional expression. In this regard, emotional numbing in depersonalization has shown itself to be a distinct and robust psychopathological concept which can be differentiated from anhedonia (Mula et al., 2010). 1.3. Anomalous subjective recall Patients with DPD often complain of subtle subjective impairments affecting recall and imagery. Although the ability to retrieve information seems unaffected, patients frequently complain that memories, particularly of personal events (i.e. episodic memory) seem to have lost any personal meaning: ‘‘I can remember things, but it seems as if what I remember did not

102

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

really happened to me’’. Such complaints would seem to correspond to a dissociation between what have been termed the ‘know/remember’ components of autobiographical memories (Gardiner & Java, 1991). In short, in addition to the retrieval of factual information about a personal event (i.e. a factual or ‘know’ component), the act of remembering also entails an awareness or particular feeling, that the experience recalled actually happened in the past and is not just being imagined or the memory of a dream. Unlike the case with ‘psychogenic amnesias’, the ‘factual’ aspect of the memory is preserved in depersonalization while it is the ‘remembering’ component which becomes disrupted in some patients. A recent study on 14 patients with DPD found that although patients did not differ from controls in a free recall performance task after watching a movie clip, they exhibited subjective and objective memory fragmentation as measured by their inability to sequence in temporal order a series of scenes extracted from the watched clip (Giesbrecht, Merckelbach, van Oorsouw, & Simeon, 2010). Another common clinical observation is that autobiographic memories in depersonalization are usually remembered from a vantage point outside of the body. That is, the event is visualised as if it had been witnessed from outside, rather than through the person’s own eyes. This type of memory distortion, which has been called ‘observers perspective’ remembering (Nigro & Neisser, 1983), has been shown to affect the recall of traumatic situations, or situations which were experienced as threatening (Sierra and Berrios, 1999). Kenny and Bryant (2006), investigated the relationship between memory vantage point and avoidance following trauma in 60 trauma survivors with differing levels of avoidance. It was found that avoidant individuals were more likely to remember their trauma from an observer perspective than individuals with a lower level of avoidance. Interestingly, avoidance did not inﬂuence vantage point for positive or neutral memories. These results support the view that the adoption of the observer vantage point for unpleasant memories may serve an avoidant function for people affected by trauma. Similar results have been reported in regards to distressing memories in depression (Williams & Moulds, 2006), and memories related to social interactions in social phobics (D’argenbeau et al., 2006). A related complaint affecting memory is that depersonalized patients often characterise their imagery as pale, colourless or completely absent. Lambert et al. (2001b) assessed visual imagery in 28 patients with depersonalization disorder using the Vividness of Visual Imagery Questionnaire (VVIQ) and the Vividness of Movement Imagery Questionnaire (VMIQ). The former is a 16-item scale consisting of descriptions of visual scenes that the subject is asked to imagine, and rate on a 5-point scale ranging from 1 = ‘perfectly clear and as vivid as normal vision’ to 5 = ‘no image at all’. The VMIQ, in turn is a 24-item scale consisting of movements that the subject is requested to imagine. Using the same 5-point scale as above, the items of this questionnaire request subjects to imagine somebody else performing a movement, and then to repeat the items this time imagining that they are themselves making the movements. As compared with a group of age and sex matched normal controls the depersonalization patients were found to have a signiﬁcant impairment of imagery on both the VVIQ and the VMIQ measures. Interestingly, patients showed more impairment on the VVIQ with those items requesting to imagine situations involving people as opposed to objects or scenery. On the VMIQ patients were more impaired at imagining themselves making movements, as compared with imagining another person making the same movement. In fact, this difﬁculty to imagine oneself making movements was found to correlate signiﬁcantly with the intensity of depersonalization as measured by the DES-Taxon. There was however a potential confounding contribution from depressed mood as the latter also correlated with impaired ability to generate visual images. A subgroup of 10 patients was further tested on a neuropsychological battery of visual perception tests and found to be unimpaired compared with normal controls and patients with obsessive compulsive disorder, despite subjective impairments in imagery (Lambert, Senior, Fewtrell, Phillips, & David, 2001a). 1.4. Derealization Most patients with depersonalization describe feelings of being cut-off from the world around, and of things around seeming ‘unreal’. Such an experience is frequently described in terms of visual metaphors (e.g. looking through a camera, mist, veil, etc.). The term derealization was coined in 1935 and ascribed to Mapother by Mayer Gross. Although it has been suggested that apparent phenomenological differences between depersonalization and derealization might simply reﬂect different descriptive angles of the same experience rather than different phenomena (Sierra et al., 2005), it might be argued that there are genuine phenomenological differences between symptoms pertaining to body, emotional and memory experiencing, and that of perception of surroundings. ‘Derealization’ commonly accompanies all the other symptom domains of DPD, and its isolated occurrence has been questioned or reported as extremely rare. Thus, Coons reported to have found only two papers which suggested that derealization can occur alone (Rosen, 1955; Krizek, 1989), but careful reading of these two case reports suggests that symptoms of derealization and depersonalization occurred together in both patients. Lambert et al. (2001a) found that among 44 patients with depersonalization derealization syndrome only four suffered from ‘‘pure derealization’’. Depersonalization has been shown to correlate with anxiety measures (Trueman, 1984), and patients with a diagnosis of DPD, a condition characterised by chronic depersonalization are often found to have high levels of anxiety (Baker et al., 2003). Additionally, it has been observed that the onset of depersonalization often coincides with stressing life-events or even life threatening situations. This has been interpreted as suggesting that depersonalization represents an anxietytriggered ‘hard wired’ inhibitory response intended to ensure the preservation of adaptative behaviour during situations normally associated with overwhelming and potentially disorganizing anxiety (Sierra & Berrios, 1998). It has been proposed that

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

103

such inhibitory response is mediated by a fronto-limbic suppressive mechanism, which would generate a state of emotional numbing, and disable the process by means of which perception (including that of one’s own body), as well as cognition become emotionally coloured. Such ‘decolouring’ will result in a ‘‘qualitative change’’ of conscious awareness, which is then reported by the subject as ‘‘unreal or detached’’. In patients with DPD this response would become abnormally persistent and dysfunctional (Sierra & Berrios, 1998). Studies carried out during the last decade seem supportive of this model. 2. Psychophysiological studies Lader and Wing (1966) ﬁrst reported anecdotal observations in patients with pathological anxiety in whom the onset of depersonalization coincided with a dramatic ﬂattening of their previously labile galvanic skin responses. An even earlier work, looking at the psychophysiological effects of repeated electrical shocks on healthy subjects, had noted that, at the time of receiving high intensity shocks, subjects often described feelings of derealization or of becoming detached observers of themselves. Coinciding with this, there was a blunting in their skin conductance recordings (Oswald, 1959). A less dramatic conﬁrmation of such ﬁndings has recently been obtained. Sixty-nine undergraduate students were exposed to a succession of 19 aversive auditory probes while their skin conductance responses were measured. It was found that the occurrence of acute dissociative experiences (including depersonalization), during the experiment was associated with a fast attenuation of skin conductance responses (Giesbrecht, Merckelbach, ter Burg, Cima, & Simeon, 2008). Kelly and Walter (1968), were probably the ﬁrst to study the sympathetic autonomic system in patients suffering with continuous, chronic depersonalization states. Using forearm blood ﬂow as a measurement of sympathetic autonomic function they found that a group of 8 ‘depersonalized patients’ had the lowest ‘basal’ recordings as compared with a whole range of clinical control groups including chronic anxiety, and depression. Although the highest basal autonomic activity was observed in the chronic anxiety group (almost four times higher than observed in the depersonalization group), both groups had similarly high subjective anxiety. The authors concluded: ‘‘The evidence suggests that the discrepancy between subjective and objective signs of anxiety is the fundamental characteristic of patients with depersonalization. In physiological terms, anxiety is experienced but is not translated into defence reaction arousal’’ (Kelly & Walter, 1968). Sierra, Senior, et al. (2002) tested the prediction that the observed attenuation of sympathetic autonomic arousal was selective to emotional stimuli as compared with neutral and non-speciﬁc stimuli. The skin conductance responses of 15 patients with DPD, 15 controls, and 11 individuals with anxiety disorders were recorded in response to non-speciﬁc elicitors of electrodermal responses (an unexpected clap and taking a sigh) and in response to pictures with neutral and both pleasant and unpleasant emotional contents. It was found that the depersonalization patients had selectively reduced autonomic responses to unpleasant pictures but not to neutral or pleasant ones (the response to these was also reduced but the difference was not statistically signiﬁcant). Further, the latency of response to these stimuli was signiﬁcantly prolonged in the group with DPD. In contrast, latency to non-speciﬁc stimuli (clap and sigh) was signiﬁcantly shorter in the depersonalization and anxiety groups than in the healthy controls. These ﬁndings suggested the presence of both inhibitory and facilitatory mechanisms on autonomic arousal, which suggests a speciﬁc disruption in emotion processing rather than a non-speciﬁc dampening effect on autonomic reactivity. Another related study compared the skin conductance responses (SCR) of DPD patients with those of anxiety disorder patients and normal controls as they watched pictures and video clips of facial expressions of disgust and happiness (Sierra, Senior, Phillips, & David, 2006). In keeping with Kelly and Walter’s study it was found that while patients in the anxiety group had increased autonomic reactivity to disgust expressions, depersonalization patients had very similar responses to those of the healthy controls in spite of showing similarly high anxiety scores to those of the anxiety group. In other words it would seem that the blunting effect on autonomic responses seen in depersonalization is not absolute but relative to anxiety levels. Another recent study compared the skin conductance level (SCL) of DPD patients and healthy controls as they watched an arousing scene from the horror movie ‘The Silence of the Lambs’. As compared with healthy controls, depersonalized participants exhibited an earlier peak followed by subsequent ﬂattening of SCLs which failed to return to baseline levels after termination of the stimulus. The existence of both excitatory and inhibitory inﬂuences on autonomic responses in depersonalization ﬁnds further support in a study, which compared levels of urinary norepinephrine of patients with DPD and healthy controls. In keeping with their higher anxiety levels, patients with depersonalization were found to have higher levels of norepinephrine than healthy controls. However, within the depersonalization group there was a striking negative correlation between norepinephrine and depersonalization scores (Simeon, Guralnik, Knutelska, Yehuda, & Schmeidler, 2003). It seems plausible to suggest that autonomic responses in patients with depersonalization are likely to reﬂect a balance between two opposing tendencies. One excitatory determined by anxiety levels, and an inhibitory one determined by depersonalization intensity. 3. Functional neuroimaging studies Over the last decade a number of functional neuroimaging studies have revealed abnormal brain activation patterns which seem functionally related to both the autonomic changes and subjective experiences already discussed. One of the ﬁrst studies used positron emission tomography (PET) to compare patterns of brain activation of 8 patients with DPD patients with normal controls as they performed a verbal memory task (Simeon et al., 2000). Although patients showed reduced metabolic activity in some association areas such as the right superior and middle temporal gyri, other areas, such

104

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

as the parietal, and occipital lobes were more active than the controls. Interestingly, one of the most signiﬁcant ﬁndings was that of an abnormally increased activation in the angular gyrus of the right parietal lobe, which correlated (r = 0.7) with ratings of depersonalization intensity. The potential signiﬁcance of abnormal parietal functioning in depersonalization is further suggested by a recent open label trial using low frequency repetitive transcranial magnetic stimulation (TMS) on the right temporoparietal junction in 12 patients with DPD (Mantovani et al., 2010). It was found that after 3 weeks treatment half of the patients showed signiﬁcant improvement. Experimental neuroimaging studies on the neural correlates of embodiment and agency feelings, have identiﬁed a network of parietal regions, which appear to play an important role in the generation of embodiment and agency feelings: the inferior parietal cortex, the temporoparietal junction, and the posterior insula. Increased activation in the angular gyrus has been observed in patients experiencing a lack of agency feelings regarding movement or the experience that movements are being controlled by an external agency (Farrer et al., 2004; Frith, Blakemore, & Wolpert, 2000). As mentioned above, subjective experiences of lack of agency feelings are often reported by patients with depersonalization. It is currently believed that the right angular gyrus computes discrepancies between intended action and subsequently experienced movement, allowing any detection of mismatch to be consciously experienced (Farrer et al., 2004, 2008). It is likely that the experience or observation of movements which do not feel as arising from the self elicits an attentional orientation response, similar to that elicited by unexpected events. In addition to the angular gyrus, the posterior insula has also been shown to play a signiﬁcant role in the integration of different input signals related to self-awareness. For example it has been shown that decreased activity in this region corresponds with a decreasing feeling of movement control. Thus, subjects with minimal posterior insula activation report such a striking absence of feelings of agency that when they move it feels to them that they are watching movements of another person (Farrer & Frith, 2002; Farrer et al., 2003). In keeping with these ﬁndings, studies in stroke patients have shown that lesions to the posterior right insula are associated with lack of ownership feelings regarding the existence, or activity of contralateral limbs (Baier & Karnath, 2008; Karnath et al., 2005). In summary, studies on the neurobiological underpinning of agency feelings have shown that while posterior insula activation correlates with the degree of self-attribution of movement, the angular gyrus in the inferior parietal cortex shows the opposite pattern so that the lower the sense of agency the greater the activity in the right inferior parietal lobe (Farrer & Frith, 2002; Farrer et al., 2003, 2008). Another related and partially overlapping parietal region, the temporoparietal junction, has been shown to play an important role in the experience of embodiment, and both pathology and electrical stimulation of this area can generate out-of-body experiences (Blanke, Landis, Spinelli, & Seeck, 2004; De Ridder, Van Laere, Dupont, Menovsky, & Van de Heyning, 2007). Other studies using functional neuroimaging have been designed to explore the role of emotional numbing in depersonalization. The ﬁrst of those studies used functional magnetic resonance imaging (fMRI) to compare the neural response of patients with DPD with that of healthy volunteers and patients with obsessive compulsive disorder. Participants were scanned as they watched a series of aversive and neutral pictures extracted from a well known and standardised picture set (the international affective picture system; IAPS). Attesting to the presence of subjective emotional numbing, depersonalized patients stated that although they could understand the content of the pictures they failed to experience any subjective emotional response. It was found that while healthy volunteers as well as obsessive compulsive disorder (OCD) patients showed activation of the anterior insula in response to unpleasant and disgusting pictures, such activation was not observed in DPD patients. Intriguingly, the anterior insula has been found to play an important role in the neurobiological underpinnings which allow autonomic body states to be consciously experienced as emotional feelings (Craig, 2009). Another key ﬁnding of this study was that depersonalization patients, but not the controls, showed an area of activation in the right ventrolateral prefrontal cortex (BA 47). This region seemed functionally coupled with the insula. Indeed, during the presentation of unpleasant stimuli there was evidence of an inverse correlation, so that prefrontal activation only occurred in the absence of insula activation. Interestingly, the prefrontal area in question has been implicated in the evaluation of negative or aversive information and on exerting control over both emotional experience and its impact on decision making (Beer, Knight, & D’Esposito, 2006). A recent fMRI study has marshalled further evidence of a fronto-limbic inhibitory mechanism in DPD, and also has shown how it might relate to the ﬁnding of attenuated autonomic responses (Lemche et al., 2007, 2008). The researchers used event-related fMRI and simultaneous skin conductance measures to compare the neural responses of 9 patients with DPD and 12 healthy controls as they viewed pictures of faces showing different intensities of sadness and happiness. The ﬁrst ﬁnding of the study was that depersonalized patients showed decreased activity in the amygdala and hypothalamus which was greater with increased emotional expression. Normal controls, in turn, yielded an opposite pattern. The second key ﬁnding was that, in the depersonalization patients but not in the controls, activity in a region of the dorsolateral prefrontal cortex (BA 9) previously implicated in emotional regulation was negatively correlated with autonomic response, hence suggesting an inhibitory role on limbic functioning. Yet another fMRI study has provided further evidence that, in keeping with their subjective complaints of emotional numbing, patients with DPD show abnormalities in the processing of emotionally salient material (Medford et al., 2006). The authors predicted that patients would not show the well established enhancing effect that emotion has on memory. It was also expected that patients would not show activation of emotion-related brain regions during encoding and recognition of emotional words. The authors compared 10 DPD patients with healthy controls while they performed an emotional memory test. It was found that while controls activated a number of emotion-relevant brain regions including the right

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

105

amygdaloid complex, hippocampus and the left temporal gyrus and anterior insula, none of these areas were activated in depersonalization patients. In fact this group showed no differences in brain activation in response to emotional vs. neutral words. In summary, in keeping with patients’ subjective complaints of non-existent or attenuated subjective emotional experiencing, fMRI ﬁndings suggest that depersonalization is consistently characterised by reduced activity in emotional-related areas such as the amygdala and the anterior insula as well as attenuated autonomic responses to arousing emotional stimuli. An inhibitory mechanism mediated by the prefrontal cortex on limbic structures and the anterior insula might explain some of the experiential aspects of depersonalization. In particular, an impairment of the process whereby emotional experience becomes integrated with perception might result in a ‘qualitative change’ on subjective conscious experiencing best described by those affected as ‘feelings of unreality’. One hypothesis arising from this model is that disabling this inhibitory network with, for example TMS, should improve the symptoms of depersonalization. The idea that the ‘immediacy’, and ‘vividness’ (i.e. ‘feelings of reality’) of experience might be determined by concomitant emotional feelings has been previously suggested in the literature. For example, as early as 1925 MacCurdy wrote: ‘the feeling of the reality attaching to any idea is proportionate to our emotional interest in it. Loss of the feeling of reality is, then only a manifestation of loss of interest which is, in turn, related to the loss of energy and stimulus susceptibility’ (MacCurdy, 1925, p. 126). This idea has found support in neurophysiological and cognitive studies. For example, it has been shown that a reduction in the affect attached to a memory can cause it to be experienced as if one had been a detached external observer at the time rather than a direct participant (Robinson & Swanson, 1993). Similarly, as described above, when evoking personal recollections patients with depersonalization often complain that memories feel as if they really didn’t happen to them. In other words, autobiographic memories retain their factual aspect (i.e. their informational load) but seem devoid of the distinct feeling that accompanies the act of remembering. A feeling that seems emotional in nature to a signiﬁcant extent (Sharot, Delgado, & Phelps, 2004). It is worth noticing that lack of ‘emotional colouring’ might also be related to feelings of ‘unreality’ regarding one’s own body. In this regard, it is perhaps not surprising that those body parts which tend to evoke a greater emotional resonance (e.g. face, hands) are the ones most commonly reported as ‘feeling unreal’ by patients with depersonalization (Shorvon, 1946), as the following case reported by Schilder (1935, p. 139): ‘‘[Depersonalization] occurs, as I have shown, especially in organs which have previously been of a great erotic signiﬁcance. I have observed a singer who showed depersonalization concerning speech and concerning the mouth, an organ to which she paid special attention, in herself as well as in others’’. The view that ‘emotional feelings’ may be a core experiential component of perception rather than just a reaction to it has been a neglected idea in neuropsychology. It is likely, however, that in addition to a pathway of information processing leading to semantic recognition (a ‘what is it? perceptual function), there is a parallel pathway in charge of assigning emotional signiﬁcance to percepts (Halgren & Marinkovic, 1994). Such an ‘emotion colouring’ mechanism is likely to be a major contributor to feelings usually described in terms of ‘immediacy’, ‘atmosphere’ or ‘vividness’ (Gloor, 1990). There is evidence suggesting that these two parallel functions take place pre-consciously (Halgren & Marinkovic, 1994), which may explain why, when perception becomes conscious, it is already ‘emotionally coloured’ (Halgren, 1992). Neuropsychological evidence is compatible with the view that these cognitive and emotional components of perception are independent of one another. On the one hand, a disruption of perceptual identiﬁcation with preservation of emotional response has been demonstrated for example in subjects with prosopagnosia, who in spite of not being able to consciously recognise pictures of relatives, show evidence of implicit emotional recognition as measured by autonomic responses (Tranel & Damasio, 1985). Similarly, it has been reported that electrical stimulation targeting temporal lobe structures such as the amygdala, often trigger hallucinatory phenomena, which although fragmentary and lacking in perceptual detail, are experienced vividly and with a strong feeling of reality and personal relevance. This has been explained in terms of the simultaneous presence of an emotional component that colours the experience (Gloor, 1990). On the other hand, a failure to display normal autonomic responses in the face of intact perceptual recognition has been shown in patients with Capgras syndrome (Brighetti, Bonifacci, Borlimi, & Ottaviani, 2007; Ellis, Young, Quayle, & De Pauw, 1997; Hirstein & Ramachandran, 1997). Thus, while these patients do not experience any problems at identifying relatives, they claim that the person in question does not feel genuine and must be an impostor. The accompanying lack of autonomic responses is revealing in that recent evidence shows them to be instrumental to the conscious experiencing of familiarity feelings (Morris, Cleary, & Still, 2008). It has been suggested that the lack of familiarity feelings in Capgras arises from a functional disconnection between face-processing areas in the temporal lobe and the limbic system (Hirstein & Ramachandran, 1997). Similar phenomena have been described with regard to a wide range of objects in addition to people, such as buildings, places, and objects. (Abed & Fewtrell, 1990; Benson, Gardner, & Meadows, 1976). Patients with depersonalization seem to experience a similar, non-delusional version of this phenomenon. ‘‘When I look at my parents I know who they are, but at the same time they feel different, as if they were people I don’t really know’’. In fact a noted high prevalence of depersonalization in patients with Capgras syndrome or reduplicative paramnesia has been interpreted as suggesting that the latter represent a delusional elaboration of depersonalization experiences (Christodoulou, 1986). The neurobiological structures and mechanisms in charge of assigning emotional signiﬁcance to percepts, and the ensuing generation of ‘emotional colouring’, is still far from being completely understood. It is clear, however, that the amygdala plays an important role. It has been clearly established that the amygdala is crucial for the perception of threat as well as the integration of fear responses. Humans with amygdala lesions show impaired aversive conditioning learning (Bechara et al., 1995; LaBar, LeDoux, Spencer, & Phelps, 1995), seem incapable of recognising facial expressions of fear (Adolphs,

106

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

Tranel, Damasio, & Damasio, 1994; Calder et al., 1996; Young et al., 1995), voice intonation expressing fear and anger (Scott et al., 1997), and the recognition of sad or scary music (Gosselin, Peretz, Johnsen, & Adolphs, 2007). In addition to fear-related functions, the amygdala also seems to have a more generic role in the processing and assignment of emotional signiﬁcance, as well as to play a modulatory role on cognitive functions such as attention, perception and memory (LeDoux, 2007). Neuroimaging studies also show activation of the amygdala during the recall of emotionally charged memories (Cahill, Haier, Fallon, et al., 1996), and it would seem that such activation contributes to the ‘feeling’ of remembering (Sharot et al., 2004). The fact that in patients with depersonalization a lack of subjective emotional feelings coexists with adequate behavioural emotional expressions gives support to the idea that in depersonalization there is a disruption of the process which allows emotions to gain conscious representation, rather than a global dysfunction of emotion processing. Recent research has identiﬁed the anterior insula as a crucial cortical region necessary for the experience of emotional feelings (Critchley, 2005; Morris, 2002). Such ﬁndings are in keeping with the reduced insula activation found in patients with DPD (Phillips et al., 2001). From an anatomical perspective the insula seems to be well placed to integrate signals from a variety of sources. It receives visceral, somatosensory, visual, auditory and gustatory inputs, and has extensive reciprocal connections with the amygdala, hypothalamus, cingulate gyrus and orbitofrontal cortex (Höistad & Barbas, 2008; Mesulam & Mufson, 1985). It has been proposed that one of the main functions of the anterior insula would be that of integrating peripheral autonomic responses with central ‘cognitive’ processing, allowing visceral responses to gain conscious representation in the form of subjective feelings (Morris, 2002). It has been shown that subjects with a condition known as pure autonomic failure, who are unable to generate autonomic responses, have a reduced capacity to experience conscious feeling including empathy (Chauhan, Mathias, & Critchley, 2008; Critchley, Wiens, Rotshtein, Ohman, & Dolan, 2004). Furthermore, it has been experimentally shown in fMRI studies that the ability to experience feelings in response to emotional pictures is directly related to activity in the anterior insula. In particular, it would seem that an inability to become aware of feelings is related to hypoactivity in the anterior insula (Silani et al., 2008). Anterior insula activation has been related to whole range of emotional feelings such as disgust, sadness, fear, reward experiences, categorization of facial emotional expressions, craving, and hunger or satiety states (Morris, 2002). It is also been involved in the experience of socially laden feelings such as the sense of ‘fairness’ (Moll et al., 2007). Such a vast array of activation correlates suggests that the role of the anterior insula in the generation of feelings is generic rather than speciﬁc to any particular emotion. In light of the evidence reviewed so far it would seem as if there are two distinct neural networks that may be relevant to the neurobiology of depersonalization. The ﬁrst system, relevant to the experience of emotional feelings, includes the amygdala, the anterior insula and possibly other limbic-related structures such as the hypothalamus and the anterior cingulate. The activity of this emotional system is strongly regulated by the prefrontal cortex, and it is suggested that in depersonalized subjects abnormal prefrontal regulatory suppression might be responsible for emotional numbing and the related inability to colour experience with feelings. This hypothesis is supported with the ﬁndings discussed above of attenuated autonomic responses, underactive amygdala and anterior insula responses, as well as related increased activation in prefrontal regions in depersonalized subjects. Of the four different symptom domains of the depersonalization syndrome reviewed above, it would seem that three of them, namely ‘derealization’, ‘anomalous subjective recall’, and ‘emotional numbing’, might be related to fronto-limbic suppression. A second neural network relevant to the experience of embodiment and feelings of agency, may be implicated in feelings of disembodiment, lack of body ownership and lack of agency feelings experienced by patients with DPD. The idea that the depersonalization syndrome might be mediated by anomalies in two distinct albeit related networks (i.e. fronto-limbic and parietal) is supported by the observation that patients with distinct neurological lesions may complain of either derealization or ‘disembodiment’. The phenomenological similarities between visual hypoemotionality, a neurological syndrome, which allegedly results from a cortico-limbic disconnection involving visual pathways (Bauer, 1982), and derealization, supports the idea that a disruption of the processes by which perception becomes emotionally coloured may be an underlying mechanism in both conditions. Likewise, phenomenological overlaps with asomatognosia suggest that the ‘disembodiment’ component of depersonalization might result from parietal mechanisms disrupting the experience of body ownership and agency (Sierra, Lopera, et al., 2002). It any event, the evidence suggest that the neural networks supporting body and emotional representation are closely linked. For example, a study using structural neuroimaging on 108 subjects with focal brain lesions found that the perception of intensity in facial emotional expressions required the integrity of the right somatosensory cortices (Adolphs, Damasio, Tranel, Cooper, & Damasio, 2000). Future studies on neurobiological mechanisms of depersonalization should address the relationship between emotion regulation systems and those underpinning embodiment and agency feelings. Acknowledgments We gratefully acknowledge a grant awarded by the Pilkington-Cecil Charitable trusts. References Abed, R. T., & Fewtrell, W. D. (1990). Delusional misidentiﬁcation of familiar inanimate objects. A rare variant of Capgras syndrome. British Journal of Psychiatry, 157, 915–917. Adolphs, R., Damasio, H., Tranel, D., Cooper, G., & Damasio, A. R. (2000). A role for somatosensory cortices in the visual recognition of emotion as revealed by three-dimensional lesion mapping. The Journal of Neuroscience, 20, 2683–2690.

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

107

Adolphs, R., Tranel, D., Damasio, H., & Damasio, A. (1994). Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature, 372, 669–672. Baier, B., & Karnath, H. O. (2008). Tight link between our sense of limb ownership and self-awareness of actions. Stroke, 39, 486–488. Baker, D., Hunter, E., Lawrence, E., Medford, N., Patel, M., Senior, C., et al (2003). Depersonalisation disorder: Clinical features of 204 cases. British Journal of Psychiatry, 182, 428–433. Bauer, R. M. (1982). Visual hypoemotionality as a symptom of visual–limbic disconnection in man. Archives of Neurology, 39, 702–708. Bechara, A., Tranel, D., Damasio, H., Adolphs, R., Rockland, C., & Damasio, A. R. (1995). Double dissociation of conditioning and declarative knowledge relative to the amygdala and hippocampus in humans. Science, 269, 1115–1118. Beer, J. S., Knight, R. T., & D’Esposito, M. (2006). Controlling the integration of emotion and cognition: The role of frontal cortex in distinguishing helpful from hurtful emotional information. Psychological Science, 17, 448–453. Benson, D. F., Gardner, H., & Meadows, J. C. (1976). Reduplicative paramnesia. Neurology, 26, 147–151. Blanke, O., Landis, T., Spinelli, L., & Seeck, M. (2004). Out-of-body experience and autoscopy of neurological origin. Brain, 127, 243–258. Bockner (1949). The depersonalization syndrome: Report of a case. Journal of Mental Science, 95, 968–971. Brighetti, G., Bonifacci, P., Borlimi, R., & Ottaviani, C. (2007). Far from the heart far from the eye: Evidence from the Capgras delusion. Cognitive Neuropsychiatry, 12, 189–197. Cahill, L., Haier, R. J., Fallon, J., et al (1996). Amygdala activity at encoding correlated with longterm, free recall of emotional information. Proceedings of the National Academy of Sciences, USA, 93, 8012–8016. Calder, A. J., Young, A. W., Rowland, D., Perret, D. I., Hodges, J. R., & Etcoff, N. L. (1996). Facial emotion recognition after bilateral amygdala damage: Differentially severe impairment of fear. Cognitive Neuropsychology, 13, 699–745. Cappon, D., & Banks, R. (1965). Orientational perception II. Body perception in depersonalization. Archives of General Psychiatry, 13, 375–379. Chauhan, B., Mathias, C. J., & Critchley, H. D. (2008). Autonomic contributions to empathy: Evidence from patients with primary autonomic failure. Autonomic Neuroscience, 140, 96–100. Christodoulou, G. N. (1986). Role of depersonalization–derealization phenomena in the delusional misidentiﬁcation syndromes. Bibliotheca Psychiatrica, 164, 99–104. Craig, A. D. (2009). How do you feel now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10, 59–70. Critchley, H. D. (2005). Neural mechanisms of autonomic, affective, and cognitive integration. Journal of Comparative Neurology, 493, 154–166. Critchley, H. D., Wiens, S., Rotshtein, P., Ohman, A., & Dolan, R. J. (2004). Neural systems supporting interoceptive awareness. Nature Neuroscience, 7, 189–195. D’argenbeau, A., Van der Linden, M., d’Acremont, M., & Mayers, I. (2006). Phenomenal characteristics of autobiographical memories for social and non-social events in social phobia. Memory, 14, 637–647. De Ridder, D., Van Laere, K., Dupont, P., Menovsky, T., & Van de Heyning, P. (2007). Visualizing out-of-body experience in the brain. New England Journal of Medicine, 357, 1829–1833. Ellis, H. D., Young, A. W., Quayle, A. H., & De Pauw, K. W. (1997). Reduced autonomic responses to faces in Capgras delusion. Proceedings of the Royal Society London B Biological Sciences, 264, 1085–1092. Farrer, C., Franck, N., Frith, C. D., Decety, J., Georgieff, N., d’Amato, T., et al (2004). Neural correlates of action attribution in schizophrenia. Psychiatry Research, 131, 31–44. Farrer, C., Franck, N., Georgieff, N., Frith, C. D., Decety, J., & Jeannerod, M. (2003). Modulating the experience of agency: A positron emission tomography study. NeuroImage, 18, 324–333. Farrer, C., Frey, S. H., Van Horn, J. D., Tunik, E., Turk, D., Inati, S., et al (2008). The angular gyrus computes action awareness representations. Cerebral Cortex, 18, 254–261. Farrer, C., & Frith, C. D. (2002). Experiencing oneself vs another person as being the cause of an action: the neural correlates of the experience of agency. NeuroImage, 15, 596–603. Frith, C. D., Blakemore, S. J., & Wolpert, D. M. (2000). Abnormalities in the awareness and control of action. Philosophical Transactions of the Royal Society of London Series B Biological Sciences, 355, 1771–1788. Gabbard, G. O., Twemlow, S. W., & Jones, F. C. (1982). Differential diagnosis of altered mind/body perception. Psychiatry, 45, 361–369. Gardiner, J. M., & Java, R. I. (1991). Forgetting in recognition memory with and without recollective experience. Memory and Cognition, 19, 617–623. Giesbrecht, T., Merckelbach, H., ter Burg, L., Cima, M., & Simeon, D. (2008). Acute dissociation predicts rapid habituation of skin conductance responses to aversive auditory probes. Journal of Traumatic Stress, 21, 247–250. Giesbrecht, T., Merckelbach, H., van Oorsouw, K., & Simeon, D. (2010). Skin conductance and memory fragmentation after exposure to an emotional ﬁlm clip in DPD. Psychiatry Research, 177, 342–349. Gloor, P. (1990). Experiential phenomena of temporal lobe epilepsy: Facts and hypotheses. Brain, 113, 1673–1694. Gosselin, N., Peretz, I., Johnsen, E., & Adolphs, R. (2007). Amygdala damage impairs emotion recognition from music. Neuropsychologia, 45, 236–244. Halgren, E. (1992). Emotional neurophysiology of the amygdala within the context of human cognition. In J. P. Aggleton (Ed.), The amygdala: Neurobiological aspects of emotion, memory and mental dysfunction (pp. 191–228). New York: Wiley-Liss. Halgren, E., & Marinkovic, K. (1994). Neurophysiological networks integrating human emotions. In M. Gazzaniga (Ed.), The cognitive neurosciences (pp. 1137–1151). Cambridge, MA: MIT Press. Hirstein, W., & Ramachandran, V. S. (1997). Capgras syndrome: A novel probe for understanding the neural representation of the identity and familiarity of persons. Proceedings Biological Sciences, 264, 437–444. Höistad, M., & Barbas, H. (2008). Sequence of information processing for emotions through pathways linking temporal and insular cortices with the amygdala. Neuroimage, 40, 1016–1033. Kelly, D. H. W., & Walter, C. J. S. (1968). The relationship between clinical diagnosis and anxiety, assessed by forearm blood ﬂow and other measurements. British Journal of Psychiatry, 114, 611–626. Kenny, L. M., & Bryant, R. A. (2006). Keeping memories at an arm’s length: Vantage point of trauma memories. Behaviour Research and Therapy, 45, 1915–1920. Krizek, G. O. (1989). Derealization without depersonalization. American Journal of Psychiatry, 146, 1360–1361. LaBar, K. S., LeDoux, J. E., Spencer, D. D., & Phelps, E. A. (1995). Impaired fear conditioning following unilateral temporal lobectomy in humans. Journal of Neuroscience, 15, 6846–6855. Lader, M. H., & Wing, L. (1966). Physiological measures, sedative drugs and morbid anxiety. Maudsley Monographs No.14. London: Oxford University Press. Lambert, M. V., Senior, C., Fewtrell, W. D., Phillips, M. L., & David, A. S. (2001a). Primary and secondary depersonalisation disorder: A psychometric study. Journal of Affective Disorders, 63, 249–256. Lambert, M. V., Senior, C., Phillips, M. L., Sierra, M., Hunter, E., & David, A. S. (2001b). Visual imagery and depersonalisation. Psychopathology, 34, 259–264. Lawrence, E. J., Shaw, P., Baker, D., Patel, M., Sierra-Siegert, M., Medford, N., et al (2007). Empathy and enduring depersonalization: The role of self-related processes. Social Neuroscience, 2, 292–306. LeDoux, J. (2007). The amygdala. Current Biology, 17, 868–874. Lemche, E., Surguladze, S. A., Giampietro, V. P., Anilkumar, A., Brammer, M. J., Sierra, M., et al (2007). Limbic and prefrontal responses to facial emotion expressions in depersonalization. Neuroreport, 18, 473–477. Lemche, E., Anilkumar, A., Giampietro, V. P., Brammer, M. J., Surguladze, S. A., Lawrence, N. S., et al (2008). Cerebral and autonomic responses to emotional facial expressions in depersonalisation disorder. British Journal of Psychiatry, 193, 222–228. MacCurdy, J. T. (1925). Psychology of emotions: Morbid and normal. London: Kegan Paul.

108

M. Sierra, A.S. David / Consciousness and Cognition 20 (2011) 99–108

Mantovani, A., Simeon, D., Urban, N., Bulow, P., Allart, A., & Lisanby, S. (2010). Temporo-parietal junction stimulation in the treatment of DPD. Psychiatry Research (online ahead of publication). Mayer-Gross, W. W. (1935). On depersonalization. British Journal of Medical Psychology, 15, 103–122. Medford, N., Brierley, B., Brammer, M., Bullmore, E. T., David, A. S., & Phillips, M. L. (2006). Emotional memory in DPD: A functional MRI study. Psychiatry Research, 148, 93–102. Mesulam, M.-M., & Mufson, E. J. (1985). The insula of Reil in man and monkey. Architectonics, connectivity, and function. In A. Peters & E. G. Jones (Eds.). Cerebral cortex (Vol. 4, pp. 179–226). New York: Plenum. Moll, J., de Oliveira-Souza, R., Garrido, G. J., Bramati, I. E., Caparelli-Daquer, E. M., Paiva, M. L., et al (2007). The self as a moral agent: Linking the neural bases of social agency and moral sensitivity. Social Neuroscience, 2, 336–352. Morris, J. S. (2002). How do you feel? Trends in Cognitive Sciences, 6, 317–319. Morris, A. L., Cleary, A. M., & Still, M. L. (2008). The role of autonomic arousal in feelings of familiarity. Consciousness and Cognition, 17, 1378–1385. Mula, M., Pini, S., Calugi, S., Preve, M., Masini, M., Giovannini, I., et al (2010). Distinguishing affective depersonalization from anhedonia in major depression and bipolar disorder. Comprehensive Psychiatry, 51, 187–192. Nigro, G., & Neisser, U. (1983). Point of view in personal memories. Cognitive Psychology, 15, 467–482. Oswald, I. (1959). Experimental studies of rhythm, anxiety and cerebral vigilance. Journal of Mental Science, 105, 269–294. Phillips, M. L., Medford, N., Senior, C., Bullmore, E. T., Suckling, J., Brammer, M. J., et al (2001). DPD: Thinking without feeling. Psychiatry Research, 108, 145–160. Priebe, S., & Rohricht, F. (2001). Speciﬁc body image pathology in acute schizophrenia. Psychiatry Research, 101, 289–301. Radovic, F., & Radovic, S. (2002). Feelings of Unreality: A conceptual and phenomenological analysis of the language of depersonalization. Philosophy, Psychiatry and Psychology, 9, 271–283. Robinson, J. A., & Swanson, K. L. (1993). Field and observer modes of remembering. Memory, 1, 169–184. Rohricht, F., & Priebe, S. (2002). Do cenesthesias and body image aberration characterize a subgroup in schizophrenia? Acta Psychiatrica Scandinavica, 105, 276–282. Rosen, V. H. (1955). The reconstruction of a traumatic childhood event in a case of derealization. Journal of the American Psychoanalytic Association, 3, 211–220. Schilder, P. (1928). Depersonalization. In Introduction to psychoanalytic psychiatry. Nerv. Ment. Dis. Monographies. Series, Vol. 50 (pp 120). Schilder, P. (1935). The image and appearance of the human body. London: Kegan Paul. Scott, S. K., Young, A. W., Calder, A. J., Hellawell, D. J., Aggleton, J. P., & Johnson, M. (1997). Impaired auditory recognition of fear and anger following bilateral amygdala lesions. Nature, 54, 57. Sharot, T., Delgado, M. R., & Phelps, E. A. (2004). How emotion enhances the feeling of remembering. Nature Neuroscience, 7, 1376–1380. Shorvon, H. J. (1946). The depersonalization syndrome. Proceedings of the Royal Society of Medicine, 39, 779–792. Sierra, M. (2009). Depersonalization: A new look at a neglected syndrome. Cambridge University Press. Sierra, M., & Berrios, G. E. (1997). Depersonalization: A conceptual history. History of Psychiatry, 8, 213–229. Sierra, M., & Berrios, G. E. (1998). Depersonalization: Neurobiological perspectives. Biological Psychiatry, 44, 898–908. Sierra, M., & Berrios, G. E. (1999). Flashbulb memories and other repetitive images: A psychiatric perspective. Comprehensive Psychiatry, 40, 115–125. Sierra, M., & Berrios, G. E. (2001). The phenomenological stability of depersonalization: Comparing the old with the new. Journal of Nervous and Mental Diseases, 189, 629–636. Sierra, M., Baker, D., Medford, N., & David, A. S. (2005). Unpacking the depersonalization syndrome: An exploratory factor analysis on the Cambridge Depersonalization Scale. Psychological Medicine, 35, 1523–1532. Sierra, M., Senior, C., Phillips, M. L., & David, A. S. (2006). Autonomic response in the perception of disgust and happiness in depersonalization disorder. Psychiatry Research, 145, 225–231. 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, 530–532. Sierra, M., Senior, C., Dalton, J., McDonough, M., Bond, A., Phillips, M. L., et al (2002). Autonomic response in DPD. Archives of General Psychiatry, 59, 833–838. Silani, G., Bird, G., Brindley, R., Singer, T., Frith, C., & Frith, U. (2008). Levels of emotional awareness and autism: An fMRI study. Social Neuroscience, 3, 97–112. Simeon, D., Guralnik, O., Knutelska, M., Yehuda, R., & Schmeidler, J. (2003). Basal norepinephrine in DPD. Psychiatry Research, 121, 93–97. Simeon, D., Guralnik, O., Hazlett, E. A., Spiegel-Cohen, J., Hollander, E., & Buchsbaum, M. S. (2000). Feeling unreal: A PET study of depersonalization disorder. American Journal of Psychiatry, 157, 1782–1788. Simeon, D., Kozin, D. S., Segal, K., Lerch, B., Dujour, R., & Giesbrecht, T. (2008). De-constructing depersonalization: Further evidence for symptom clusters. Psychiatry Research, 157, 303–306. Tranel, D., & Damasio, A. R. (1985). Knowledge without awareness: An autonomic index of facial recognition by prosopagnosics. Science, 8, 1453–1454. Trueman, D. (1984). Anxiety and depersonalization and derealization experiences. Psychological Reports, 54, 91–96. Watanabe, H., Takahashi, T., & Tonoike, T. (2003). Cenesthopathy in adolescence. Psychiatry and Clinical Neuroscience, 57, 23–30. World Health Organization (1992). F48.1. Depersonalisation-derealisation syndrome. In The ICD-10 Classiﬁcation of mental and behavioural disorders, Clinical descriptions and diagnostic guidelines (pp. 171–173). Geneva, World Health Organisation. Williams, A. D., & Moulds, M. L. (2006). Cognitive avoidance of intrusive memories: Recall vantage perspective and associations with depression. Behavioral Research and Therapy, 45, 1141–1153. Young, A. W., Aggleton, J. P., Hellawell, D. J., Johnson, M., Broks, P., & Hanley, J. R. (1995). Face processing impairments after amygdalotomy. Brain, 118, 15–24.

Consciousness and Cognition 20 (2011) 109–119

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The self as phenotype Philippe Rochat ⇑ Department of Psychology, Emory University, 36 Eagle Row, Atlanta, GA 30322, USA

a r t i c l e

i n f o

Article history: Available online 8 December 2010 Keywords: Self-awareness Infant and child development Brain development

a b s t r a c t Self-awareness is viewed here as the phenotypic expression of an interaction between genes and the environment. Brain and behavioral development of fetuses and newborn infants are a rich source of information regarding what might constitute minimal selfawareness. Research indicates that newborns have feeling (subjective) experience. Unlike automata, they do not just sense and respond to proximal stimulations. In light of the explosive brain growth that takes place inside and outside of the womb, ﬁrst signs of feeling as opposed to sensing experience are discussed. Feeling experience is considered as the necessary condition for having minimal self-awareness. Both would co-emerge in development. However, minimal self-awareness is rapidly supplemented with an awareness that is not just perceptual, but also conceptual and ethical, primarily deﬁned in relation to and by others. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction What is a self and what qualiﬁes for self-awareness? Much can be gained by addressing these questions in the light of ontogeny and the dynamic of changes. Development is too often forgotten in the debate on the nature and origins of self. Here, I want to re-visit these questions from a developmental perspective, in relation to both behavior and brain growth. As a general framework, I propose to think of the self as a phenotype, in the literal sense of an organism emerging from the interaction of the genotype and the environment. At the origins, and at a basic level, it is perceived as something that has form and unity, a Gestalt that is more than the sum of its parts. The self is indeed an organism, in the dictionary sense of ‘‘a form of life composed of mutually interdependent parts that maintain various vital processes” or ‘‘a complex system having properties and functions determined not only by the properties and relations of its individual parts, but by the character of the whole that they compose and by the relations of the parts to the whole” (Random House, Unabridged Dictionary). Here, the goal is to consider the origins of self-awareness in the light of the fast, highly programmed development of brain and behavior, within the maternal womb and beyond. We are interested in the predictable maturation of the nervous system’s anatomy and functions that accompanies and supports less predictable, more open-ended behavioral development. Taking a developmental stance and parallel look at brain and behavior from conception to the point where children start expressing shame and self-consciousness, helps pondering the question of what qualiﬁes for self-awareness and what changes in the course of early development. A developmental approach helps capturing the self as phenotype: an emerging and changing expression, rather than a static and disembodied form. This is the bet. I will start discussing the basic criteria for having self-awareness. To capture the origins of such criteria, I try to link behavior and brain during the pre- and neonatal period of development. For the rest of the article, I review important transitions in the development of self-awareness, leading the child to become eventually self-conscious, ultimately conscientious (ethical) toward others from approximately 3 years of age. I consider some of the structural and functional brain changes ⇑ Fax: +1 404 727 0372. E-mail address: [email protected] 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.012

110

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

that might be implicated in such development leading children toward becoming ‘‘free” (autonomous) moral agents. I view the phenotypic emergence of ‘‘conscientiousness” as a moral kind of self-awareness, the ultimate ﬁnality of child development demanded by all cultures.

2. Criteria for having self-awareness It is common to link self-awareness to higher order cognitive processes associated with late developing cortical structures. From this perspective, one should not attribute any self- or other kinds of awareness to creatures that are not, or not yet endowed with higher cortical structures, namely a mammalian forebrain. But this view is challenged by empirical evidence of goal-directed, purposive behaviors in decorticated animals and anencephalic children (children born without cortex, see Merker (2007) for a cogent review). In general, it appears that self-awareness does not have to be explicit and mediated by higher symbolic processes like language. Infants as well as most animals could be endowed with the potential for self-awareness, depending on what criteria are used. In his argument in support of a consciousness without cortex, Merker (2007) proposes that in ‘‘the most basic and general sense” what qualiﬁes for consciousness (here by extension awareness and self-awareness), is ‘‘the state or condition presupposed by any experience whatsoever” (p. 63). This would mean that only ‘‘experiencing” creatures could be self-aware: those who are capable of having subjective ‘‘what it is like” experience. Such proposal, however, begs the question of what qualiﬁes for having ‘‘any experience whatsoever?” In particular, what is having experience vs. not having experience? I submit that to have experience is to have an affective life that is made of perceived values (or qualities) such as pain, pleasure or cravings. Having experience, hence having an affective life, should transcend the mere recording of raw sensations as ‘‘signals” by an organism. Inversely, not having experience is to sense the world by simply recording raw sensory signals that trigger automatic, predictable responses like in the output behaviors of computers or thermostats. These automata sense but do not feel. One entails minimal subjective experience (feelings), not the other, even though output behaviors might be homologous for third party observers (e.g., ﬂeeing of a cockroach vs. the ﬂeeing of a foot soldier under attack). Feelings as opposed to raw sensory signals are expressed in emotions via particular bodily movements (e.g., facial expressions; particular dynamic of motion or signaling like in screaming). It is the affective life revealed in emotions that distinguishes sentient creatures from mere machines and automata. The former feel and are potentially conscious, the latter are not. Feeling experience adds values to an organism’s encounter with the world. More importantly, in relation to selfawareness, it gives purpose and orientation to the actions performed by that organism. This, I will suggest, is the constitutive element of any form of what we refer as consciousness (the state or condition presupposed by experience), and by extension the necessary pre-requisite of minimal self-awareness. We will see, for example, that it is in the feeling experience of purpose and orientation that infants from birth express an early ecological ‘‘feeling experience” of themselves as differentiated, situated, substantial, and eventually by 2 months as agent entities among other ‘‘distal” entities in the world. Affectivity and emotions guide actions and elevate behaviors above the register of mere automatic and mechanistic responses. Automata respond, they do not act proper. Contrary to sentient (feeling) organisms, they respond like thermostats do in relation to temperature ﬂuctuations within a pre-set, calibrated range. Thermostats do not act to avoid hot or cold, nor are they oriented toward a gain of comfort (pleasure or good feelings maintenance). Machines have however the superior power to ﬁnd and apply algorithms to resolve complex problems with reliable success (see the chess match between I.B.M. Deep Blue machine and Garry Kasparov). Machines can, in some cold sense, think and resolve problems, but they cannot feel and for this reason have no potential for self-awareness, except maybe in science ﬁctions (e.g., computer HAL 9000 experiencing ‘‘fear” in Clarke and Kubrick ‘‘2001 Space Odyssey”). The necessary qualiﬁer for having minimal self-awareness is, accordingly, the feeling experience that elevates organisms from mere responders to volitional actors (goal oriented and purposeful beings in a world made of affective values: pleasures, pains, envies). Self-awareness would co-emerge with feeling experience, both mutually necessary and co-deﬁning. Self-awareness is therefore affective and emotional at the origin, rather than cognitive in the etymological sense of consciousness (‘‘con-scientia” or ‘‘with-knowledge”). If self-awareness has anything to do with knowledge or cognition, it is with ‘‘hot”, not cold cognition: what feels good or bad, what is pleasurable or not, ultimately what feels right and what feels wrong. Children in their development, as I will try to show, demonstrate that the explicit conscious ‘‘content” of the self (explicit and conceptual awareness of Me) rests upon and is eventually added to this basic pre-reﬂective and implicit affective process that distinguishes us from machines. But this implies major steps in brain and behavior development. Within this theoretical framework, the question is: when can we say that an organism has feeling experiences, hence ﬁlls the criteria for having minimal self-awareness? Do cockroaches feel? Are cold blood species (ﬁsh, octopus, reptiles) and those who did not evolve a mammalian brain precluded from having self-awareness? Do birds feel? What about human fetuses or young infants? When do we start to feel, hence have the potential to be self-aware? Looking at the brain, isolating neural networks and brain structures that are predictably associated in time with feeling experiences can help determining whether an organism ‘‘feels” something and might be endowed with subjective experience, therefore qualiﬁes for potential self-awareness. Inversely, just looking at behavior and emotional expressions that are predictably associated with feeling experiences (e.g., pain) can do the same. But ﬁrst, ‘‘peripheral” receptor systems

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

111

(e.g., pain, auditory, vestibular, light, olfactory, or gustatory sensitivity) have to be functional and connected (projecting) to some higher centers of the nervous system for further processing and motor/vegetative responses. A rich body of evidence points to the fact that, in humans, the potential for self-awareness might already be expressed at birth, possibly even during the last weeks of gestation. This is rather revolutionary considering that not so long ago, the idea of feeling-less infants (i.e., non verbal children) was the default assumption. The Zeitgeist was to deny infants any form of worthwhile feeling experience (phenomenal awareness). Proof of it is that in the 1940s and 1950s, surgery without anesthesia was routinely performed on infants and young children. Modern surgeons conveniently paralyzed squirming infants by injection of Curare or similar paralytic agents. Under such circumstances, adults recalled excruciating pain during surgery. But patients were not believed and the practice went on for 20 years. As pointed out by Dennett (1981): ‘‘The fact that most of the patients were infants and small children may explain this credibility gap” (p. 201). Even today, local anesthetics are not routine in painful procedures on newborns such as heel prick and circumcision, even by pediatricians practicing in state of the art maternity hospitals. The Zeitgeist continues to be that infants have either no feelings, less feelings, or that feeling experience at this early stage might not be as consequential for lack of memory (infantile amnesia). Such rationale raises questions when looking at the brain and behavior in pre- and post-natal development. 3. Great momentum of brain growth The basic facts about brain development that I brieﬂy present are meant to remind us of the great biological momentum, epigenetic force and programming behind brain growth. What might correspond to the basic, necessary neurological prerequisites for feeling experiences, is put in place only 8 weeks after conception. Here are some cardinal facts. It takes only 4 weeks from conception for the neural tube to be formed from layers of cells on the embryonic disc (Hepper, 2002). Only one extra week is needed for the basic ﬁve parts structure of the brain to be anatomically differentiated and clearly visible (i.e., Telecephalon and Diencephalon of the Forebrain, Midbrain, Hindbrain and the Spinal cord) (Carlson, 1994). By 11 weeks, Medulla, Cerebellum, Inferior and Superior colliculus, as well as both Cerebral hemispheres covering the Diencephalon are also clearly visible. From then on and for a few years, both hemispheres grow in surface areas via folding grooves and convolutions. This growth reﬂects rapid and exponential connection network among synapses as well as myelination of axons providing insulation (fatty ‘‘white matter”) for better transmission of electrochemical nerve impulses. By 2 years of age, the child’s brain weighs already 80% of its lifetime maximum weight (Kretschmann, Kammradt, Krauthausen, Sauer, & Wingert, 1986). In terms of neural growth, between 10 and 26 weeks gestational age, neurons are produced at a rate of 2500 000 a minute, leading to overproduction. Beyond 26 weeks, more than half of the produced nerve cells are selectively pruned and die. The surviving 100 billions will eventually form the adult brain (Oppenheim, 1991). Regarding connection between cells, there is also an overproduction of synapses that continues beyond birth, with peaking periods that vary across brain regions (Rakic, 1972). Synaptogenesis continuing after birth is not homogeneous and synchronous across brain regions. For example, post mortem data indicate that synaptic density peaks earlier in the auditory cortex (3 months) compared to the middle frontal gyrus (15 months, Huttenlocher & Dabholkar, 1997). This kind of growth asynchrony is reﬂected, for example, in the sequential development of sense modalities in the womb and beyond (i.e., vision lag). In short, these facts remind us that brain emergence is remarkably fast and programmed, literally an explosive growth. This development puts in place within 8 weeks the potential for fetuses to sense the world, eventually by the end of gestation also to feel it, hence to have the potential for minimal self-awareness. However, feeling experience rests on the pre-requisite of an ability to sense the world via systems that are in place and functional, the basis for primary sensitivity. 4. Emerging fetal sensitivity Fetal sensitivity matures sequentially depending on the modality (Lecanuet & Schaal, 1996), recapitulating the evolutionary order of the main sensory systems (Gottlieb, 1971). Somesthetic sensitivity (skin and body feelings) matures ﬁrst. This sensitivity corresponds to tactile (skin pressure), vestibular (posture and balance), and pain stimulation (tissue damage, see Merskey & Bogduk, 1994). It is followed by the maturation of chemosensory sensitivity that combines olfaction and gustation (i.e., smell and taste), followed by audition (pitch, amplitude, and phrasing of sounds), and ﬁnally by vision (light and optic array). Below are some relevant facts on the emergence of each sensory system, in the order of their functional emergence in pre-natal development. 4.1. Skin and body Tactile and somatic sensitivity is already expressed by 8 weeks gestational age. Eight week-old, externalized fetuses display head movements away from an object touching their lips (Hooker, 1952). Nociceptive (pain) receptors appear ﬁrst in and around the mouth area at around 7 weeks post conception. They rapidly extend to the palmar surface of the hands by 11 weeks, and the rest of the skin and mucosal surfaces by 20 weeks (Brusseau & Mashour, 2007; Smith, 1996).

112

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

4.2. Taste and smell Fetuses are documented swallowing amniotic ﬂuid by 12 weeks of gestation when chemo-receptors already pave the inside of the nose and the oral cavity. Chemo-sensation (sense of taste and smell combined) is evident when a sweet substance (sugar) is injected into the amniotic ﬂuid, swallowing increases. It decreases with the injection of a ‘‘noxious” (iodinated poppy seed) substance (Schaal, Orgeur, & Rognon, 1995). In multiple studies, Marlier and collaborators (Marlier, Schaal, & Soussignan, 1998a, 1998b) demonstrate that newborns manifest not only taste and smell discrimination, but also chemosensory preferences (preferential postural orientation of nose and mouth) that have been learned in the womb (e.g., smell of maternal amniotic ﬂuid), a preference that perseverates days after birth. 4.3. Sounds and voices Fetuses display bodily movements in response to sounds from 22 to 24 months. These movements reﬂect a complex auditory sensitivity to variations of sound’s frequency, intensity, as well as duration (Hepper & Shahidullah, 1994). From 30 weeks gestational age, fetuses display heart rate acceleration at the onset of external airborne sounds they hear through the uterine wall, as well as vibroacoustic stimulations applied against the mother’s abdomen (Kisilevsky & Hains, 2010; Lecanuet, Granier-Deferre, & Busnel, 1988). Based on habituation/dishabituation paradigms with extra-uterine speakers placed close to the mother, fetuses learn from approximately 32 weeks to discriminate structure characteristics of speech sounds. Changes in the order of two syllables composing a word like ‘‘babi” and ‘‘biba” are correlated with heart rate deceleration indexing an orienting response of the fetus (Lecanuet et al., 1987, 1988). The womb is not sound proof, sounds travel through the amniotic ﬂuid with the voice of the mother particularly ampliﬁed. In their niche, fetuses learn and develop preferences for familiar noise conﬁgurations. Based on an operant sucking paradigm, few hour old newborns are shown to discriminate and actively prefer to hear their mother’s voice compared to another female’s uttering the same phrases with the same intensity. This preferential discrimination is based on what infants heard and eventually learned in-Utero regarding their mother’s voice characteristics (DeCasper & Fifer, 1980). 4.4. Light and dark Light experience being greatly limited in the darkness of the womb explains the maturational delay of vision. However, fetal eye activities are recorded from 13 weeks gestational age, starting with downward movements of both eyes (12.5 weeks), slow eye movements (16 weeks), eye closing (20 weeks), rapid eye movements (23 weeks) and conjugate lateral eye movements by 24 weeks (deVries, Visser, & Prechtl, 1985). From approximately 26 weeks, change in heart rate and overall bodily movements are recorded in the fetus when a bright light is ﬂashed on the mother’s abdomen (Hepper, 2002). Compared to audition, which is almost adult-like at birth, vision is noticeably poor in comparison. Babies are born with poor visual acuity and contrast sensitivity, in need of the rich illumination and much larger view outside of the womb to develop. Vision improves dramatically in the weeks following birth and within 6 months becomes close to adult-like, with detection of colors’ full spectrum, sensitivity to movement parallax, contrast sensitivity, convergence and accommodation for objects at far distances (Kellman & Arterberry, 2006). 5. Pre-natal signs of feeling experience By 30 weeks gestational age, fetuses display marked changes in their habituation to acoustic vibrations, coupling of movement to heart rate, as well as some indications that they might begin experiencing pain (Anand & Hickey, 1987). Such changes suggest a budding minimal experience of ‘‘what it is like” or phenomenal (P) consciousness (Block, 2007). We read pleasure and pain in others, perceiving as well as eventually inferring feeling states. We do so primarily by perceiving organized patterns of outward bodily movements (e.g., facial expressions), either directly or linking them to perceived circumstances or events in the environment such as loud sound = startle = fear; needle prick = cry = pain; or tickle = giggle = pleasure. Animals evolved patterns of such recognizable bodily movements (emotions) rendering their feeling states public, accessible to others in communication (Darwin, 1872/1965). There is indeed an inherited equation between particular feeling experience and the expression of speciﬁc bodily patterns. This equation is directly perceived in communication (e.g., threat or fear bodily expressions as signs of aggressive or avoidant feeling states), but can also be inferred and made explicit in conversation at more advanced stages of development (e.g., the typical content of gossips and other folk theories that furnish our social lives). In relation to issue of the self in development and within the proposed framework, the question is when do emotional patterns that are readily perceivable as standing for particular feeling states emerge in epigenesis? What might be the actual origins of emotional patterns that we perceive as standing for particular feeling states like pain, fatigue, or pleasure, all of which could be linked to minimal self-awareness? Newborns do cry when unmistakably in pain and smile when satiated, but what about fetuses?

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

113

New technological progress in real-time sonographic imaging of fetuses (2D-4D ultrasound, Hata, Dai, & Marumo, 2010) allow to study with great precision facial expressions that are universally recognized as standing for basic mental states. Ultrasonic imaging demonstrate that from 30 weeks gestational age, fetuses display highly organized facial expressions that stand unambiguously for what we perceive as experiences of pleasure (smiling), displeasure (scowling), and fatigue (yawning). These patterns of emotional expressions are also readily observable immediately after birth. When newborns fuss before feeding or smile away following a feed, we can assume that they are feeling either pain or intense pleasure, not just ‘‘on or off” hunger sensations that would set off some sort of thermostatic responses. Mental states and affective values most likely motivate newborns’ behaviors, not just sensation as I will try to show next. Newborns do experience the world, including themselves, beyond the raw ‘‘proximal” sensations transducted on the surface of the sense receptors. They are not just stimulus-bound. If clear, unambiguous emotional expressions can be detected in fetuses in the same way that they are readily observable in newborns, the inference of feeling states is not too far fetched, particularly in light of the abundance of data showing the remarkable behavioral continuity of pre-natal and post-natal development (Prechtl, 1984). Recent use of ﬁne grain 4-D ultrasonographic recordings conﬁrm that no movements observed in the fetus during the last trimester of pregnancy (from at least 30 weeks) are not also expressed by neonates (deVries et al., 1985; Hepper, 2002; Prechtl, 1984). These include also, aside of unambiguous smiling and scowling, isolated eye-blinking, hand and ﬁnger movements, tongue protrusion, and all reﬂex responses displayed by newborns with the exception of the Moro reﬂex that requires space for the spreading of arms (Andonotopo & Kurjak, 2006; Andonotopo, Stanojevic, Kurjak, Azumendi, & Carrera, 2004). Furthermore, by 36 weeks fetuses also show four well-deﬁned grouping of behaviors (Sleep-Wake behavioral states) that would correspond to distinct levels of consciousness (levels of ‘‘feeling” experience). These patterns are stable and show clear transitions, respectively: Quiet Sleep (stable heart beat, no eye movements); Active Sleep (eye movements, body movements with heart rate acceleration); Quiet Awake (eye movements, stable heart rate), Active Awake (eye movements, unstable heart rate and bouts of tachycardia) (Hepper, 2002; Nijhuis, Prechtl, Martin, & Bots, 1982; Prechtl, 1977). These patterns of behavioral states are observed in newborns and beyond via applicable electro-encephalographic (EEG) recordings. Starting a trend that will continue beyond birth, the proportion of time fetuses spend in an Active Alert state augments signiﬁcantly during the last 4 weeks of gestation (from 6% to 9% of the time, Hepper, 2002; Prechtl, 1977). This developmental trend continues in the weeks following birth, infants spending an increasing amount of time in an active and awake behavioral ‘‘feeling” state (Wolff, 1987). Finally, sensory evoked potential recorded in infants born premature, 10 week before term, indicates that minimal level of phenomenal consciousness might be present already by 30 weeks of gestational age as thalamo-cortical connections become functional (Klimach & Cooke, 1988), although continuing to develop markedly through adulthood with documented changes in childhood and adolescence (Fair et al., 2010). As suggested by Fair et al., cortical-subcortical interactions must play a role in ‘‘the shift from reﬂexive, stimulus-bound behavior in childhood, to the goal-directed and more ﬂexible functioning found in adulthood” (Fair et al., 2010, p. 2). However, cortico-subcortical interactions remain scarcely mapped in the perspective of fetal and infant development. However, as we will see next, early behavioral signs of such shift clearly exist in infancy, with precursor signs possibly manifested already by 32 weeks gestational age (see discussion above). In this context, noteworthy is the observation that the signiﬁcant increase of full range emotional expressions and facial movements observed in the fetus from 32 weeks of gestation is associated with a decrease of overall movements (Kurjak et al., 2006; Andonotopo et al., 2004; Hata et al., 2010). This growing expressive speciﬁcity is consistent with the normal neurological development of the fetus (Prechtl, 1997). In summary, recent progress in fetal psychology research suggests that there are pre-natal signs of feeling experience. The well organized emotional expressions combined with the remarkable continuity of pre-natal and post-natal development supports the idea that ﬁrst feeling experience, therefore the potential for minimal self-awareness, might emerge 8–10 weeks before birth (30–32 weeks gestational age). Keeping in mind the striking continuity of behaviors observed during the last 10 weeks of gestation and what can be readily observed and tested after birth (Prechtl, 1984), what can be seen in the newborn could stand also for what is not readily testable in the womb, from at least 32 weeks when fetal behaviors show all the aspects of what is observed after birth. I now turn to such demonstration. 6. Newborn feeling experience Infancy research of the past four decades changed our views on the starting state of mental life, namely what is it like to be a newborn. Until then, developmental theorists tended to endorse the view, in their own ways, of an initial state of confusion with the environment, the famous initial ‘‘blooming buzzing confusion” proposed by James (1890). Neonates were presented as stimulus-bound, sensing but not feeling the world, their behavior primitively reduced to ready-made, evolved automatisms (reﬂexes or pulsions). Newborns’ were thought to experience a world that was not yet objectiﬁed or differentiated, subjectivity and objectivity confounded and in need of progressive integration through experience (e.g., cognitive distancing and construction in the case of Piaget (1952, 1955), Ego development in the case of Freud’s pulsion theory (Freud, 1905/2000).

114

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

More recent research shows that, in fact, healthy newborns do perceive the world objectively and are not in a state of subject–object confusion. From birth they express a difference between what pertains to their own body and what pertains to the world ‘‘out there”. Within the proposed framework, infants at birth do not just sense and respond based on reﬂex-like mechanisms. More than thermostats functioning on the basis of on/off close-loop feedback, they also feel and act on objects that they experience as differentiated, distal, and situated in relation to them. This research demonstrates that infants from birth are capable of ‘‘feeling” not just ‘‘sensing” the world, demonstrating the potential for minimal embodied self-awareness. For sake of space, only a few evidence are reviewed here (for more, see Rochat (2001, 2010b)). Although babies are born with poor contrast sensitivity and grating acuity (Banks & Shannon, 1993; Kellman & Banks, 1997), infancy researchers investigating newborn vision demonstrate that despite the obvious developmental lag of the modality, active perceptual processing does take place at birth. For example and relevant to our discussion, using habituation and novelty preference paradigms researchers have established that newborn infants, only a few hours old, when awake and alert, perceive the real (distal) size of objects, not the varying (proximal) sizes projected onto the retina. Newborns perceive size constancy of objects (Granrud, 1987; Slater, Mattock, & Brown, 1990), most likely via visuo-proprioceptive convergence cues from both eyes as they line their gazes and focus onto the distal object (Kellman & Arterberry, 2006). What is particularly relevant to our discussion is that this kind of empirical evidence suggests that newborn infants have feeling experience, and are not just limited to sensing what is recorded at the proximal level of the receptors (i.e., the retina). But what does it say about self-awareness? If feeling experience is required for minimal self-awareness, what kind of evidence is there that newborns actually express such awareness? Research shows that infants from birth are capable of perceiving their own body as an entity among other entities. An entity that has unity, is differentiated, occupies space, and is substantial. In addition, from at least 2 months of age, there is good evidence that infants have a sense of their own agency on objects, aware of themselves as embodied agent in the world (Rochat, 2001). For example, we were able to show that newborn infants do discriminate between self-stimulation and stimulations coming from the outside world, suggesting that they are not in a state of confusion with the world outside. They root (i.e. orient head and mouth) signiﬁcantly more toward the ﬁnger of an experimenter touching their cheek than their own hand spontaneously brought in contact with the peri-oral region of the face (Rochat & Hespos, 1997). We also showed that 2 month-olds are attentive and systematically explore the auditory consequences of their own action while sucking on a sound-producing paciﬁer (Rochat & Striano, 1999). They differentiate between sounds that are perfectly contingent but that are either analog or non-analog to the physical pressures they apply on the paciﬁer. In the context of our research, from 2 months of age (not at birth) infants show clear signs that they perceive themselves as agent of what they hear. Other empirical observations demonstrate further the minimal self-awareness of neonates who seem to experience the world with an implicit differentiated sense of themselves as embodied perceivers. In their behavior, newborns conﬁrm J.J. Gibson’s (1979) idea that perceiving the world is co-perceiving the self. For example, there is some evidence that from birth infants differentiate movements of the own body (ego motions), from movements of objects and things in the world that occur independently of the self (allo motions). Newborns pick up visual information that speciﬁes ego-motion or movements of their own body while they, in fact, remain stationary. These studies indicate that neonates experience an illusion of moving, adjusting their bodily posture according to changes in direction of an optical ﬂow that is presented on TV monitors in the periphery of their visual ﬁeld (Jouen & Gapenne, 1995). This kind of observations point to the fact that from birth, infants are endowed with the perceptual, qua inter-modal capacity to pick up and process meaningfully self-specifying information. Other research indicates that neonates and young infants display an a priori proprioceptive sense of their own body in the way they act and orient to meaningful affordances of the environment (Ball & Tronick, 1971; Carroll & Gibson, 1981; E.J. Gibson, 1995; Van der Meer & Lee, 1995) as well as in the way they detect visual information that speciﬁes ego motion, adjusting their posture appropriately in direction and amplitude to compensate for surreptitious changes in gravitational forces (Butterworth, 1992; Butterworth & Hicks, 1977). In summary, newborns demonstrate minimal, implicit self-awareness in relation to physical objects, what Neisser (1988, 1991) ﬁrst coined as the ecological self. Empirical research indicate that infants are born with a sense of themselves as differentiated and situated entities among other entities in the world they perceive as distal and distinct from the feeling experience of their embodied self. By 2 months, correlated with a sudden increase in the proportion of time spent awake and alert (Wolff, 1987), infants also manifest a sense of their own agency on both physical objects and people, what I view as indexing the ‘‘two month-revolution” (Rochat, 2001). This important transition has been identiﬁed and correlated with brain maturation, in particular the transition from the dominant control of behaviors by subcortical systems, to higher order cortical systems (McGraw, 1942). A major index of the two-month revolution is the emergence of a more contemplative stance taken by infants in their attention to events and things, together with (and not the least), the emergence of socially elicited smiling in relation to people (Wolff, 1987; Rochat, 2001). However, the implicit minimal self-awareness changes rapidly, from the time it is expressed by neonates, and possibly even older fetuses during the last 10–8 weeks of gestation. It is supplemented with an awareness of self that becomes explicit and conceptual. We now turn to this development.

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

115

7. Emerging idea of Me From around 18 months, parallel to the emergence of grammatically articulated language as well as the explosion of a children’s vocabulary (including personal pronouns and adjective like ‘‘Me” and ‘‘Mine!” (Bates, 1990; Tomasello, 1998), children start to express much more than implicit (minimal) self-awareness. They become explicitly self-conscious. They are newly capable of re-cognizing themselves for themselves, inclined to work on their self-presentation with others in mind, hence begin to manifest radically novel feeling experiences that are determined by what children now perceive and construe of the evaluative eyes of others onto themselves (Rochat, 2009). They show ﬁrst explicit signs of self-conscious emotions such as embarrassment, but also shame, pride, contempt, guilt, even hubris and contempt (Kagan, 1981; Lewis, 1992). From this point on, minimal self-awareness is not simply replaced and does not disappear as a kind of feeling experience. It remains implied in perceptions and actions that are not under explicit conscious control, which is the case for most perceptions and acts that are ritualized, automatic, and routinely performed all through the lifetime. However, from the middle of the second year, minimal self-awareness is supplemented, and often in competition with a self-awareness that is explicit and conceptual, what James (1890) referred to as the distinct idea of Me that he contrasted with the implicit sense of I, here referred as minimal self-awareness. From the point of view of brain growth, there is a developmental synchrony between emerging meta-cognitive abilities around 2–3 years of age, potentially turned toward the self, and the documented ontogenetic maturation of the rostrolateral region of the prefrontal cortex. The growth of this region would correlate with the development of new levels of consciousness, in particular the transition from minimal to meta-cognitive levels of self-consciousness (Bunge & Zelazo, 2006; Zelazo, Hong Gao, & Todd, 2007). Likewise, and in a related fashion, same prefrontal regions of the cortex that come on line in development are correlated with progress in the cognitive control underlying ‘‘executive function”: the ability to pause and reﬂect before a decision to act, the capacity inhibit ﬁrst impulse for action. Similar models exist in the animal literature in relation to the search of hidden objects by rhesus monkeys and young infants in the context of object permanence and the famous A-Not-B error ﬁrst described by Piaget (Diamond & Goldman-Rakic, 1989; Piaget, 1936/1952). These prefrontal cortex regions are known to develop steadily, but at different rates, coming chronologically on-line through childhood (see Gogtay et al., 2004). Each of these prefrontal cortex regions would be linked to particular levels of cognitive control achieved by the child (Zelazo, 2004; Zelazo et al., 2007). Bunge and Zelazo (2006) distinguish four types of rules in a sorting card game they use to test children (from simple stimulus-reward to complex higher order ‘‘meta” rules), indexing various levels of cognitive control children achieve in early development. These levels of cognitive control would, for these authors, correspond to levels of self-awareness as they are directly linked to children’s executive functioning when for example they try to resolve a problem or anticipate events. Children would develop self-consciousness and recursive consciousness by ‘‘the iterative reprocessing of the contents of consciousness via thalamocortical circuits involving regions of prefrontal cortex” (Zelazo et al., 2007, p. 224). Each reprocessing of the content of consciousness, starting with minimal consciousness and self-consciousness at birth, would require the recruitment and ‘‘excitability” of yet another region of the prefrontal cortex (Rochat, 2010a). Four cortical regions are identiﬁed as maturing in succession by Zelazo et al. (2007): the orbitofrontal, ventrolateral, dorsolateral, and the rostrolateral regions of the prefrontal cortex. Based on both developmental neuroscience (EEG, PET), animal models, and neurological case studies, each of these regions would control for particular levels of executive functioning and rule use, extended to the development of self-awareness: from simple to more complex, eventually reﬂective and evaluative self-awareness, the latter particularly linked to the maturation of the rostrolateral region of the prefrontal cortex (Bunge, 2004; Bunge & Zelazo, 2006). The development of self-consciousness and bodily awareness, like the development of the ability to use rules to inhibit inappropriate behaviors at higher levels of complexity, would ‘‘mirror the protracted developmental course of the prefrontal cortex” (Zelazo et al., 2007, p. 412). An important aspect of the proposed brain-based model of developing self-consciousness is that such development starts off with the innate prescription of a minimal level consciousness. In relation to the own body and self-consciousness in general, such development does not start from scratch, but rather rests on the primary requirement of a minimal experiential awareness of the embodied self that would be already signiﬁed prenatally (Rochat, 2010a). At a behavioral level, the mirror mark test (self-directed behaviors toward a mark surreptitiously put on the face and discovered in the mirror) established some 40 years ago by (Amsterdam, 1968, 1972; and Gallup, 1970) continue to be viewed as the ‘‘acid acid test” of self-consciousness in both the developmental and the comparative literature (see Gallup, 1970; Lewis, 1995). In general, the rationale is that self-directed behavior toward the mark would presuppose some form of explicit and conceptualized self-awareness (but see the recent critic of such purely cognitive view in Rochat and Zahavi (2010)). A majority of children are documented passing the mirror mark test from around 21 months (Amsterdam, 1972; Lewis & Brooks-Gunn, 1979), although recent ﬁndings suggest that it might greatly vary depending on the cultural and developmental niche of the child (Broesch, Callaghan, Henrich, & Rochat, 2010). In her pioneer study of children’s reaction to the mirror in the context of the mark test, Amsterdam (1968, 1972) describes four developmental phases emerging in succession between 3 and 24 months. In a ﬁrst period extending from 3 to 11 months, children treat their own image as a playmate, expressing

116

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

mainly social invitations toward the specular image. In a second period (11–12 months) children start to explore the mirror proper, its surface and texture, often searching behind it. By 13 months, children begin to show marked increase in withdrawal behaviors, avoiding looking at the mirror, hiding from it and even sometime crying. From 14 months and peeking by 20 months, children show unmistakable signs of embarrassment and coy glances toward the specular image. These observations point to the complex cognitive and affective aspects of children’s developing reactions to their own mirror reﬂection (Amsterdam & Greenberg, 1977; Amsterdam & Levitt, 1980). In particular, the generalized embarrassment children express by either by hiding from or clowning in front of the mirror (two opposite forms of explicit self-consciousness) demonstrate the growing social connotation of mirror self-experience: the fear or weariness of being ultimately judged (Rochat & Zahavi, 2010). It appears that the evaluative gaze of others begin to weigh heavily on the child’s mind, indexing genuine emergence of self-consciousness, in the literal English sense of being aware and weary of how others perceive and represent the self. From the middle of the second year, children begin to show signs that they care about reputation, a uniquely exacerbated human trait and a major determinant of self-awareness beyond infancy and through the lifetime (Rochat, 2009). The social dimension of self-consciousness is unmistakable in the fact that as children begin to show explicit self-concept using personal pronouns, engaging in pretense, covering up, putting up faces, or working on self-presentation, they also display a growing sense of rules and norms, the way things ‘‘ought to be” (Kagan, 1981). By the time children begin to recognize themselves in mirrors (21 months), they also tend to assert their own territory and possession over things, stating that ‘‘it’s Mine!”, implying ‘‘. . . it’s not yours!” (Rochat, in press). This development is signiﬁcant and marks what can be viewed as an ultimate step: an awareness of self that is situated in a moral space made of shared values. 8. Emergence of moral self-awareness by 3–5 years Charles Taylor (1989) makes the point that developing self-awareness is ultimately becoming conscientious and mindful of self in relation to others: ‘‘What we are constantly losing from sight (. . .) is that being a self is inseparable from existing in a space of moral issues, to do with identity and how one ought to be. It is being able to ﬁnd one’s standpoint in this space, being able to occupy a perspective in it” (Taylor, 1989, p. 112). This aspect might be an ultimate achievement and a primary goal of children’s socialization that constrains them to share, reciprocate, and to tame a natural inclination toward selﬁshness. From 5 years of age, children start to factor the perspectives of others on the world, their mind states, and motives, allowing them to predict for example whether they hold correct or false beliefs (Wellman & Liu, 2004). This development is universal (Callaghan et al., 2005) and continues beyond 5 years of age with increased expression of inequity aversion (Fehr, Bernhard, & Rockenbach, 2008) as well as more complex considerations of what constitutes equity and relative sacriﬁce in sharing (McCrink, Bloom, & Santos, 2009). Parallel to the emergence of explicit theories of mind between 3 and 5 years, children develop an ethical stance and become ‘‘principled”, ready to sacriﬁce some of their own resources to make a ‘‘moral” point (Robbins & Rochat, 2010). The ethical stance taken by children from 5 years of age indexes a new kind of self-awareness that emerges from the ethical feeling experience of what feels right and what feels wrong relative to others. Moral self-awareness and the taking of an ethical stance by which one acts ‘‘principled” depends on some control over selﬁsh propensities: the propensity to maximize gains for self and to be primarily centered on one’s own motives and perspectives. In relation to the brain, such development is likely to be linked to the growing ability in executive functioning, in particular the ability to inhibit basic ‘‘selﬁsh” propensities and immediate self-gratiﬁcation to consider the motives, mind states, and perspectives of others, a ‘‘mentalizing” ability that is lacking in autism (Frith & Frith, 2003). In the context of theories of mind and false belief understanding, research indicates that such executive functioning and mentalizing process are linked to prefrontal regions of the brain that would presumably come on-line at around 5 years of age in the healthy child (Gallagher & Frith, 2003). Such regions include the medial prefrontal cortex, once again a late maturing brain system linked to higher order processing of neural signals (Gallagher & Frith, 2003; but see also Saxe & Kanwisher, 2003; Saxe, Moan, Scholz, & Gabrieli, 2006). 9. Summary and conclusions Infancy and fetal psychology research supports the idea that minimal self-awareness is deeply rooted in epigenesis, possibly manifested already by 30–32 weeks gestational age. I proposed that such embodied awareness co-emerge with the ability of an organism to have ‘‘feeling” experience. Research points to the fact that newborns, and possibly older fetuses are more than automata. They do not just sense the world by responding to proximal stimulation hitting their receptors. Aside from reﬂexes, they perceive and act toward distal objects and events that are not confounded with their own subjective experience. Self-world differentiation is indeed an early fact of life. Such evidence needs to be considered in light of the explosive brain growth that takes place from the embryo stage of child development, as part of a powerful epigenetic momentum. Self-awareness needs to be construed as the phenotypic expression of such momentum, exacerbated and protracted in humans beyond birth (Konner, 2010). Minimal awareness

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

117

of the body as a differentiated, situated, and substantial entity among other entities (implicit ecological self) expressed at birth is rapidly supplemented with an implicit awareness of the self as agent in the world. By 2 months, infants explore systematically the perceptual consequences of their own actions. They adopt a contemplative stance toward objects and people. A cardinal index of this new stance is the emergence of socially elicited smiling, aside from signiﬁcantly more time spent in an alert and awake state. This ‘‘two-month revolution” has been linked to the maturation of thalamo-cortical connections, allowing greater cortical control of behaviors. By 18 months, self-awareness develops to become conceptual in addition to being perceptual. Children start to demonstrate explicit recognition and signs of embarrassment (self-consciousness) when viewing themselves in mirrors, in parallel to ﬁrst uses of personal pronouns and adjectives like ‘‘Me” and ‘‘Mine”. Increasingly, they appear to behave with others in mind, construing how others might represent and evaluate them. Self-consciousness is associated with the maturation and the coming on-line of various regions of the prefrontal cortex that continues to mature up to 5 years of age and probably beyond. As children become self-conscious, they begin to care about reputation and demonstrate growing concerns regarding norms, rules, and regulations. As part of such growing concerns, children become also proactive in asserting their own entitlement to things (i.e., claimed possessions). This creates a rich soil for the ultimate development of the moral self-awareness emerging between 3 and 5 years as children show ﬁrst signs of taking an ethical stance. They become conscientious of others in addition to being self-conscious and self-assertive. They show ﬁrst signs of becoming ‘‘autonomous or free” moral agents. From then on, children start the lifelong quest of ﬁnding a perspective in a space of moral values that is shared and recognized by others. In development, the moral self is an ultimate expression of the interaction between genes and the environment, an expression demanded by all human cultures. So, what is a self and what qualiﬁes for self-awareness? I tried to show that it is helpful to consider these perennial questions in the perspective of early, even pre-natal development as organisms grow from just sensing to actually feeling the world. Feeling experience would give any organism the potential for being minimally self-aware, something that appears to be readily demonstrated by neonates. The question of how minimal self-awareness (‘‘I”) relates to later developing conceptual and moral self (‘‘Me”) remains an open question that should guide future research. What is certain is that much can be gained from considering how brain, body, and environment interact to produce the self as phenotype. Acknowledgments Gratitude is expressed (by alphabetical order) to Michel Heller, Claudia Passos-Ferreira, and Erin Robbins for their constructive criticisms and feedback on an earlier version of this article. References Amsterdam, B. K. (1968). Mirror behavior in children under two years of age doctoral dissertation. Univ. North Carolina. Order no. 6901569; University Microﬁlms, Ann Arbor, MI. 48106. Amsterdam, B. (1972). Mirror self-image reactions before age two. Developmental Psychobiology, 5, 297–305. Amsterdam, B. K., & Greenberg, L. G. (1977). Self-conscious behavior of infants. Developmental Psychobiology, 10, 1–6. Amsterdam, B. K., & Levitt, M. (1980). Consciousness of self and painful self-consciousness. Psychoanalytic Study of the Child, 35, 67–83. Anand, K. J. S., & Hickey, P. R. (1987). Pain and its effects in the human neonate and fetus. New England Journal of Medicine, 317, 1321–1329. Andonotopo, W., & Kurjak, A. (2006). The assessment of fetal behavior of growth restricted fetuses by 4D sonography. Journal of Perinatal Medicine, 34, 471–478. Andonotopo, W., Stanojevic, M., Kurjak, A., Azumendi, G., & Carrera, J. M. (2004). Assessment of fetal behavior and general movements by four-dimensional sonography. Ultrasound Review of Obstetrics and Gynecology, 4, 103–114. Ball, W., & Tronick, E. (1971). Infant responses to impending collision: Optical and real. Science, 171, 818–820. Banks, M. S., & Shannon, E. S. (1993). Spatial and chromatic visual efﬁciency in human neonates. In C. E. Granrud (Ed.), Carnegie-Mellon symposium on cognitive psychology (pp. 1–46). Hillsdale, NJ: Erlbaum. Bates, E. (1990). Language about me and you: Pronominal reference and the emerging concept of self. In D. Cicchetti & M. Beeghly (Eds.), The self in transition: Infancy to childhood (pp. 165–182). Chicago: University of Chicago Press. Block, N. (2007). Consciousness, accessibility, and the mesh between psychology and neuroscience. Brain and Behavioral Sciences, 30, 481–548. Broesch, T., Callaghan, T., Henrich, J., & Rochat, P. (2010). Cultural variations in children’s mirror self-recognition. Journal of Cross-Cultural Psychology, in press. Brusseau, R. R., & Mashour, G. A. (2007). Subcortical consciousness: Implications for fetal anesthesia and analgesia. Behavioral and Brain Sciences, 30(1), 86–87. Bunge, S. A. (2004). How we use rules to select actions: A review of evidence from cognitive neuroscience. Cognitive, Affective, and Behavioral Neuroscience, 4, 564–579. Bunge, S. A., & Zelazo, P. D. (2006). A brainbased account of the development of rule use in childhood. Current Directions in Psychological Science, 15, 118–121. Butterworth, G. (1992). Origins of self-perception in infancy. Psychological Inquiry, 3(2), 103–111. Butterworth, G., & Hicks, L. (1977). Visual proprioception and postural stability in infancy: A developmental study. Perception, 6, 255–262. Callaghan, T., Rochat, P., Lillard, A., Claux, M. L., Odden, H., Itakura, S., et al (2005). Synchrony in the onset of mental-state reasoning: Evidence from ﬁve cultures. Psychological Science, 16(5), 378–384. Carlson, B. M. (1994). Human embryology and developmental biology. St. Louis: Mosby. Carroll, J. J., & Gibson, E. J. (1981). Differentiation of an aperture from an obstacle under conditions of motion by 3-month-old infants. In Paper presented at the meetings of the society for research in child development, Boston, MA. Darwin, C. (1872/1965). The expression of the emotions in man and animals. Chicago: Chicago University Press. DeCasper, A. J., & Fifer, W. P. (1980). Of human bonding: Newborns prefer their mother’s voices. Science, 208, 1174–1176. Dennett, D. C. (1981). Brainstorms: Philosophical essays on mind and psychology. Cambridge, MA: M.I.T. Press Bradford Books Series. deVries, J. I. P., Visser, G. H. A., & Prechtl, H. F. R. (1985). The emergence of fetal behavior: II. Quantitative aspects. Early Human Development, 12, 99–120. Diamond, A., & Goldman-Rakic, P. S. (1989). Comparison of human infants and rhesus monkeys on Piaget’s A not B task: Evidence for dependence on dorsolateral prefrontal cortex. Experimental Brain Research, 74, 271–294.

118

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

Fair, D. A., Bathula, D., Mills, K. L., Costa Dias, T. G., Blythe, M. S., Zhang, D., et al (2010). Maturing thalamocortical functional connectivity across development. Frontiers in Systems Neuroscience, 4(10), 1–10. Fehr, E., Bernhard, H., & Rockenbach, B. (2008). Egalitarianism in young children. Nature, 454, 1079–1084. Freud, S. (1905/2000). Three essays on the theory of sexuality. NY: Basic Books Classics series. Frith, U., & Frith, C. D. (2003). Development and neurophysiology of mentalizing. Philosophical Transactions of the Royal Society of London Series B – Biological Sciences, 358(1431), 459–473. Gallagher, H. L., & Frith, C. D. (2003). Functional imaging of ‘theory of mind’. Trends in Cognitive Sciences, 7(2), 77–83. Gallup, G. G. (1970). Chimpanzees: Self-recognition. Science, 167, 86–87. Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton Mifﬂin. Gibson, E. J. (1995). Are we automata? In P. Rochat (Ed.), The self in infancy: Theory and research (pp. 1–23). Amsterdam: North-Holland/Elsevier Science. Gogtay, N., Giedd, J. N., Lusk, L., Hayashi, K. M., Greenstein, D., Vaituzis, A. C., et al (2004). Dynamic mapping of human cortical development during childhood through early adulthood. Proceedings of the National Academy of Sciences USA, 101, 8174–8179. Gottlieb, G. (1971). Ontogenesis of sensory functions in birds and mammals. In A. Tobach, L. R. Aronson, & E. Shaw (Eds.), The biopsychology of development (pp. 67–128). NY: Academic Press. Granrud, C. E. (1987). Size constancy in newborn human infants. Investigative Opthalmology and Visual Science, 28(Suppl.), 5. Hata, T., Dai, S. Y., & Marumo, G. (2010). Ultrasound for evaluation of fetal neurobehavioural development: From 2-D to 4-D ultrasound. Infant and Child Development, 19(1), 99–118. Special issue: Towards a fetal psychology. Hepper, P. G. (2002). Prenatal development. In A. Slater & M. Lewis (Eds.), Introduction to infant development. NY: Oxford University Press. Hepper, P. G., & Shahidullah, S. (1994). Development of fetal hearing. Fetal and Maternal Medicine Review, 6, 167–179. Hooker, D. (1952). The prenatal origin of behavior. Kansas: University of Kansas Press. Huttenlocher, P. R., & Dabholkar, A. S. (1997). Regional differences in synaptogenesis in human cerebral cortex. Journal of Comparative Neurology, 387(2), 167–178. James, W. (1890). The principles of psychology. NY: Henry Holt & Company. Jouen, F., & Gapenne, O. (1995). Interactions between the vestibular and visual systems in the neonate. In P. Rochat (Ed.), The self in infancy: Theory and research (pp. 277–302). Amsterdam: North-Holland/Elsevier Science. Kagan, J. (1981). The second year: The emergence of self-awareness. Harvard University Press. Kellman, P. J., & Arterberry, M. E. (2006). Infant visual perception. In D. Kuhn, R. S. Siegler, W. Damon, & R. M. Lerner (Eds.), Handbook of child psychology (6th ed.. Cognition, perception, and language (Vol. 2, pp. 109–160). Hoboken, NJ: Wiley. Kellman, P., & Banks, M. S. (1997). Infant visual perception. In R. Siegler & D. Kuhn (Eds.), Handbook of child development (5th ed.. Cognition, perception, and language (Vol. 2, pp. 103–146). New York: Wiley. Kisilevsky, B. S., & Hains, S. M. J. (2010). Exploring the relationship between fetal heart rate and cognition. Infant and Child Development, 19(1), 60–75. Special issue: Towards a fetal psychology. Klimach, V. J., & Cooke, R. W. (1988). Maturation of the neonatal somatosensory evoked response in preterm infants. Developmental Medicine and Child Neurology, 30(2), 208–214. Konner, M. (2010). The evolution of childhood. Cambridge, MA: Harvard University Press. Kretschmann, H. J., Kammradt, G., Krauthausen, I., Sauer, B., & Wingert, F. (1986). Brain growth in man. Bibliotheca Anatomica, 28, 1–26. Kurjak, A., Andonotopo, W., Hafner, T., Salihagic-Kadic, A., Stanojevic, M., Azumendi, G., Ahmed, B., Carrera, J. M., & Troyano, J. M. (2006). Normal standards for fetal neurobehavioral developments – longitudinal quantiﬁcation by four-dimensional sonography. Journal of Perinatal Medicine, 34, 56–65. Lecanuet, J.-P., Granier-Deferre, C., & Busnel, M.-C. (1988). Fetal cardiac and motor responses to octave-band noises as a function of central frequency, intensity and heart rate variability. Early Human Development, 18, 81–93. Lecanuet, J.-P., Granier-Deferre, C., DeCasper, A. J., Maugeais, R., Andrieu, A.-J., & Busnel, M.-C. (1987). Perception et discrimination foetales de stimuli langagiers, mise en évidence à partir de la réactivité cardiaque, résultants préliminaires. Compte-Rendus de l’Académie des Sciences, Paris (III), 305, 161–164. Lecanuet, J. P., & Schaal, B. (1996). Fetal sensory competencies. European Journal of Obstetrics & Gynecology and Reproductive Biology, 68, 1–23. Lewis, M. (1992). Shame: The exposed self. New York: Free Press. Lewis, M. (1995). Aspects of the self: From systems to ideas. In P. Rochat (Ed.), The self in infancy: Theory and research (pp. 95–116). Amsterdam: NorthHolland, Elsevier Publishers. Lewis, M., & Brooks-Gunn, J. (1979). Social cognition and the acquisition of self. New York: Plenum Press. Marlier, L., Schaal, B., & Soussignan, R. (1998). Neonatal responsiveness to the odor of amniotic and lacteal ﬂuids: A test of perinatal chemosensory continuity. Child Development, 69(3), 611–623. Marlier, L., Schaal, B., & Soussignan, R. (1998b). Bottle-fed neonates prefer an odor experienced in utero to an odor experienced postnatally in the feeding context. Developmental Psychobiology, 33, 133–145. McCrink, K., Bloom, P., & Santos, L. R. (2009). Children’s and adults judgments of equitable resource distributions. Developmental Science, 13, 37–45. McGraw, M. B. (1942). The neuromuscular maturation of the human infant. NY: Columbia University Press. Merskey, H., & Bogduk, N. (Eds.). (1994). Classiﬁcation of chronic pain: Descriptions of chronic pain syndromes and deﬁnitions of pain terms (pp. 209–214). NY: IASP Press. Merker, B. (2007). Consciousness without a cerebral cortex: A challenge for neuroscience and medicine. Behavioral and Brain Sciences, 30, 63–134. Neisser, U. (1988). Five kinds of self-knowledge. Philosophical Psychology, 1, 35–59. Neisser, U. (1991). Two perceptually given aspects of the self and their development. Developmental Review, 11(3), 197–209. Nijhuis, J. G., Prechtl, H. F. R., Martin, C. B., & Bots, R. (1982). Are there behavioral states in the human fetus? Early Human Development, 6, 177–195. Oppenheim, R. W. (1991). Cell death during development of the nervous system. Annual Review of Neuroscience, 14, 453–501. Piaget, J. (1936/1952). The origins of intelligence in children. New York: International Universities Press. Piaget, J. (1938/1955). The construction of reality in the child. New York: Routledge and Kegan Paul. Prechtl, H. F. R. (1977). Assessment and signiﬁcance of behavioural states. In S. R. Berenberg (Ed.), Brain, fetal and infant: Current research on normal and abnormal development (pp. 79–90). Den Haag: Martinus Nijhoff. Prechtl, H. F. R. (Ed.). (1984). Continuity of neural functions: From prenatal to postnatal life. Oxford: Blackwell Scientiﬁc Publications. Prechtl, H. F. (1997). Is neurological assessment of the fetus possible? European Journal of Obstetrics and Gynecology and Reproductive Biology, 75, 81–84. Rakic, P. (1972). Model of cell migration to the superﬁcial layers of the fetal monkey neocortex. Journal of Comparative Neurology, 145, 61–84. Robbins, E., & Rochat, P. (2010). Emerging signs of strong reciprocity in human ontogeny, submitted for publication. Rochat, P. (2001). The infant’s world. Cambridge: Harvard University Press. Rochat, P. (2009). Others in mind: Social origins of self-consciousness. NY: Cambridge University Press. Rochat, P. (2010a). The innate sense of the body develops to become a public affair by 2–3 years. Neuropsychologia, 48(3), 738–745. Rochat, P. (2010b). Emerging self-concept. In J. G. Bremner & T. D. Wachs (Eds.), Blackwell handbook of infant development (2nd ed.). London: Blackwell Publishers. Rochat, P. (in press). Possession and morality in early development. New Directions in Child and Adolescent Development. Rochat, P., & Zahavi, D. (2010). The uncanny mirror: A re-framing of mirror self-experience. Consciousness & Cognition, in press. Rochat, P., & Hespos, S. J. (1997). Differential rooting response by neonates: Evidence of an early sense of self. Early Development and Parenting, 6(3-4), 105–112. Rochat, P., & Striano (1999). Emerging self-exploration by 2 month-old infants. Developmental Science, 2(2), 206–218.

P. Rochat / Consciousness and Cognition 20 (2011) 109–119

119

Saxe, R., & Kanwisher, N. (2003). People thinking about thinking people: The role of the temporo-parietal junction in ‘theory of mind’. NeuroImage, 19, 1835–1842. Saxe, R., Moan, 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(3), 229–234. Schaal, B., Orgeur, P., & Rognon, C. (1995). Odor sensing in the human fetus: Anatomical, functional, and chemio-ecological bases. In J.-P. Lecanuet, W. P. Fifer, N. A. Krasnegor, & W. P. Smotherman (Eds.), Fetal development: A psychobiological perspective (pp. 205–237). Hillsdale, NJ: Lawrence Erlbaum Associates. Slater, A., Mattock, A., & Brown, E. (1990). Size constancy at birth: Newborn infants’ responses to retinal and real size. Journal of Experimental Child Psychology, 49, 314–322. Smith, S. (1996). Commission of inquiry into fetal sentience. CARE. Taylor, C. (1989). Sources of the self: The making of modern identity. Cambridge: Harvard University Press. Tomasello, M. (1998). One child early talk about possession. In J. Newman (Ed.), The linguistic of giving. Amsterdam: John Benjamins. Van der Meer, A., & Lee, D. (1995). The functional signiﬁcance of arm movements in neonates. Science, 267, 693–695. Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75(2), 523–541. Wolff, P. H. (1987). The development of behavioral states and the expression of emotions in early infancy. Chicago: University of Chicago Press. Zelazo, P. D. (2004). The development of conscious control in childhood. Trends in Cognitive Sciences, 8, 12–17. Zelazo, P. D., Hong Gao, H., & Todd, R. (2007). The development of consciousness. In Cambridge handbook of consciousness. NY: Cambridge University Press.

Consciousness and Cognition 20 (2011) 120–129

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The origins and uses of self-awarenesss or the mental representation of me q Michael Lewis ⇑ Institute for the Study of Child Development, UMDNJ-Robert Wood Johnson Medical School, 97 Paterson Street, New Brunswick, NJ 08903, United States

a r t i c l e

i n f o

Article history: Available online 14 December 2010 Keywords: Self-recognition Machinery of self Mirror recognition Pretend play Consciousness Knowledge of the self

a b s t r a c t This paper explores the meaning and the development of consciousness in the human child. The idea of a self is made up of at least two major aspects. These can be referred to as the machinery of the self and the mental state of the idea of ‘‘me’’. The machinery of the self involves all unconscious, unreferenced action of the body, including its physiology and its processing of information that in turn includes cognitions and emotional states, which are unavailable to consciousness. The mental state or the idea of ‘‘me’’ is that part of the self that makes reference to itself. This mental state develops over the ﬁrst 2 years of life and is a function of both brain maturation processes as well as socialization. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction In order to discuss the origins of self-awareness or the representation of me, which is a developmental question, I must divert brieﬂy to try and place my theory within what I will call a brain-culture framework. This requires that we ﬁrst discuss what is and what is not a self. I do this to emphasize that the confusion of the meaning of self hampers our understanding of its development. The following four questions are essential to understand our notion of self. For the sake of clarity, selfawareness is part of what I mean by self, in particular it is a mental representation of me. What is a self? (1) Here I wish to point out that part of the self is awareness of itself, some call this implicit consciousness while other have considered that part of the self that is unaware, implicit consciousness as well as unconsciousness. If T-cells are capable of some level of intention and go after foreign proteins and kill them, and are also capable of selfversus-nonself recognition, do these cells have a self? Infants soon after birth engage in what some have called nonverbal communication, others have called social reﬂexive behavior, and still others, intersubjectivity. Do these infants have a self? Even more important, from the perspective of human ontogeny, do both the newborn infant who imitates its mother or the 20 month old child who says ‘‘me drink bottle’’ while reaching for the bottle have selves? If they do – a claim made by many – then what develops? The problem is solved by considering that the term self is inadequate to the task. A self is made up of many features (e.g., Lewis, 1979, 1992a, 1992b; Neisser, 1988; Stern, 1985). For example, all living creatures have selfregulating and self-organizing capabilities, and all creatures that have social lives have the ability to interact with regard to other conspeciﬁcs. Whether or not we call these processes ‘‘self’’ does not alter the fact that human and nonhumans share these features.

q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap.

⇑ Fax: +1 732 235 6189.

E-mail address: [email protected] 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.11.002

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

121

What we might not share with most other creatures is our cognitive ability to reﬂect on ourselves (self-awareness) and a culture which provides the content of that awareness. It is the aspect of self which can be called self-awareness, the mental representation of ‘‘me’’. We also are not likely to share with other living creatures the ability to have a symbolic-linguistic understanding of recursive propositions such as ‘‘I know, that you know, that I know how to chop wood.’’ The role of language and conversation and the role of play and pretend play do bear on the feature of self described in this particular way. It is clear that the ﬁeld remains plagued by deﬁnitional vagrancies for the different aspects of a self, self-awareness or a mental representation of me. This representation is different from all other representations since the subject and object is the same. There is one problem in particular, namely, that even the most basic machine aspects of the self are often discussed in ways that assume that they exhibit mental representations of me, that is, in fact, found only in toddlers and older humans. This is a serious problem in the study of development which needs to be addressed and deﬁnitional standards set. Our language and our quest for meaning are likely to confuse us. Consider for example, newborn imitation. We call the tongue protrusion of a newborn to the tongue protrusion of an adult, imitation. We also call imitation when the 6 year old says, ‘‘I want the same sweater as my friend Sarah’’. Piaget (1952), who also observed early ‘‘imitation’’, suggested that we need to distinguish between this early imitation, which is a reﬂective behavior and later imitation that has intention. He argued, as I have over 40 years ago (Lewis, 1967), that the same behavior may serve very different processes and different processes, the same behavior. The developmental literature is full of such examples of the same behavior served by different processes (see for example Kagan, 2008). Another problem is that humans have the capacity and may be even need to attribute humanness to animate and inanimate things. This anthropomorphizing has to be considered when we talk about infants and the origins of self-awareness. For example, humans have attributed human qualities to inanimate objects like trees, rivers, and rocks. Even the wind has been given human attributes. Moreover, as we are meaning seekers we ﬁnd in clouds and mountains, recognizable shapes and creatures which we name. The star constellations are still another example of this human tendency. Certainly we do it with domestic animals like our pets, dogs, cats, birds, and even goldﬁsh. We also are likely doing it with infants. Given our adult knowledge of situations and the emotions these situations invoke in us, we readily ﬁnd these emotions in our infants. As this knowledge is in part culturally given, different cultures should produce different parental beliefs in regard to their infants’ emotions. For example, research on cultural differences in parental beliefs shows that in American culture, parents believe that infants show more and earlier expressions of anger than do Japanese parents. Our tendency for meaning seeking and anthropomorphizing is an important feature in the human child’s development. The meaning the parents give affects the meaning the child learns. If I ﬁnd more anger in my child than another parent, my child is likely to be more angry later in life because my attributions are conveyed to the child in many ways, including our verbal labeling of the infants emotion (Lewis & Michalson, 1983). What we attribute to them is, in large part, what the child will learn and include such cultural aspects as emotional labels, expression interactions, as well as rules, goals, and standards. I tried in several recent pieces to deﬁne what I considered a self to be and have argued that self-awareness or the mental representation of me is but one aspect of our selves (Lewis, 1997, 2003, 2010). This mental state is ﬂexible in that it can be evoked or not. Indeed it is adaptive sometime to have self-awareness (Mandler, 1975) and sometime not (Wegner, 2009). It is not useful to think about me when I am trying to bring my car swiftly to a stop on a highway after a tire blow-out. My attention is drawn to the tasks of looking at other cars, the feel of the wheel, and the sounds of the tire. It is only after bringing my car to a stop by the side of the road that I can focus on myself. To do so earlier is non-adaptive. For this discussion, self-awareness, an aspect of consciousness will be the focus. (2) How do we measure a self? Of course, measurement must follow from theory. If that aspect of self we wish to observe is related to self-other differentiation, self-regulation, social reﬂexive behavior (called by some intersubjectivity) or self-recognition in mirrors, the way we choose to measure the self will differ. It would be foolish to confuse measurement of one feature of self with another, and to make the claim that one measure reﬂects all aspect of self. Imitation, perceptual-motor organization, language, role taking or self directed mirror behavior, are all methods that should inform us about various aspects of the self. Measurement issues, as always, must be tied to the particular construct of the self studied. Since we are interested in self-awareness, it is that aspect which will be discussed. Brieﬂy, the mental representation of me is measured by self-recognition in mirrors, personal pronoun use such as me or mine, and pretend play that involves another. (3) Are there cultural and historical differences in a self? The issue of cultural differences in the concept of self, certainly those features requiring self-awareness, varies markedly (e.g., Lewis, 1992a, discussion of the cultural and historical changes in this aspect of the self). On the other hand, one might wish to argue for some universal features of a self. For example, self-other discrimination, self-regulating, and self-other interactions observed in the very young infant are likely to be found across cultures and historical time. Moreover, other features of the self also may be pan-cultural. If, as some believe, the left temporal lobe, is necessary for a mental representation of me, then the development that occurs in this brain area in the middle of the second year of life may also exist across cultures (Carmody & Lewis, 2010; Lewis & Carmody, 2008). Its expression, however, may be quite different depending on cultural speciﬁcs. The ‘‘terrible two’s’’, described in our culture as part of the emergence of the toddler’s will independent of the parent’s wishes, may not be as marked in cultures less interested in individual autonomy and more interested in a we-self culture (see Geertz, 1984). William James suggested over 100 years ago that it is probably wise to separate features of structure from features of content. The problem of cross cultural differences is likely best approached by arguing that some structures are pan cultural but

122

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

that most content is culturally determined. There is no reason to assume that a structural feature of the self (e.g., the emergence of self-awareness) is much affected by cultural or by caregiving techniques. Except for severely bad treatment, children are likely to develop this structural feature. What is affected, however, is how children come to view themselves. Poor or violent care is likely to result in different and negative and even pathological content of the self. However, it should not affect the emergence of self-awareness or the mental representation of me. Likewise the content of our self-awareness may include the Western idea of a single self, standing alone and in opposition to other such selves – the I self – versus the we-self found in Japan and Indian cultures (Roland, 1988). (4) What does a self do? This question has to do with why we need the construct of self at all. For example, instead of the term self-regulation, we could simply use the term system regulation. Self regulation implies something unique about the self. For each of the features of self that we articulate, we can have the same question; what would the child be like if it does not have that self feature? It is obvious that for the earliest features of self their absence would result in such maladaptive behavior that the organism could not survive. Thus, failures of self-other differentiation or self-other interaction results in disorders that we think of as autism or some type of retardation. More important for this discussion is what would happen if self-awareness or the mental representation of me did not emerge? What are the differences between children who have self-awareness and those who are developmentally less advanced? From our theoretical point of view, if I found no difference between a child who does not have self-awareness and those that do, I would need to revise the theory. If, on the other hand, there was a difference, this difference should inform us as to the role and importance of this capability. In case of a self-aware organism, we could expect the child to show, in addition to a variety of capabilities, such as empathy and embarrassment, and later shame, guilt, and pride, self-referential behaviors such as self-recognition, personal pronoun usage and pretend play (Lewis, 1992a). Prior to the emergence self-awareness, emotional capacities of empathy or embarrassment are not present, while the basic emotions like fear and anger, are. In a study by Lewis, Sullivan, Stanger, and Weiss (1989) we have shown that prior to self-referential behavior, infants show anger or fear but not embarrassment. I have argued, as has Darwin (1872/1965) that embarrassment exists only after we have a selfaware of itself, an ‘‘I’’ that can be the object of others attention. Darwin wrote, ‘‘It is not the simple act of reﬂecting on our own appearance (a mental representation of me) but the thinking what others think of us, which excites a blush’’ (Darwin, 1872/1965, p. 327). From my point of view, humans and nonhumans share many features of self. That feature that we do not share is a mental representation of me and its cognitive elaboration through cultural learning that makes us different from other creatures.

2. The self-system Because we use the term ‘‘self’’ in a wide variety of ways, for example in reference to plants and to cells, when we talk about newborn infants (Kopp, 1982), intersubjectivity in 6-month-olds (Rochat, 2009; Stern, 1985), an ‘‘I’’ self or a ‘‘we’’ self (see Geertz, 1984), as well as in multiple selves rather than a single self (Ross, 1989), we need to return to consider what selves are. To deal with these differences, I will turn to the idea of a system. The following are some propositions which address the characteristics of any complex system and why some behaviors are not the same as self-awareness. I think we need to think of the human child, body and mind as a complex system. The thing about systems is that they have certain features (Von Bertalanffy, 1967). 1. All living systems self-regulate. By this we mean that within any living system there is need for communication between parts of that system. This can include a unit as small as a cell, a plant or animal, or even more complex organism. As I sit here writing, my systems are regulating my temperature, producing shivering as the room cools, or regulating my blood sugar level which if I pay attention to ‘‘me’’ will inform ‘‘me’’ that I am hungry. Self-regulation is a property of living matter. 2. Some minimal differentiation between self and other is a necessary condition for the system to act. Whether this differentiation is a product of experience or part of the process of action – including perceiving – is unknown (see Butterworth, 1992). What appears to be so is that organisms cannot act without at some level being able to distinguish between self and other. The ability to self-regulate or to distinguish self from other is part of the core processes of all living systems. 3. Even higher order functions such as perception, thinking, and even problem solving as well as complex actions, such as driving a car, can be carried out by adult humans without invoking a mental state of me, that is, without their being able to reﬂect on, look at and observe the processes which allow these behaviors to be carried out. I cannot watch myself think. I can only look at the product of my thinking. It is probably the case that once this capacity emerges developmentally, it can or cannot be invoked. 4. A unique aspect of some self-systems is a mental representation of the system. By mental representation I mean the capacity of a self to know it knows or to remember it remembers. It also means that I have the capacity to know that there are things I do not know about myself. I can know through learning about my amygdala’s reaction when I am fearful, something I did not know about me 25 years ago. I know about my amygdala and presumably my amygdala does not know about me. It is the ‘‘meta’’ ability which we refer to when we say self-awareness.

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

123

Given this distinction between self and self-awareness, we need to address the following question: Is all information which is system known capable of being known? The answer is, not always and, in fact, most often not. System processes may be available to us but only at times. I may be able to learn to focus on some of my core processes, such as controlling my heart rate or blood pressure; however, when I focus on these I am likely unable to focus on other aspects of the system’s core processes. Some system knowledge may never be available to us. The process of thinking is an example. This suggests the likelihood that our self-awareness is limited. This problem also can be stated from an epistemological point of view. When I say ‘‘I know X,’’ is it the case that I must know that I know X. If it is the case that I can know that I know X, when is it the case that I do know that I do not know X? These kinds of epistemological questions require different ways of knowing, self-awareness only one, however it is an important aspect of consciousness. This analysis leads to the view that there are at least two aspects of self, a mental representation of self and a systems self. From an epistemological point of view, knowledge can be found at both aspects of the self, but the knowledge of the knowledge is always a self-awareness aspect. Knowledge of the knowledge is the capacity of the self to reﬂect upon itself. This ability to reﬂect upon or have a mental representation of me is what makes the self so important in our understanding of our emotional lives. The notion of a self is most helpful in understanding the development of emotions. Below I will discuss emotions in more detail but for now let us suggest that emotions include emotional states and experiences (Lewis, 1992a, 2003; Lewis & Michalson, 1983). Emotional states are innate or learned action patterns including facial and bodily expression as well as physiological changes. People can have certain emotional states and yet not be aware that they have them: that is, they have state but no experience of their state. Darwin and others, Izard (1977) for example, considered emotional states to be action patterns evolutionarily derived and adaptive as such, part of the self-system requiring no mental representation. These processes can have goals, can learn and proﬁt from experience, can control functions, and can react to events including people. The experience of our emotional states, however, refers to a mental representation or what William James and more recently Damasio (2003) refer to as ‘feelings’. They require a mental representation or self-awareness, and that this self-awareness develops over the ﬁrst few years of life (Darwin, 1872/1965). Wanting to write this essay requires a ‘‘me wanting,’’ it requires a plan. The very act of writing these words, although planned and involving self reﬂection, is carried out by the system machinery of my body. How could it be that I can write (a motor act) almost effortlessly complicated phrases and thoughts without giving much attention to the processes that give rise to them? I certainly know, as I sit here writing (notice the self reﬂection), that I have a plan to write this and I have an outline which I have made to help me formulate my thoughts. It is clear that I have intentions and desires and presumably the ability to carry out the task of thinking and writing, yet the acts themselves seem to emerge from me almost effortlessly. Indeed, if I focus my attention on them I ﬁnd that doing so interrupts the very act that I am performing (see Wegner, 2009). We can also see this in the problem of not being able to do as in Akrasia (Aristotle, 1953), where an intention to do X does not lead to X or may lead to Y instead. The idea of intentions failed suggests that one way to understand the self is to assume the idea of multiple aspects of the self, only one being a mental representation of me or as I have called it, consciousness. Thus, Akrasia as well as self deception force us to view the self as a complex system, a modular system with features not in total communication with each other (see Gazzaniga, 1988); not a particularly new idea about what a self might be (see Wylie, 1961, for a historical review of this issue). 3. The development of a mental representation of me or self-consciousness Given this idea of what an adult self might be, and given the idea that competence can exist without mental comprehension, how are we to treat the idea of the development of self? Here our brain maturation-cultural perspective differs from others, incorporating both brain maturation and socialization. As we have argued, the claim of intersubjectivity (see Rochat, 2009; Stern, 1985; Trevarthen, 1980) revolves around the observation of coordinated actions between the infant and its caregiver. However, complex social action patterns does not mean comprehension nor a mental representation, and as Dennett (2009) has argued, this can be the cause of much confusion and may be at the heart of our different perspectives. This problem, for me, resides in our loose use of the term self. Newborn imitation, the process of intersensory integration, be it mirror neurons or some other process, does not mean that the ‘‘idea of me,’’ the mental representation of myself, has to exist. Nevertheless, alternative views about the mental representation of me exist and for the most part involve the idea of intersubjectivity. A recent example of this idea of the existence of both aspects of the self comes from Rochat (2009), who suggests that the infant is born with an implicit embodied self, based on a ‘‘ﬁrst-person perspective, formed on privileged perceptual information. This ﬁrst-person self becomes a third-person self based on the infants interaction with others.’’ This sounds similar to the idea here expressed as a self-system versus mental representation. But here is the difﬁculty, since Rochat argues that both the ﬁrst- and third-person selves have self-knowledge which are always at odds with themselves. By this we are to understand that Rochat gives to the ﬁrst-person self, conceptual and evaluative ability equal to the a mental representative self even from birth? As I understand it, this is similar to Freud’s Id and Ego description and the conﬂict between them. But Freud understood that the ego or mental abilities develop. It also reminds me of Freud’s critique of Rank’s theory of birth trauma. Rank claimed that the child could watch itself come into and out of being (existence), and this caused the infant to be fearful. Freud cautioned not to give the newborn ego functions, that is, the ability to think about itself from the beginning of life.

124

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

Often the bodily aspects of a self are given mental attributes of a later self. Take for an example, how our language usage gets us into trouble, the distinction between intentionality and intention. Intentionality refers to goals and processes built into a system that may not require any elaborate mental state, for example a thermostat controlling the temperature of the room. Intentions, however, are mental states. A ﬂower turning toward the sun or a newborn imitating a tongue protrusion might be said to have intentionality but the ﬂower does not have intentions as a mental state. If our language is not precise we may attribute mental state about goals to actions without them (Roitblat, 1990). 4. The self-conscious emotions From my perspective, self-conscious emotions, in particular shame, can be used to distinguish the point of view presented here from those who would attribute mental representation of me rather than system processes. While it is the case that shame can be elicited by many causes, the claim that it exists earlier than 2.5 years is similar to the proposition that all emotions, including the self-conscious ones can be seen in the early months of life (see Masciuch and Kienapple (1989), for the example of jealousy). Previously, it was thought that the self-conscious emotions, such as shame, took time to develop. Freud, for one, required an ego and a superego in order to have guilt and shame. Tomkins’s (1963) book on shame has as its central theme, the idea of thoughts, how children think of others’ thoughts about them as the elicitor of shame. Children’s thinking about others thinking about them needs to develop. Darwin’s theorizing about the self-conscious emotions required a developed cognitive system where a mental representation of me is needed. Darwin related the self-conscious emotions to the process of thinking about what others are thinking about oneself. He wrote about the nature of the mental states which induce blushing. For example, ‘‘These consist of shyness, shame and modesty; the essential element in all being selfattention.’’ Notice, self-attention, what I have called self-awareness or the ability to reﬂect on oneself. He believed that ‘‘Many reasons can be assigned for thinking that originally self-attention directed to personal appearance, in relation to the opinion of others, was the exciting cause; the same effect being subsequently produced, through the force of association, by self-attention in relation to moral conduct. It is not the simple act of reﬂecting on our own appearance, but the thinking what others think of us, which excites a blush’’ (Darwin, 1872/1965, pp. 326–327; italicize added). We are back to our recursive idea ‘‘I think about what you think of me.’’ All of these theories about the emergence of self-conscious emotions are cognitive in nature to the extent that thoughts about oneself, in relation to thoughts about others thoughts about the self, are the elicitors of embarrassment, shame, guilt, and pride. This suggests that the developmental sequence of I know, I know I know, I know you know, and ﬁnally I know you know I know, should not be confused nor thought to exist at birth but should be considered as a developmental progression. ‘‘I know’’ is bodily or system knowledge (newborn imitation being an example). ‘‘I know I know’’ is the onset of a mental representation, what I have considered the onset of self-awareness or consciousness. It may be at this time that the child also has some mental representations of what another knows. However, it is ‘‘I know, you know, I know’’ that the more adult-like human cognitions appear. I believe that if we keep in mind that competence is not to be equated with comprehension, we can allow for the processes of development to occur. This avoids the world view of the infant as a miniature adult and avoids as well the pitfalls of nativism. Now let us return to the self-conscious emotions, for it is here that the need for a mental representation of me or self-awareness plays a central role. I have spent considerable time studying the relation between emotion and self-representation and thus here I will only brieﬂy comment on it (Lewis, 1992b, 1997, 2000, 2001; Lewis & Michalson, 1983). Brieﬂy, I have proposed a model of emotional development where the mental representation of me gives rise to two sets of self-conscious emotions, those I refer to as self-conscious exposed emotions and self-conscious evaluative emotions (Lewis, 1992b, 2002). All of these self-conscious emotions require a mental representation of me or consciousness, although as in the case of the evaluative emotions, more is required. While the emotions that appear early – such as joy, sadness, fear, and anger – have received considerable attention, the set of later-appearing emotions that I wish to consider has received relatively little attention. There are likely to be many reasons for this; one reason is that these self-conscious emotions cannot be described solely by examining a particular set of facial movements, necessitating the observation of bodily action as well as facial cues. A second reason for the neglect of study of these later emotions is the realization that there are no clear speciﬁc elicitors of these particular emotions. While happiness can be elicited by seeing a signiﬁcant other, and fear can be elicited by the approach of a stranger, there are few speciﬁc situations that will elicit shame, pride, guilt, or embarrassment. These self-conscious emotions are likely to require classes of events that only can be identiﬁed by the individuals themselves. The elicitation of self-conscious emotions involves elaborate cognitive processes that have, at their heart, mental states about the self or explicit consciousness. While some theories, such as psychoanalysis (Erikson, 1950; Freud, 1963; Tomkins, 1963), have argued for some universal elicitors of self-conscious emotions, such as failure at toilet training or exposure of the backside, the idea of an automatic non-cognitive elicitor of these emotions does not make much sense. Cognitive processes must be the elicitors of these complex emotions (Lewis, 1992a). It is the way we think or what we think about that becomes the elicitor of these emotions. There may be a one-to-one correspondence between thinking certain thoughts and the occurrence of a particular emotion; however, in the case of this class of emotions, the elicitor is a cognitive event. This does not mean that the earlier emotions, those called primary or basic, are elicited by non-cognitive events. Cognitive factors may play a role in the elicitation of any emotion; however, the nature of cognitive events are much less articulated and differentiated in the earlier ones (Plutchik, 1980).

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

125

The need for cognitive elicitors having to do with the self was known to Darwin (1872/1965). He suggested that these emotions were a consequence of our thoughts about other’s thoughts of us, there being therefore no clear or universal elicitors. Darwin saw these latter emotions as involving the self, although he was not able to distinguish among the various types (see also Tomkins (1963) and Izard (1977) for similar problems). His observation in regards to blushing indicates his concern with the issue of appearance and the issue of explicit consciousness. He repeatedly makes the point that these emotions depend on sensitivity to the opinion of others, whether good or bad. We have attempted to clarify those speciﬁc aspects of self that are involved in self-conscious emotions. First let us consider self-conscious exposed emotions. The self-conscious exposed emotions have been differentiated from the selfconscious evaluative emotions since the latter require fairly elaborate cognitions around standards, rules, and goals and around attributions relevant to the self (Lewis, 1992b; Lewis & Michalson, 1983). These evaluative emotions will be considered next. The exposed emotions consist, at least, of embarrassment, empathy, and jealousy. While some work has been done looking at empathy (Bischof-Kohler, 1991), most of the work has been conducted on embarrassment (Lewis, 1995). I have tried to distinguish between two different types of embarrassment – one related to exposure and one related to evaluation which has much in common with shame. Exposure embarrassment emerges once self-recognition (a measure of the mental representation of me) can be shown, around 15–24 months, while evaluative embarrassment does not emerge until 2½ years. An example of exposure embarrassment is the embarrassment that occurs when one is complimented. Praise rather than a negative evaluation is the source of this type of embarrassment. Another example of this type of embarrassment can be seen in our reactions to public display. When people observe someone looking at them, they are apt to become self-conscious, look away, and touch or adjust their bodies. In few cases do the observed people look sad. If anything, they appear pleased by the attention. This combination – gaze turned away brieﬂy, no frown, and nervous touching – is exposure embarrassment. Another example of embarrassment as exposure can be seen in the pointing to a child. In a series of studies, we have demonstrated the effectiveness of complimenting, pointing to the child, and asking him/her to perform – dance to music – in front of us as three different elicitors of exposure embarrassment (Lewis, Stanger, & Sullivan, 1989; Lewis, Stanger, Sullivan, & Barone, 1991). The relation between self-recognition measuring a mental representation of me and exposure embarrassment has been explored and the ﬁndings are quite clear. Exposure embarrassment is signiﬁcantly more likely to be seen once the child shows self-recognition. However, the earlier more basic emotions, such as wariness or fearfulness, are unaffected by the child’s emerging self-awareness (Lewis, Stanger, et al., 1989; Lewis, Sullivan, et al., 1989). Thus, while the basic emotions such as joy, sadness, fearfulness, disgust, anger, and interest all emerge prior to self-recognition and explicit consciousness, the exposed self-conscious emotions require its emergence. Looking at another nonevaluative self-conscious emotion, empathy, ﬁnds a similar result (Bischof-Kohler, 1991). This should not be surprising given that adult empathic responses require that one be able to place one’s self in the role of the other, an ability which obviously requires a mental representation of me. Self-conscious evaluative emotions not only require a mental representation but also require an elaborate set of other cognitive capacities. Because of this, these emotions do not emerge until 2½–3 years (Lewis, 1992b). They all require knowledge about standards, rules, or goals. These standards are inventions of the culture which are transmitted to the child and involve the child’s learning of and willingness to consider these as their own. This process of incorporating the standards has been discussed by Stipek, Recchia, and McClintic (1992). What is apparent from this work is that learning starts quite early in life. Standards, rules, and goals imply self-evaluation for it would make little sense if we had standards but no evaluation of our action vis a vis them. Having self-evaluative capacity allows for two distinct outcomes; we can evaluate our behavior and hold ourselves responsible for the action which is being evaluated, or we can hold ourselves not responsible. In the attribution literature, this distinction has been called either an internal or an external attribution (Weiner, 1986). If we conclude that we are not responsible, then evaluation of our behavior ceases. However, if we evaluate ourselves as responsible, then we can evaluate our behavior as successful or unsuccessful vis a vis the standard. Finally, global self-attributions refer to the whole self, while speciﬁc self-attributions refer to speciﬁc features or actions of the self (Dweck & Leggett, 1988; Weiner, 1986). These are sometimes referred to as performance versus task orientation (Dweck, 1996). In every one of these processes, the mental representation of me needs to be considered. The terms global and speciﬁc are used to specify the tendency of individuals to make speciﬁc evaluations about themselves (Beck, 1967, 1979; Seligman, 1975). Global evaluations about themselves refers to an individual’s focus on the total self and on their performance. Thus, for any particular behavior violation, an individual can focus on the totality of the self; and then use such self-evaluative phrases as, ‘‘Because I did this, I am bad [or good].’’ Janoff-Bulman’s (1979) distinction is particularly relevant here. In global attributions, the focus is upon the self and performance. The self becomes embroiled in the self. The focus is not upon the self’s behavior as in task focus, but upon the self. There is little wonder that in using such global attribution one can think of nothing else, and one becomes confused and speechless (Lewis, 1971). We turn to focus upon ourselves, not upon our actions. Because of this, we are unable to act and are driven from the ﬁeld of action into hiding or disappearing. Speciﬁc, in contrast, refers to the individual’s propensity to focus on speciﬁc actions of the self and on the task. It is not the total self that has done something wrong or good, it is speciﬁc behaviors in context that are judged. Individuals use such evaluative phrases as, ‘‘My behavior was wrong, I must not do it again.’’ Notice that the individual’s focus is on the task in a speciﬁc context, not on the totality of the self. These cognitions, which focus on the self, give rise to the self-conscious

126

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

evaluative emotions. Our research indicates that these emotions do not emerge until after the onset of self-awareness (in the middle of the second year of life) and not until the child is capable of the complex cognitions associated with standards. By 2½–3 years, these cognitive capacities are present and so is the emergence of these self-conscious evaluative emotions (Lewis, 1992b). We can see, therefore, that the role of a mental representation of me in both classes of emotion is quite elaborate, involving: (a) the mental representation of me; (b) knowledge of standards, rules, and goals; (c) evaluation of one’s behavior vis a vis the standards; (d) distribution of the blame to oneself or to others; and (e) attribution focus, either global or performance focus or speciﬁc and task focus. In each one of these processes, mental structures about me are required. 5. The development of a representational self I have called my view a brain maturation-cultural theory that requires the socialization of values since once a representation emerges it has to be about something. The something are ideas and ideas are cultural artifacts, inseparable from it. Below, then is a discussion of the development of a representational self. We have studied its development using three measures which we believe taken together reﬂects the emerging representational self. These are the use of personal pronouns, including ‘‘me’’, ‘‘mine’’, and eventually ‘‘I’’ (Harter, 1983; Hobson, 1990). Another measure of a representation of self is mirror self-recognition (Lewis & Brooks-Gunn, 1979a, 1979b; Lewis & Michalson, 1983; Lewis & Ramsay, 2004). As has been widely reported and found to be the case across different cultures, this aspect, self-recognition, has been associated with brain maturation (Lewis, 2003) and which we have shown in several recent studies to be associated with myleniation of the left temporal region, in particular the left temporal parietal lobe (Carmody & Lewis, 2006, 2010; Lewis & Carmody, 2008). Pretend play appears to be an early manifestation of the toddler’s ability to understand mental states including its own and others. It also appears to be related to personal pronoun usage as well as mirror recognition (Lewis & Ramsay, 2004). The use of these three measures appears to be the best measure of a mental representation we have found. It is apparent that with the development, self-representation increasingly becomes a more complex and multifaceted phenomenon that progressively includes other cognitive and evaluative aspects of knowledge such as gender identity (Lewis & Brooks-Gunn, 1979a) and age (Lewis, 1985). Nonetheless, our research suggest that in terms of emergent time, self-recognition appears within the 15–24 month window of time as does the other measures of pretend and personal pronoun usage. Children, whose development of these measures of self-representation does not occur typically, can be found in the cases of children with Down’s syndrome and autism. Consistent with these ﬁndings is work that indicates children’s emerging understanding of a theory of mind by the middle of the second year of life. For example, Meltzoff (1995) reports that 18-month-old toddlers have the ability to understand the intentions of others. After observing adult models demonstrate the intention to act in a certain way by starting, but not completing, a given activity, the toddlers, when given the opportunity, performed the complete acts the adult intended. Similarly, Asendorph and Baudonniere (1993) and Asendorph, Warkentin, and Baudonniere (1996) have found increases in imitative play linked to the presence of self-recognition in 20-month-old infants. Indeed, there are many studies linking self-recognition to other abilities that mark more broadly the emergence of self-representation. For example, self-recognition is related to children’s self-conscious emotions, in particular embarrassment (Lewis, Sullivan, Stanger, & Weiss, 1989) and empathy (Bischof-Kohler, 1994), as well as altruism (Zahn-Waxler, Radke-Yarrow, Wagner, & Chapman, 1992). Self-recognition is related to autobiographical memories (Harley & Reese, 1999) and is associated with imitation (Asendorph, 2002). Pretense is an early manifestation of the ability to understand mental states, including one’s own (Leslie, 2002; Piaget, 1952), as well as an understanding of a negation by the self that ‘‘this is not what I pretend it to be’’ rather than what it actually is. 6. Mental representation of me develops in the ﬁrst three years of life To supply a time line and to integrate our view of development I will return to our model brieﬂy mentioned before. The proposition is that there are 3 or 4 phases to the development of a representational self. Level 1 is called knowing (or I know). This level exists from birth until the second year of life when new skills are added and is likely to be driven by basic processes common to other animals. It is based on system’s ability and involves little or no language; it is not supported by the mental state of the idea of ‘‘me’’. Many organisms share in this kind of knowledge. For example, when an object in the visual ﬁeld rapidly expands, infants, as well as adults and animals, show surprise and discomfort. This response is simply built into the core features of the system’s perceptual-motor knowledge. Likewise early social interactions are controlled by systems processes. For example, they are controlled by contagion, where the behavior of one conspeciﬁc triggers, much like a yawn, behavior in another. Newborn crying increases when other newborns cry (Sagi & Hoffman, 1976). Complex social exchanges are built into the system, there is intentionality but no intention on the part of the infant, as both are present in the adult. In the last 30 years there has been an expanded list of infants’ competences which reﬂect such a knowledge level. Infants’ competences, however, are not the same as understandings and distinction between competence and understanding is necessary in order not to confuse this level of knowledge from the next level. Level 2 is I know I know. This level involves self-referential behavior. Brain maturation, more speciﬁcally in the left hemisphere and in particular the temporal lobes is related to this level and it emerges between 15 and 24 months of age in typically developing children. It is based on a mental state of ‘‘me’’ and allows for the capacity to reﬂect on one’s self and to

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

127

reﬂect on what one knows. This mental state is a meta-representation. It is similar to a memory of a memory. Whereas a child at the ﬁrst level may have a memory, it is at the second level that meta-memory is possible. Here the child remembers that he or she remembers. As we have seen, this capacity emerges somewhere in the middle of the second year of life (Lewis, 1990a, 1990b). Level 3 is I know you know. This form of knowing takes into account the mental state that not only do I know something, but I believe others know it as well; it is the ability and basis of shared meaning. Adult-like interactions, not system based ones, are possible. This representation, that you know what I know, does not need to be accurate. Adults know more than children know; thus the child may not really know what the adult or another knows. The child is likely to make errors, something we might call egocentric errors. That is, the child assumes that what it knows is what the other knows. At this level children know that they know, and they also know you know. What they cannot yet do is to place themselves in opposition to what they know. This level, in combination with the earlier ones, accounts in part for the early ability to deceive. A 2½year-old child who deceives knows that he knows and he knows that you know; thus deception is possible (Lewis, Stanger, & Sullivan, 1989). It is also the reason why children are likely to make the traditional false belief error. Before going onto the fourth level, it is worth mentioning that the third level may not be distinct from the one before it in which children knows they know. It is possible that the mental state of the idea of me and what I know may emerge at the same time as the mental state of what I know about what others know. In other words, it is possible that what I know about me is part of what I know about the other (Merleau-Ponty, 1964). If this indeed is the case, then a separate level might not be called for. Level 4 is the more adult-like level. It addresses the interactive and recursive nature of cognition. It is characterized as I know, you know, I know. At this level, not only are there two actors, as at Level 3, but each actor has a perspective. These perspectives are cultural artifacts affecting the content of the two actors. They can be similar or different. It is when there are two perspectives that one has the ability to recognize false belief. Only when one has reached the level of knowing that ‘‘they know I know’’ that your knowledge about what they know can be corrected, since you can check their knowledge of what they know about you against what you know. That is, once a child knows that it can be the subject and also the object of the knowledge of another, it is capable of recognizing the difference in perspectives between individuals. It is at this ﬁnal level of perspective-taking that mature meta-knowledge can emerge. It is important to note that in this sequence, these levels do not replace one another, rather they are added to the earlier levels. Thus by the fourth level all ways of knowing exist in the child as in the adult. 7. Mental representation as consciousness and the integration of domains of knowledge Before closing I should brieﬂy like to argue that a mental representation of me has particular use in connecting the various domains of social, cognitive and emotional capacities (Lewis, 2010). I have argued that the organization of development follows from the assumption that social, emotional, and cognitive knowledge are features of the same uniﬁed relational development system that is fundamental to the individual’s consciousness. Individuals develop social, emotional, and cognitive knowledge in relational and bidirectional interactions with each other. Moreover, I understand development as a gradual differentiation among the various domains (Werner, 1961). The change from a uniﬁed system of knowledge based on the emergence of consciousness to one which is differentiated, integrated, and specialization occurs as a further function of development. I see this system like a tree, the trunk representing the uniﬁed and integrated system generated by consciousness, while the branches represent the separate areas of knowledge, some of which are interrelated whereas others are independent. This model allows for both the integration of knowledge from a developmental perspective and functional independence as the end product. Thus, as a central premise, the development of consciousness provides the scaffolding for the development, integration, and separation of the various other behaviors of the child. To begin with, these domains are supported by the system processes but with the onset and development of consciousness they become adult human attributes. Thus, perceptual-sensory knowledge becomes a theory of mind, social interactions become social relationships, and the early or basic emotions become the self-conscious emotions, all possible because of consciousness (Lewis, 2003). Throughout this paper I have made reference to Freud who I have always considered having a deterministic view of development: Early events effect later events, there is an end point, and therefore development has a direction. While this is true, recently in a conversation with the psychoanalyst Shelly Bach, I was reminded of Freud’s idea of Nachtraglich. As you might recall the idea behind this term was that events which take place earlier in ones life can reappear and can be given new interpretation, ones quite different than given before. Our ability to reinterpret past events and change their meaning resides in our complex self-system and the mental representation of me. It is this capacity to think about myself, both as I am, I was, and what I would like to be, which gives us our ability to change our selves. This is made possible only by a conscious mind, one that utilizes the representation of ‘‘me’’ to construct its cognitive, social, and emotional world. This aspect of self is what I see developing both as a function of the biology of our brains but also the content of our cultures. References Aristotle (1953). The Nichomachean Ethics. London: Penguin Books. Asendorph, J. B. (2002). Self-awareness, and secondary representation. In A. N. Meltzoff & W. Prinz (Eds.), The imitative mind: Development, evolution, and brain bases. Cambridge studies in cognitive perceptual development (pp. 63–73). New York: Cambridge University Press.

128

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

Asendorph, J. B., & Baudonniere, P. M. (1993). Self-awareness and other-awareness: Mirrorself-recognition and synchronic imitation among unfamiliar peers. Developmental Psychology, 29(1), 88–95. Asendorph, J. B., Warkentin, V., & Baudonniere, P. M. (1996). Self-awareness and other-awareness. II: Mirror self-recognition, social contingency awareness, and synchronic imitation. Developmental Psychology, 32(2), 313–321. Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical aspects. New York: Times Mirror. Beck, A. T. (1979). Cognitive therapy and emotional disorders. New York: Times Mirror. Bischof-Kohler, A. (1991). The development of empathy in infants. In M. E. Lamb & H. Keller (Eds.), Development: Perspectives from German-speaking countries (pp. 245–273). Hillsdale, NJ: Lawrence Erlbaum. Bischof-Kohler, D. (1994). Self-objectiﬁcation and other-oriented emotions: Self-recognition, empathy, and prosocial behavior in the second year. Zeitschrift fur Psychologie, 202, 349–377. Butterworth, G. (1992). Origins of self-perception in infancy. Psychological Inquiry, 3, 103–111. Carmody, D. P., & Lewis, M. (2006). Brain activation when hearing one’s own and others’ names. Brain Research, 1116, 153–158. Carmody, D. P., & Lewis, M. (2010). Regional white matter development in children with autism spectrum disorders. Developmental Psychobiology, 52, 755–763. Damasio, A. (2003). Looking for Spinoza: Joy, sorrow, and the feeling brain. London: Heinemann. Darwin, C. R. (1872/1965). The expression of the emotions in man and animals. Chicago: University of Illinois Press [Original work published in 1872]. Dennett, D. C. (2009). Darwin and the evolution of ‘why’?. Darwin festival, Cambridge, July 8, 2009 [Abstract]. Dweck, C. S. (1996). Social motivation: Goals and social-cognitive processes. In J. Invonen & K. R. Wentzel (Eds.), Social motivation: Understanding children’s school adjustment (pp. 181–195). New York: Cambridge University Press. Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95, 256–273. Erikson, E. (1950). Childhood and society. New York: Norton. Freud, S. (1963). The problem of anxiety. New York: Norton [H.A. Bunker, trans., original work published in 1936.]. Gazzaniga, M. S. (1988). Brain modularity: Towards a philosophy of conscious experience. In A. J. Marcel & E. Bisiach (Eds.), Consciousness in contemporary science. New York: Oxford University Press. Geertz, C. (1984). On the nature of anthropological understanding. In R. A. Shweder & R. A. Levine (Eds.), Cultural theory: Essays on mind, self, and emotion. Cambridge, England: Cambridge University Press. Harley, K., & Reese, E. (1999). Origins of autobiographical memory. Developmental Psychology, 35, 1338–1348. Harter, S. (1983). Developmental perspectives on the self-system. In E. M. Hetherington (Ed.). Handbook of child psychology. Socialization, personality and social development (Vol. IV, pp. 275–385). New York: Wiley. Hobson, R. P. (1990). On the origins of self and the case of autism. Development and Psychopathology, 2, 163–181. Izard, C. E. (1977). Human emotions. New York: Plenum Press. Janoff-Bulman, R. (1979). Characterological versus behavioral self-blame: Inquiries into depression and rape. Journal of Personality and Social Psychology, 37, 1798–1809. Kagan, J. (2008). In defense of qualitative changes in development. Child Development, 79(6), 1606–1624. Kopp, C. B. (1982). Antecedent of self-regulation: A developmental perspective. Developmental Psychology, 18, 199–214. Leslie, M. (2002). Pretense and representation revisited. In N. L. Stein & P. J. Bauer (Eds.), Representation, memory, and development: Essays in honor of Jean Mandler (pp. 103–114). Mahwah, NJ: Erlbaum. Lewis, M. (1967). The meaning of a response, or why researchers in infant behavior should be oriental metaphysicians. Merrill-Palmer of Behavior and Development, 13(1), 7–18. Lewis, H. B. (1971). Shame and guilt in neurosis. New York: International Universities Press. Lewis, M. (1979). The self as a developmental concept. Human Development, 22, 416–419. Lewis, M. (1985). Age as a social dimension. In T. Field & N. Fox (Eds.), Social perception in infants (pp. 299–319). New York: Academic Press. Lewis, M. (1990a). Social knowledge and social behavior. Special Issue of Merrill-Palmer Quarterly, 36(1), 93–116. Lewis, M. (1990b). Thinking and feeling – The elephants tail. In C. A. Maher, M. Schwebel, & N. S. Fagley (Eds.), Thinking and problem solving in the developmental process: International perspectives (the work) (pp. 89–110). Hillsdale, NJ: Lawrence Erlbaum. Lewis, M. (1992a). Shame: The exposed self. New York: Free Press. Lewis, M. (1992b). Shame, the exposed self. Zero to Three, 7(4), 6–10. Lewis, M. (1995). Embarrassment: The emotion of self exposure and evaluation. In J. P. Tangney & K. W. Fischer (Eds.), Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride (pp. 198–218). New York: Guilford Press. Lewis, M. (1997). Altering fate: Why the past does not predict the future. New York: The Guilford Press. Lewis, M. (2000). The emergence of human emotions. In M. Lewis & J. Haviland-Jones (Eds.). Handbook of Emotions (Vol. 2nd ed., pp. 265–280). New York: Guilford Press. Lewis, M. (2001). The origins of the self-conscious child. In R. Crozier & L. E. Alden (Eds.), International handbook of social anxiety: Concepts, research and interventions relating to the self and shyness (pp. 101–118). Sussex, England: John Wiley & Sons, Ltd. Lewis, M. (2002). Early emotional development. In A. Slater & M. Lewis (Eds.), Introduction to infant development (pp. 192–209). England: Oxford University Press. Lewis, M. (2003). The emergence of consciousness and its role in human development. In J. LeDoux, J. Debiec, & H. Moss (Eds.). The self: From soul to brain (Vol. 1001, pp. 1–29). New York: Annals of the New York Academy of Sciences. Also appearing on the annals online <www.annalsnyas.org>. Lewis, M. (2010). The emergence of consciousness and its role in human development. In R. Lerner & W. Overton (Eds.), Cognition, biology, and methods. The handbook of lifespan development (Vol. 1, pp. 628–670). MA: Wiley. Lewis, M., & Brooks-Gunn, J. (1979a). Social cognition and the acquisition of self. New York: Plenum. Lewis, M., & Brooks-Gunn, J. (1979b). Toward a theory of social cognition: The development of self. In I. Uzgiris (Ed.), New directions in child development: Social interaction and communication during infancy (pp. 1–20). San Francisco, CA: Jossey-Bass. Lewis, M., & Carmody, D. (2008). Self representation and brain development. Developmental Psychology, 44(5), 1329–1334. Lewis, M., & Michalson, L. (1983). Children’s emotions and moods: Developmental theory and measurement. New York: Plenum. Lewis, M., & Ramsay, D. (2004). Development of self-recognition, personal pronoun use, and pretend play during the 2nd year. Child Development, 75, 1821–1831. Lewis, M., Stanger, C., & Sullivan, M. W. (1989). Deception in three-year-olds. Developmental Psychology, 25(3), 439–443. Lewis, M., Stanger, C., Sullivan, M. W., & Barone, P. (1991). Changes in embarrassment as a function of age, sex, and situation. British Journal of Developmental Psychology, 9, 485–492. Lewis, M., Sullivan, M. W., Stanger, C., & Weiss, M. (1989). Self-development and self-conscious emotions. Child Development, 60, 146–156. Mandler, G. (1975). Mind and emotion. New York: Wiley. Masciuch, S., & Kienapple, K. (1989). The emergence of jealousy in children 4 months to a 7 years of age. Journal of Social and Personal Relationship, 10, 421–435. Meltzoff, A. N. (1995). Understanding the intentions of others: Re-enactment of intended acts by 18-month-old children. Developmental Psychology, 31, 838–850. Merleau-Ponty, M. (1964). Primacy of perception. Evanston, IL: Northwestern University Press [J. Edie (Ed.), W. Cobb, Trans.]. Neisser, U. (1988). Five kinds of self-knowledge. Philosophical Psychology, 1, 35–59. Piaget, J. (1952). The origins of intelligence in children. New York: International Universities Press [M. Cook, trans., original work published in 1936].

M. Lewis / Consciousness and Cognition 20 (2011) 120–129

129

Plutchik, R. (1980). A general psychoevolutionary theory of emotion. In R. Plutchik & H. Kellerman (Eds.). Emotion: Theory, research, and experience (Vol. 1, pp. 3–33). New York: Academic Press. Rochat, P. (2009). Others in mind: Social origins of self-consciousness. Cambridge: University Press. Roitblat, H. L. (1990). Causation, intentionality, and cognitive action theory. Psychological Inquiry, 1(3), 263–265. Roland, A. (1988). In search of self in India and Japan. Princeton, NJ: Princeton University Press. Ross, C. A. (1989). Multiple personality disorder. New York: Wiley. Sagi, A., & Hoffman, M. (1976). Empathic distress in the newborn. Developmental Psychology, 12(2), 175–176. Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco: Freeman. Stern, D. (1985). The interpersonal world of the infant. New York: Basic Books, Inc. Stipek, D., Recchia, S., & McClintic, S. (1992). Self-evaluation in young children. Monographs of the society for research in child development (Vol. 57), Serial no. 226. Tomkins, S. S. (1963). Affect imagery and consciousness. The negative affects (Vol. 2). New York: Springer. Trevarthen, C. (1980). The foundations of intersubjectivity: Development of interpersonal and cooperative understanding in infants. In D. Olson (Ed.), The social foundations of language and thought. New York: W.W. Norton & Co. Von Bertalanffy, L. (1967). Robots, men, and minds. New York: Brazilles. Wegner, D. (2009). How to think, say, or do precisely the worst thing for any occasion. Science, 325, 48–50. Weiner, B. (1986). An attributional theory of motivation and emotion. New York: Springer-Verglag. Werner, H. (1961). Comparative psychology of mental development. New York: Science Editions. Wylie, R. C. (1961). A self-concept: A critical survey of pertinent research literature. Lincoln: University of Nebraska Press. Zahn-Waxler, C., Radke-Yarrow, M., Wagner, E., & Chapman, M. (1992). Development of concern for others. Developmental Psychology, 28, 126–136.

Consciousness and Cognition 20 (2011) 130–140

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The role of the self in mindblindness in autism Michael V. Lombardo ⇑, Simon Baron-Cohen Autism Research Centre, Douglas House, 18B Trumpington Rd., Cambridge CB2 8AH, UK

a r t i c l e

i n f o

Article history: Available online 6 October 2010 Keywords: Autism Self Mindblindness Theory of mind Mentalizing Simulation Egocentrism

a b s t r a c t Since its inception the ‘mindblindness’ theory of autism has greatly furthered our understanding of the core social-communication impairments in autism spectrum conditions (ASC). However, one of the more subtle issues within the theory that needs to be elaborated is the role of the ‘self’. In this article, we expand on mindblindness in ASC by addressing topics related to the self and its central role in the social world and then review recent research in ASC that has yielded important insights by contrasting processes relating to both self and other. We suggest that new discoveries lie ahead in understanding how self and other are interrelated and/or distinct, and how understanding atypical selfreferential and social-cognitive mechanisms may lead to novel ideas as to how to facilitate social-communicative abilities in ASC. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction If history is any metric of importance, then the ‘self’ should be ranked not only among one of the most important topics in psychological inquiry, but also among one of the most important topics in autism research. Inquiry into the self dates back to as early as the ancient Greeks. The aphorism gno¯thi seauton (‘‘know thyself”) refers to the idea that to understand human behavior, morals, and thought one must ﬁrst understand oneself. Similarly, autism spectrum conditions (ASC) are named after the Ancient Greek word autos, which literally translates as ‘self’. The ﬁrst characterizations of autism by Kanner (1943) and Asperger (1944) introduced notions of ‘extreme egocentrism’ that lead to ‘autistic aloneness’. From the start, these early notions suggested the concept of ‘self’ is fundamentally altered in autism and is integral to the hallmark difﬁculties in social and communicative domains. If we fast forward to the mid-1980s, it was discovered that individuals with ASC are profoundly impaired in understanding minds; an normative ability we now refer to as ‘theory of mind’ or ‘mentalizing’ (Baron-Cohen, Leslie, & Frith, 1985). This theory of mind impairment in ASC is not explained simply as a deﬁcit in general meta-representation (e.g., decoupling events from reality) as individuals with ASC pass tests of ‘out of date’ pictorial representations (e.g., false photos) even whilst failing tests about understanding ‘out of date’ beliefs (e.g., false beliefs) (Charman & Baron-Cohen, 1992; Leslie & Thaiss, 1992). Furthermore, the theory of mind impairment in ASC extends into an understanding of one’s own mental states (Baron-Cohen, 1989; Perner, Frith, Leslie, & Leekam, 1989) and has been shown to be more impaired in the self-referential than the social domain (Williams & Happe, 2009). Rather than a complete lack of theory of mind, meta-analytic evidence suggests that many (though not all) individuals with ASC do develop a rudimentary explicit mentalizing ability, albeit at a very delayed point in development (Happe, 1995). However, even here, this explicit mentalizing ability developed at later ages may mask the true deﬁcits in understanding and attributing mental states, as studies of automatic or implicit mentalizing ﬁnd deﬁcits in ASC all the way up to adulthood (Abell, Happe, & Frith, 2000; Klin, 2000; Senju, Southgate, White, & Frith, 2009; Senju et al.,

⇑ Corresponding author. Fax: +44 01223 746 033. E-mail address: [email protected] (M.V. Lombardo). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.09.006

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

131

2010). Thus, the marked impairments speciﬁcally in representing mental states was formulated as the ‘mindblindness’ theory of autism and still stands today as one of the primary cognitive explanations behind social-communicative difﬁculties in ASC (Baron-Cohen, 1995; Frith, 2001; Hamilton, 2009). One of the main premises behind the mindblindness theory is the idea that while the general population possesses an intact mechanism for representing or attributing mental states to both self and other, this mechanism (Carruthers, 2009) or set of mechanisms (Baron-Cohen, 1995; Leslie, 1994) is profoundly impaired in ASC. Of particular note is that the proposed mechanisms for mindblindness in ASC have been very ‘other’-centric in nature (Baron-Cohen, 1995; Leslie, 1994), focusing on how individuals read social cues from others (e.g., facial expressions, eye gaze, body postures), or have been agnostic with respect to the target of mentalizing (Baron-Cohen, 1995; Carruthers, 2009) (e.g., intentionality detection, mental state representation). While informative in their own right, ‘other’-centric or ‘target-agnostic’ mechanisms have left a gap in terms of mechanisms that may be responsible for atypical self-referential processes in ASC and their integration into the bigger picture of how individuals navigate and interact with the social world. Thus, rather than adopting the a priori stance that the mechanisms for mindreading are ‘other-centric’ or ‘target-agnostic’, we pose the question of whether there is anything to be gained by exploring similarities and differences in self-referential vs. other-referential processes in ASC. Approximately 15 years ago when the ‘mindblindness’ account was expanded to a consideration of underlying mechanisms, we alluded to the idea that there is much to be gained by taking into account mechanisms for both understanding our own and other minds (see p. 130 in Baron-Cohen (1995)). Along with work by many others, the self has begun to garner more attention in accounts of mindblindness in ASC (Carruthers, 2009; Frith, 2003; Frith & de Vignemont, 2005; Frith & Happe, 1999; Goldman, 2006; Happe, 2003; Hobson, 1990; Hobson, Chidambi, Lee, & Meyer, 2006; Hobson & Meyer, 2005; Hurlburt, Happe, & Frith, 1994; Nichols & Stich, 2003; Williams, 2010). With the growing momentum of research on the self in social cognition (not just in autism) over the past 2–3 decades, we take this opportunity to update the mindblindness account by examining some additional factors that take into account the importance of self-referential processing. In this article we begin by highlighting ideas from social psychology. These include ideas such as the relationship between self and other, egocentrism, simulation, asymmetry of informational sources about self and other, perceived similarity, and distinguishing self from other – just a few ways in which knowledge about the self in social cognition can start to provide new insights into the topic of mindblindness and social difﬁculties in ASC. Our hope is to steer future work in the direction of taking a more balanced approach, looking at how both atypical self-referential and social-cognitive mechanisms contribute to the social difﬁculties in ASC. 2. Relational multidimensional selves People do not exist in a vacuum. We are ‘selves’ embedded in a rich social world full of other ‘selves’. As social psychological and personality research demonstrates, ‘selves’ are not unidimensional; we are multidimensional constructs (Goldberg, 1990; Greenwald & Pratkanis, 1984). Our self-concept is represented as an array of traits in multidimensional space and this multidimensionality allows us to share some variance with others, yet at the same time still be unique – a property which we refer to as the ‘duality of self’ (Lombardo & Baron-Cohen, 2010; Lombardo et al., 2010a). This duality of self (being both similar to and yet different from others at the same time) is something we believe is important in social-cognitive processes such as mentalizing and we later discuss how the push and pull of similarities and differences between self and other can be used to overcome gaps we have in perceiving self and other. As multidimensional ‘selves’ embedded within a rich social world replete with other multidimensional ‘selves’, it is also well known that the relations we have with others subsequently affect our multidimensional percept of both self and other (Andersen & Chen, 2002; Aron, Aron, & Smollan, 1992; Aron, Aron, Tudor, & Nelson, 1991; Brewer, 1991; Kenny, 1994). For example, the unique relationship you develop with a close other differentially affects how you perceive that (now individuated) person independently of general trends you might exhibit when you generally perceive others. In social psychological models such as the inﬂuential ‘social relations model’ (SRM; Kenny & La Voie, 1984), this is a type of ‘relationship-effect’ that is an emergent property of the interaction between the target (e.g., another person) and the perceiver (e.g., oneself). This type of interaction is independent of the more general ‘main effects’ due to characteristics that just the ‘perceiver’ or ‘target’ bring to the equation. In another example, imagine two strangers engaged in repeat social interactions. Both individuals build perceptions of the other person that are emergent properties of the history of interactions with that person. Thus, part of the interactive process of mentalizing can be conceptualized in a computational framework modeling both ‘relationshipeffects’ as well as what the ‘target’ or ‘perceiver’ generally add to the equation. Computational approaches applied to mentalizing processes have recently emerged, with studies showing that speciﬁc parameters modeling the emergent properties of social interactions (e.g., relationship effects1) map onto variability in neural systems involved in mentalizing (Behrens, Hunt, & Rushworth, 2009; Behrens, Hunt, Woolrich, & Rushworth, 2008; Hampton, Bossaerts, & O’Doherty, 2008). It is also worth noting that this type of effect is more characteristic of naturalistic social interactions in the real world and thus gets us closer to the external validity we strive to attain.

1 The usage of the phrase ‘relationship effects’ should not be interpreted in the traditional way a ‘relationship effect’ is interpreted under the social relations model as the modeling strategies from these studies differ from those put forth by Kenny and La Voie (1984). However, the phrase is useful here since these modeling studies are assessing parameters that reﬂect the relations between people that emerge as a property of the growing interaction between them.

132

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

Adding to the complexity of emergent ‘relationship-effects’ in how we perceive ourselves and others, there are also general factors to account for. James (1890) was the ﬁrst to suggest that how we see ourselves is in part based on how we think we look in the eyes of others. This idea was developed into the concept of the ‘Looking Glass Self’ (Cooley, 1902; Mead, 1934) and refers to the fact that the perceptions of others have a substantial impact on how we perceive ourselves (Brewer, 1991; Srivastava & Beer, 2005). Outside the context of being embedded in a real-world social interaction, if you know that others generally see you as being ‘trustworthy’, you may then perceive yourself to be a trustworthy person, even though you have enough privileged information to know whether this is actually an accurate reﬂection of yourself. Conversely, the general perception we have of ourselves affects how we then perceive others (Srivastava, Guglielmo, & Beer, 2010). A person who is generally upbeat and positive is likely to judge others as positive and upbeat in a biased manner congruent with their own self-perception. Thus, how we construe both our own intrapersonal as well the interpersonal social world depends in part on a variety of these more general factors as well as factors built upon relationships we have with others. As noted earlier, much of the research on social difﬁculties in autism is inherently one-sided, focusing mainly on reading social cues from others (e.g., face-processing, eye gaze, biological motion, emotion recognition, action understanding, pragmatic aspects of language). What is needed is a more relational approach in order to reveal the deeper complexities involved in interpersonal relations. It is increasingly apparent that these basic mechanisms for related to reading social cues of others, are unlikely to account for all the variability in the real-world social difﬁculties in ASC. This is especially true for emergent properties of social interactions such as ‘relationship-effects’ that are a given outside the laboratory setting. To illustrate this point consider a recent study by Chiu et al. (2008) where participants with ASC were scanned with fMRI while playing the ‘trust game’ with another person. The ‘trust game’ is a social interaction where emergent properties of the interaction play a major role in shaping the neural mechanisms elicited in both the investor and trustee’s brain (King-Casas et al., 2006), but do not shape the same neural mechanisms when the interaction is non-social, such as when playing against a computer (Tomlin et al., 2006). Chiu and colleagues found that an area outside the traditional mentalizing system (the middle cingulate cortex; MCC) was speciﬁcally underactive when participants with ASC had to decide how much to invest in the other person. This result is important because the psychological processes involved in making this type of decision are relevant to what we do when we mentalize. In this particular scenario, participants likely think about what they want to do, but also need to take into account the consequences that their decision will have on the other person’s perception of them (what Frith and Frith (2008) dub as ‘reputation management’). Given that ‘reputation management’ must involve mentalizing, it is striking that Chiu and colleagues did not ﬁnd typical mentalizing system regions underactive in ASC. Instead, a region not typically involved in traditional mentalizing studies emerged: the MCC. This type of interaction in the trust game is a clear example of a ‘relationship-effect’ since it is an emergent property of being in a social interaction. Thus, one could say that the MCC result from the Chiu et al., study highlights a new neural mechanism for mindblindness in ASC that arises speciﬁcally from the emergent properties of being embedded in a social interaction. The approach demonstrated by Chiu and colleagues demonstrates a very different approach from the traditional way in which mentalizing or theory of mind is assessed in autism research. Past studies have typically assessed the participant making judgments about the mental states of another individual outside of the context where they are interacting with them. This traditional approach may have highlighted more general effects that are part of the mentalizing process, such as the capacity to represent mental states, but they are most likely independent of some of the ‘relationship-effects’ that are likely to emerge as a property of the social interaction itself. Thus, we take this opportunity to point out that both investigating mentalizing within and outside of social interactions are important and provide complimentary information about the mechanisms involved in mindblindness in ASC. However, this case represents an example where the mechanism for general representation of mental states in ASC may not be able to account for the deeper complexity embedded in the emergent properties that come out of interpersonal interactions. In this case, the basic idea of ‘mindblindness’ still holds, but the mechanism behind it is likely to be one that integrates the information gained from the interaction of the relationship between self and other, rather than just a mechanism that attributes a propositional attitude to an agent. A second piece of evidence that highlights the role of the MCC region highlighted by Chiu et al. (2008) involves comparing the response of this region while participants with ASC mentalize independently about themselves or another person. Despite the general premise that mindblindness extends to impairments in understanding one’s own and another’s mind, this premise has not been extensively tested within the brain and within the same paradigm in ASC. According to this premise, there should be no neural mechanisms asymmetrically involved in mentalizing more for self compared to others (or vice versa) (Carruthers, 2009). However, if self-referential processing is important for explaining some of the mechanisms involved in mindblindness in ASC, we should expect that there will be speciﬁc areas of the brain that respond atypically for mentalizing speciﬁcally about self or other. We showed that the MCC is one of these special regions. In the general population, MCC responds more when information is self-relevant (Lombardo et al., 2010a; Moran, Macrae, Heatherton, Wyland, & Kelley, 2006) but also responds more in self-relevant decisions embedded in the context of a social interaction (King-Casas et al., 2006; Tomlin et al., 2006). When asked to mentalize separately about self and other, typical participants activate MCC more when mentalizing about self than when mentalizing about other. However, individuals with ASC display the opposite pattern; MCC responded more when mentalizing about others compared to mentalizing about themselves (Lombardo et al., 2010a). Therefore, target-agnostic mechanisms involved in mindblindness irrespective of whether the target is self or other (Carruthers, 2009), cannot account for the speciﬁcity of this result. Along with the evidence from Chiu et al. (2008) this highlights that there is more complexity involved in mindblindness in ASC than just one general-purpose mechanism accounting for all deﬁcits in mentalizing about self and other.

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

133

3. Asymmetry in informational sources about self and other Beyond this idea that the relations we have with the social world considerably affects our perception of it, there are also fundamental considerations regarding differences in the kind of information we use to make inferences about self and other. The information we use for ourselves is largely introspective and interoceptive while the information available for making sense of others is largely extrospective and exteroceptive. That is, we have direct and privileged access to every sensation, emotion, and thought we have. However, when it comes to others, there is a dearth of direct information that we can collect about another’s embodied experience. Instead, our experience of others’ phenomenology is largely dominated by observing their external behaviors (Pronin, 2008). This asymmetry in informational sources plays a major role in shaping the kinds of mechanisms that are involved in how we understand our own and others’ phenomenology. To demonstrate how different kinds of inputs necessitate the need for different mechanisms, take the example of a car engine. A car that runs on diesel has one particular type of mechanism (an internal combustion engine) to deal with this input. This mechanism cannot easily handle different types of input (e.g., electric power). However, a hybrid car has two integrated mechanisms that can deal with both types of input: these being the electrical motor and an internal combustion engine as well as the computer that regulates the integration of both. In this scenario, the fuel or electrical power is the input to the system and these inputs are fundamentally different. Because they are different, they require different mechanisms to get the same output. Thus, the difference in the quality/type of input makes a difference to how efﬁciently (if at all) the operations can run (as you would sadly discover if accidentally you put unleaded petrol into your diesel car) as well determining the kinds of mechanisms needed to operate on such fundamentally different kinds of input. This example is not to say that for mentalizing information to be processed it must be handled by one special kind of mechanism for each type of input. However, given the complexity involved in performing operations on multiple types of input, one unitary mentalizing mechanism (Carruthers, 2009) or a set of other-centric or target-agnostic mentalizing mechanisms (Baron-Cohen, 1995; Leslie, 1994) are necessarily ill-equipped for handling qualitatively different types of information available for self-referential processing, such as introspective or interoceptive information. From what we know about the speciﬁc neural mechanisms involved in dealing with introspective/interoceptive compared to extroceptive/exteroceptive input (e.g., Olsson & Ochsner, 2008) it is more likely that the mentalizing system needs mechanisms that handle the different types of inputs as well as mechanisms where convergence and integration occurs across the different types of input. Because of this asymmetry in informational sources for self and other, it is necessary to pay close attention to the differences between mindblindness for self and mindblindness for other in ASC. Difﬁculty in understanding one’s own mental states might be due to differences in the unique types of information available for self-referential processing and thus the distinct mechanisms that handle such input. On the other hand, mindblindness for others may be due to mechanisms aligned with dealing with input available when perceiving others. However, a third alternative not mutually exclusive to the others, is that of a mechanism responsible for mentalizing operations on both self and other. One way to begin to understand whether the mechanisms behind mindblindness for self and other are similar or different is to examine them both within the same study. Because of this consideration about the differences in informational sources, we should not necessarily assume a priori that the mechanisms at work for understanding our own minds are the same as those working to understand other’s minds. It is likely there will be some common ground, but it is also likely there will be some important differences. Interoception is a good example to demonstrate the importance of casting a spotlight on the kind of information used to make inferences about self and other. Interoceptive information is only directly accessible to the agent experiencing it. This type of embodied information (e.g., somatosensory, visceral) is processed by a mechanism that allows the agent to become conscious or aware of its own bodily or physiological state. While important for the agent to gauge their own bodily or physiological state, the same neural mechanisms that handle this interoceptive function for ourselves also come online when we simply observe others in situations where their experience (e.g., pain or disgust) might evoke extreme somatosensory or visceral states (Craig, 2002; Singer et al., 2004; Wicker et al., 2003). Thus, accurately interpreting and using interoceptive information seems to be a basis for not only understanding our own experience, but also for empathically jumping into the experiences of others. Given these arguments, it is surprising that we do not yet know if interoceptive processing is speciﬁcally impaired in ASC or whether this mechanism is inﬂuencing the difﬁculties in empathizing with others. Although we know introspection is difﬁcult in ASC (Hill, Berthoz, & Frith, 2004; Hurlburt et al., 1994; Lombardo, Barnes, Wheelwright, & Baron-Cohen, 2007; Toichi et al., 2002) there is as yet no test of interoceptive accuracy in ASC. Individuals with ASC have alexithymic tendencies that hint at difﬁculty in interoception (Hill et al., 2004; Lombardo et al., 2007) but may simply reﬂect difﬁculties due to introspection. Supporting the idea that introspective processes may be involved is the observation that appraisal of negative emotion is different in ASC compared to controls, despite the two groups showing equivalent physiological arousal to the emotion eliciting stimuli (Ben Shalom et al., 2006). On the other hand, corticospinal excitability to viewing other’s in pain is reduced in ASC (Minio-Paluello, Baron-Cohen, Avenanti, Walsh, & Aglioti, 2009). This neurophysiological difference may be the ﬁrst step in demonstrating a disconnect between physiological or bodily state information and interoceptive manipulation of such information in an empathic way. The ambiguous nature of existing data suggests that this is an important area for future research in ASC as it may clarify the extent of difﬁculties in access to self-referential information as well as possibly highlighting critical mechanisms that may underlie difﬁculties in empathy (Bird et al., 2010; Minio-Paluello, Lombardo, Chakrabarti, Wheelwright, & Baron-Cohen, 2009; Silani et al., 2008).

134

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

4. Egocentrism We also need to pay more attention to the idea of egocentrism in ASC. While autism may seem like ‘egocentrism in the extreme’ (Asperger, 1944; Frith & de Vignemont, 2005), we must keep in mind that some degree of egocentrism itself is in fact the norm within the typical population. Social psychological research has repeatedly demonstrated the egocentric ways in which we make sense of and interpret the world (Greenwald, 1980; Krueger, 2003; Nickerson, 1999). Nowhere is egocentrism more obvious than in how it manifests when we try to predict what others think and know. When asked to estimate the general consensus on a topic, most people tend to overestimate that the consensus will conform in a congruent way with what they believe; an effect known the ‘false consensus effect’ (Krueger & Clement, 1994; Ross, Greene, & House, 1977). We also tend to overestimate how much attention is being paid to us (e.g. the ‘spotlight’ effect) (Gilovich, Medvec, & Savitsky, 2000) and how transparent we think we are in the eyes of others (the ‘illusion of transparency’) (Gilovich, Savitsky, & Medvec, 1998). Both children and adults show signs of egocentrism in social prediction. In the ‘curse of knowledge’ paradigm, people tend to overestimate the probability that naïve others will look for an object in a location that they themselves know the object to be in. Although in this paradigm it is easy to step back and consider what the naïve other person knows independently of what ‘I’ know (Birch & Bloom, 2003, 2007), even adults make egocentric errors on this paradigm or variants of it (Dumontheil, Apperly, & Blakemore, 2010; Epley, Morewedge, & Keysar, 2004b; Keysar, Lin, & Barr, 2003). These are just a few examples of egocentrism applied in the social domain. Given that some degree of egocentrism is the norm in the general population, how do we know that egocentrism in autism is an extreme version of this? While there are hints of egocentrism scattered throughout the autism literature (Lombardo and Baron-Cohen (2010), there is little work on this topic in autism with relation to standard social psychology paradigms that elicit marked egocentrism in social prediction in the general population. A more systematic examination of egocentrism in autism, especially in relation to egocentrism in social prediction, will be an important avenue for future research. 5. Simulation: the upshot of egocentrism and privileged access Although we possess privileged, yet asymmetrical informational access to embodied information and are highly prone to egocentric biases, there are clear beneﬁts for having both. Both enable us to readily use ourselves as proxies for predicting the social world. Since we have ready and direct access to our own experiences and mental life, we exploit this information to our advantage when given an impoverished environment with a lack of direct access to what others think and feel. Philosophers have coined this process simulation (Goldman, 2006; Gordon, 1992) and many accounts of ‘simulation’ (sometimes under another guise or label) can be found from philosophy (Gallese, 2001; Goldman, 2006; Gordon, 1992; Heal, 1986; Humphrey, 1984; Hurley, 2008), social cognition (Ames, 2004a, 2004b; Epley, Keysar, Van Boven, & Gilovich, 2004a; Nickerson, 1999), embodied cognition (Aziz-Zadeh & Damasio, 2008; Niedenthal, 2007; Niedenthal, Barsalou, Winkielman, Krauth-Gruber, & Ric, 2005), to neuroscience (Buckner & Carroll, 2006; Decety & Grezes, 2006; Iacoboni & Dapretto, 2006; Keysers & Gazzola, 2007; Mitchell, Banaji, & Macrae, 2005; Mitchell, Macrae, & Banaji, 2006; Rizzolatti & Craighero, 2004). Despite all the different varieties of simulation accounts, they all share one common quality: we use privileged access to our own phenomenology to gain a window of insight into the phenomenology of others. Thus, we can bypass being behaviorists (e.g., using ‘behavior rules’) or disembodied ‘theorists’ in interpreting the extrospective and exteroceptive information we get from others, by looking inward and projecting or simulating that other person as if we were them. This is not the only way we come to know the minds of others (Gopnik & Wellman, 1992), but it is one important path towards the difﬁcult problem of getting into another’s mind and experiences, especially in instances where others are perceived to be similar to oneself and/or when not much individuating information is known about the target person (Ames, 2004a, 2004b; Epley, 2008). 6. Simulation as anchoring and adjustment As the multiple varieties of simulation out there might suggest, the basic idea of using oneself as a proxy for understanding others has not been very consistent in how it is speciﬁed. Thus, an idea that may be useful here is to frame the process of simulation as both automatic, and yet effortful and controlled. An integral idea here is the general process in the judgment and decision making literature that Tversky and Kahneman dubbed the ‘anchoring and adjustment’ heuristic (Tversky & Kahneman, 1974). That is, when people make judgments under uncertainty, they start from an initial anchor value automatically set based on information that is most readily available and then go through an effortful and controlled process of serial adjustment away from that anchor in order to come up with an answer that in the end will be sufﬁcient to them. However, this process of serial adjustment away from the anchor point is usually insufﬁcient for honing in on the correct answer and is usually biased towards the initial starting values (Slovic & Lichtenstein, 1971). So what are the starting values for social judgments? In the general formulation of the anchoring and adjustment heuristic, Tversky and Kahneman (1974) noted that the anchor is usually spontaneously self-deﬁned, based on the most available information present at the time (known as the ‘availability heuristic’) (Tversky & Kahneman, 1973). Thus, as noted in the section on egocentrism as well as Meltzoff’s developmental ‘like me’ framework (Meltzoff, 2007a, 2007b), information about the self is usually the most available information and tends to be the automatic default starting value or anchor point

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

135

for social judgments. A caveat here is that self is usually the default anchor for social judgments in instances of uncertainty about the other person. The more information is known about a particular target the more it may be the case that other types of information may be more readily available than just what we know of ourselves (Epley, 2008). Using self as a habitual reference point in the social domain can be exempliﬁed in studies that ﬁnd asymmetries in similarity judgments depending on how the similarity statement is linguistically speciﬁed. When the self is speciﬁed as the referent and another person as the subject (e.g., ‘How similar are you to X?’), similarity ratings are much higher than when the other person is the referent and the self is the subject (e.g., ‘How similar is X to you?’) (Srull & Gaelick, 1983). The explanation behind this is that the ‘self’ is a supraordinate object of knowledge, whilst others are a subordinate object of knowledge. Comparing a supraordinate object to a subordinate object of knowledge will render more matches than making the comparison in the other direction (see Fig. 1). However, rather than this simply occurring because of linguistic convention, further manipulations can be performed to show that the self is indeed a habitual reference point in the social domain. Catrambone, Beike, and Niedenthal (1996) replicated the asymmetry in similarity judgments mentioned above. However, they also included a manipulation where the referent and subject were not forced upon the participant based on linguistic conventions. By simply asking people ‘How similar are these two people in general?’ and then offering self and the other person under this statement in a vertical arrangement, they found that these non-forced similarity judgments rendered similarity ratings identical to those where the self is placed as the referent. Thus, this simple set of studies illustrates the frequent observation throughout the egocentrism literature, that people automatically use themselves as the habitual reference point in social judgments (Alicke, 1993; Clement & Krueger, 2000; Dunning & Cohen, 1992; Dunning, Meyerowitz, & Holzberg, 1989; Meltzoff & Brooks, 2008; Nickerson, 1999). Furthermore, as the ‘spotlight’ effect and the ‘illusion of transparency’ demonstrate, people use their own phenomenology when estimating how they will be evaluated by others (Chambers, Epley, Savitsky, & Windschitl, 2008; Epley, Savitsky, & Gilovich, 2002; Gilovich et al., 1998, 2000). This is not particularly surprising given that information for self typically rests on a wealth of access to one’s own phenomenology, but observing others rests predominantly on externally observable (e.g., behavioral) information (Pronin, 2008). Although we automatically use ourselves as a plausible anchor or reference point to facilitate social inference, the second hurdle of adjustment from our anchor point is essential for accurate social inference. This adjustment process, unlike the automatic nature of anchoring on oneself, is effortful and controlled. Because adjustment requires effort and control, it is highly susceptible to disruption and thus, becomes insufﬁcient. In the context of social judgment, when adjustments are insufﬁcient they are biased toward the starting value; ourselves. Epley and Gilovich (2006) suggest that adjustments are usually insufﬁcient because this process terminates when we settle on a plausible (though not always correct) value, and because in most contexts individuals are not motivated enough to go through the extra search for more information that will provide evidence for further adjustment towards an accurate estimate. However, give an individual a ﬁnancial incentive

A

B

Fig. 1. Self-as-a-supraordinate and Other-as-a-subordinate object of knowledge. This ﬁgure illustrates Self and Other as circles. The size of the circle depicts the amount of knowledge one has about each. When Self is compared to Other, the overlap (represented by the area of the shadow on the subject) is larger than when Other is compared to Self. This is used as an explanation for the phenomenon that similarity between self and other is rated as higher when Self is the ‘referent’ and Other is the ‘subject’ (e.g., ‘How similar are you to hOtheri?’) compared to when Other is the ‘referent’ and Self is the ‘subject’ (e.g., ‘How similar is hOtheri to you?’) (Srull & Gaelick, 1983). When people are not forced to make the similarity ratings in a manner where referent and subject are pre-speciﬁed linguistically (e.g., ‘How similar are these two people?’), people tend to default by placing the Self as the ‘referent’ because similarity ratings in this case are identical to when Self is designated as the ‘referent’ (Catrambone et al., 1996).

136

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

to be accurate and adjustments tend to become more accurate. Given these ideas that adjustment is effortful and controlled and that insufﬁcient adjustment will lead to judgments closer to self-referential anchors, one way to determine if people use anchoring and adjustment is to intervene at the adjustment stage. The prediction here is that manipulations that disrupt serial adjustment away from self-referential anchor points will lead to increasingly egocentric biases in social prediction. However, manipulations that motivate the participant to continue the adjustment process (e.g., ﬁnancial incentives) should lead to less egocentric bias. Through a series of studies Epley et al. (2004a) were able to show that systematic manipulations to intervene in the adjustment stage can systematically alter how much of an egocentric bias in social judgment is expressed. Across four studies, Epley and colleagues were able to ﬁrst demonstrate egocentric biases in social prediction and that individuals serially adjust from their self-referential starting point. Furthermore, manipulations that disrupted the effortful and controlled adjustment process led to more pronounced egocentric biases, while ﬁnancial incentive manipulations led participants to adjust towards more accurate social predictions. The most dramatic result in this set of studies was that people who reported hearing backward masked messages in music (e.g., hearing ‘‘Its fun to smoke marijuana” while listening to the Queen song Another One Bites the Dust played backwards) would egocentrically predict that naïve others would also hear the backward masked message, and that adjustment processes away from this egocentric anchor could be disrupted simply by answering questions about what other’s would hear while nodding their head (an implicit act of acceptance). The idea here is that people start from the egocentric anchor that others will be like me and hear the backward masked message, and head nodding leads people to accept values earlier in the adjustment process (Wells & Petty, 1980), thus leading to more pronounced egocentrism in predicting whether others could also hear the message. Given the idea that simulation is a key process for how individuals gain insight into the social world, it is striking that there is such a dearth of work on simulation2 in the context of mentalizing research in autism. This is perhaps the case since the trend has typically been to study other-referential or self-referential processing separately, rather than within the same study. However, recent studies consistently show that individuals with autism are impaired in both self-referential and social-cognitive processing (Lombardo et al., 2007; Williams & Happe, 2009, 2010), suggesting at the very least, that some of the same mechanisms are involved in both self-referential and social cognition, and perhaps implicating that simulation approaches may be worthwhile in testing for common mechanisms underlying self-referential and social deﬁcits in autism. Our recent work tied together a variety of self-referential and social-cognitive measures and found that individuals with ASC have concurrent and related impairments in both domains (Lombardo et al., 2007). We followed this behavioral study with an fMRI study to investigate whether there were any common neural mechanisms involved in mindblindness for self and other. In the general population, a distributed network of brain regions is recruited for both mentalizing about self and other (Lombardo et al., 2010b). One of these shared mentalizing regions known to be speciﬁcally involved in the general representation of mental states, the right temporo-parietal junction (RTPJ) (Saxe & Kanwisher, 2003; Saxe & Powell, 2006), was hypoactive in ASC across both mentalizing about self and other. Furthermore, the magnitude of early childhood social impairments was related to decreased activation of this region across both mentalizing about self and other (Lombardo, Chakrabarti, Bullmore, MRC AIMS Consortium, & Baron-Cohen, in preparation). Thus, the premise that mindblindness in ASC may stem from similar mechanisms deployed for self and other seems to be conﬁrmed. This study clariﬁes that this mechanism may be involved speciﬁcally with the representation of mental states itself. However, as we argue in this paper, there are likely to be considerably more mechanisms that contribute to mindblindness for self and other, and some may affect self, other, or the unique interaction of the two within a social interaction. 7. Perceived similarity and distinguishing between self and other The ﬁnal notion we will underscore as important for considering the role of the self in mindblindness in ASC is the idea of whether others are perceived as similar to or different from self. Perceived similarity between self and other is one of the primary factors in determining how much adjustment away from self-referential anchors will be required in order to successfully simulate the experience of others. While one may want to anchor on self for others who are similar to oneself, they will not want to engage in much adjustment away from the self-referential anchor point. Alternatively, those who are perceived to be dissimilar to self will likely be those where adjustment processes may become more important. Ames highlighted this notion in a series of two papers where he describes his ‘similarity contingency model’ of social inference (Ames, 2004a, 2004b). In this model he describes two forms of social inference: projection (i.e. ‘simulation’) and stereotyping. His prediction was that the deployment of such forms of social inference would depend on the perceived similarity of the target to oneself. In cases where one is similar to self, projection would be deployed, while stereotyping would be deployed for dissimilar others. Across a range of different manipulations, Ames showed that indeed, the more similar perceivers are to target others, the more the perceiver’s own attitudes and opinions are projected onto the target other. However, when the target is dissimilar, perceivers engage more in stereotyping processes. Ames’ results are largely congruent with other ﬁndings in the false consensus literature which highlight that social categorization signiﬁcantly

2 We do not discuss areas of research on ‘simulation’ such as mirror neurons as this extends discussion into topics such as action understanding and while relevant to discussing its relation to the social deﬁcits in autism, is a bit beyond the scope of discussing ‘simulation’ speciﬁcally within the context of mindblindness in autism.

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

137

mediates the degree to which the false consensus effect occurs (Clement & Krueger, 2000). The take home message of Ames’ studies is that while egocentrism creates many biases in social perception, it also has the upshot of facilitating an understanding of others, depending on how well one serves as a proxy for others. In the scope of the anchoring and adjustment perspective, Ames’ work demonstrates that when one is similar to oneself, the adjustment process is much easier than when adjusting far from the anchor point for dissimilar others. The deployment of other heuristics for dissimilar others, such as more rule-based stereotyping, is important to highlight since this may be a shortcut around the effortful process of serial adjustment away from self-referential anchors and highlights that anchoring and adjustment may be part of a larger toolkit people use when making sense of the social world. Perceiving similarities and differences between self and others relies on speciﬁc neural mechanisms. fMRI studies have consistently demonstrated that one area of the brain crucial for making such a distinction is the ventromedial prefrontal cortex (vMPFC). Activity in vMPFC increases in linear fashion when information is perceived to be increasingly self-relevant (Moran, Heatherton, & Kelley, 2009; Moran et al., 2006) and thus responds on average more when thinking about oneself compared to others (Kelley et al., 2002). When individuals with autism reﬂect on themselves or others, vMPFC responds atypically, in an egocentrically equivalent fashion for both self and other (Kennedy & Courchesne, 2008; Lombardo et al., 2010a). This lack of a neural self-other distinction mirrors behavioral evidence that individuals with ASC show markers of atypical self-other distinctions (Lee & Hobson, 2006; Lee, Hobson, & Chiat, 1994; Mitchell & O’Keefe, 2008). Furthermore, underscoring the idea that mechanisms for distinguishing self from other are important in the social domain, we found that the degree to which vMPFC responds selectively for self compared to other was predicted by the degree of early childhood social impairment, such that individuals whose vMPFC responded most egocentrically to the mental characteristics of self and other were the most socially impaired in early childhood (see Fig. 2) (Lombardo et al., 2010a). Because perceiving similarities and differences between self and other are integral in simulation aspects of mindreading, this insight further illustrates the importance of taking the self into account when addressing mindblindness and the hallmark social difﬁculties in ASC.

Fig. 2. Atypical neural self-other distinction in ASC and its relationship to social impairment (Taken from Lombardo et al. (2010a)). (A) Results from a quantitative meta-analysis highlighting regions of the brain consistently recruited more self-referential cognition compared to thinking about others. Ventromedial prefrontal cortex (vMPFC) is the maximal peak from this meta-analysis. (B) While vMPFC responds more to thinking about self compared to others in healthy Controls, it responds equivalently to self and other in ASC. (C) Magnitude of self-other distinction expressed by vMPFC in the mentalizing domain (red) but not the physical domain (blue) correlates with social impairment in ASC. Those individuals whose vMPFC made the largest self-other distinction were least socially impaired, while those whose vMPFC made little to no distinction between self and other were most socially impaired. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

138

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

In conclusion, we have provided several arguments that illustrate why it is important to consider the role of the self in mindblindness. Integrating knowledge from social psychology is an important ﬁrst step in this topic, as many considerations about the self discussed here may play an important role in mentalizing in ASC. However, new ideas raise many new questions. We have highlighted some of these new directions as well as some recent research that begins to take this new approach of integrating the self into social cognition in ASC. Further work along these lines will yield not only new insights into mindblindness and social-communication difﬁculties in ASC, but also inform us as to how critical this topic is in typical social development. In the clinical domain, paying further attention to mechanisms involved in mindblindness and how they are related to self-referential processing may provide new ways of facilitating both self-referential and social-communicative deﬁcits in ASC. References Abell, F., Happe, F. G., & Frith, U. (2000). Do triangles play tricks? Attribution of mental states to animated shapes in normal and abnormal development. Cognitive Development, 15, 1–16. Alicke, M. D. (1993). Egocentric standards of conduct evaluation. Basic and Applied Social Psychology, 14, 171–192. Ames, D. R. (2004a). Inside the mind reader’s tool kit: Projection and stereotyping in mental state inference. Journal of Personality and Social Psychology, 87, 340–353. Ames, D. R. (2004b). Strategies for social inference: A similarity contingency model of projection and stereotyping in attribute prevalence estimates. Journal of Personality and Social Psychology, 87, 573–585. Andersen, S. M., & Chen, S. (2002). The relational self: An interpersonal social-cognitive theory. Psychological Review, 109, 619–645. Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63, 596–612. Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships as including other in the self. Journal of Personality and Social Psychology, 60, 241–253. Asperger, H. (1944). Die autistischen Psychopathen im Kindesalter. Archiv fur Psychiatrie und Nervenkrankheiten, 117, 76–136. Aziz-Zadeh, L., & Damasio, A. (2008). Embodied semantics for actions: Findings from functional brain imaging. Journal of Physiology Paris, 102, 35–39. Baron-Cohen, S. (1989). Are autistic children ‘‘behaviorists”? An examination of their mental–physical and appearance-reality distinctions. Journal of Autism and Developmental Disorders, 19, 579–600. Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘‘theory of mind”? Cognition, 21, 37–46. Behrens, T. E., Hunt, L. T., & Rushworth, M. F. (2009). The computation of social behavior. Science, 324, 1160–1164. Behrens, T. E., Hunt, L. T., Woolrich, M. W., & Rushworth, M. F. (2008). Associative learning of social value. Nature, 456, 245–249. Ben Shalom, D., Mostofsky, S. H., Hazlett, R. L., Goldberg, M. C., Landa, R. J., Faran, Y., et al (2006). Normal physiological emotions but differences in expression of conscious feelings in children with high-functioning autism. Journal of Autism and Developmental Disorders, 36, 395–400. Birch, S. A., & Bloom, P. (2003). Children are cursed: An asymmetric bias in mental-state attribution. Psychological Science, 14, 283–286. Birch, S. A., & Bloom, P. (2007). The curse of knowledge in reasoning about false beliefs. Psychological Science, 18, 382–386. Bird, G., Silani, G., Brindley, R., White, S., Frith, U., & Singer, T. (2010). Empathic brain responses in insula are modulated by levels of alexithymia but not autism. Brain, 133, 1515–1525. Brewer, M. B. (1991). The social self: On being the same and different at the same time. Personality and Social Psychology Bulletin, 17, 475–482. Buckner, R. L., & Carroll, D. C. (2006). Self-projection and the brain. Trends in Cognitive Sciences, 11, 49–57. Carruthers, P. (2009). How we know our own minds: The relationship between mindreading and metacognition. Behavioral and Brain Sciences, 32, 121–138. Catrambone, R., Beike, D., & Niedenthal, P. (1996). Is the self-concept a habitual referent in judgments of similarity? Psychological Science, 7, 158–183. Chambers, J. R., Epley, N., Savitsky, K., & Windschitl, P. D. (2008). Knowing too much: Using private knowledge to predict how one is viewed by others. Psychological Science, 19, 542–548. Charman, T., & Baron-Cohen, S. (1992). Understanding drawings and beliefs: A further test of the metarepresentation theory of autism: A research note. Journal of Child Psychology and Psychiatry, 33, 1105–1112. Chiu, P. H., Kayali, M. A., Kishida, K. T., Tomlin, D., Klinger, L. G., Klinger, M. R., et al (2008). Self responses along cingulate cortex reveal quantitative neural phenotype for high-functioning autism. Neuron, 57, 463–473. Clement, R. W., & Krueger, J. (2000). The primacy of self-referent information in perceptions of social consensus. British Journal of Social Psychology, 39, 279–299. Cooley, C. H. (1902). Human nature and the social order. New York: Scribner. Craig, A. D. (2002). How do you feel? Interception: The sense of the physiological condition of the body. Nature Reviews Neuroscience, 3, 655–666. Decety, J., & Grezes, J. (2006). The power of simulation: Imagining one’s own and other’s behavior. Brain Research, 1079, 4–14. Dumontheil, I., Apperly, I. A., & Blakemore, S. J. (2010). Online usage of theory of mind continues to develop in late adolescence. Developmental Science, 13, 331–338. Dunning, D. A., & Cohen, G. L. (1992). Egocentric deﬁnitions of traits and abilities in social judgment. Journal of Personality and Social Psychology, 63, 341–355. Dunning, D. A., Meyerowitz, J. A., & Holzberg, A. D. (1989). Ambiguity and self-evaluation: The role of idiosyncratic trait deﬁnitions in self-serving appraisals of ability. Journal of Personality and Social Psychology, 57, 1082–1090. Epley, N. (2008). Solving the (real) other minds problem. Social and Personality Psychology Compass, 2(3), 1455–1474. Epley, N., & Gilovich, T. (2006). The anchoring-and-adjustment heuristic: Why the adjustments are insufﬁcient. Psychological Science, 17, 311–318. Epley, N., Keysar, B., Van Boven, L., & Gilovich, T. (2004a). Perspective taking as egocentric anchoring and adjustment. Journal of Personality and Social Psychology, 87, 327–339. Epley, N., Morewedge, C. K., & Keysar, B. (2004b). Perspective taking in children and adults: Equivalent egocentrism but differential correction. Journal of Experimental Social Psychology, 40, 760–768. Epley, N., Savitsky, K., & Gilovich, T. (2002). Empathy neglect: Reconciling the spotlight effect and the correspondence bias. Journal of Personality and Social Psychology, 83, 300–312. Frith, U. (2001). Mind blindness and the brain in autism. Neuron, 32, 969–979. Frith, U. (2003). Autism: Explaining the enigma (2nd ed.). Malden, MA: Blackwell. Frith, U., & de Vignemont, F. (2005). Egocentrism, allocentrism, and Asperger syndrome. Consciousness and Cognition, 14, 719–738. Frith, C. D., & Frith, U. (2008). The self and its reputation in autism. Neuron, 57, 331–332. Frith, U., & Happe, F. (1999). Theory of mind and self-consciousness: What it is like to be autistic. Mind and Language, 14, 1–22. Gallese, V. (2001). The ‘shared manifold’ hypothesis: From mirror neurons to empathy. Journal of Consciousness Studies, 8, 33–50. Gilovich, T., Medvec, V. H., & Savitsky, K. (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one’s own actions and appearance. Journal of Personality and Social Psychology, 78, 211–222. Gilovich, T., Savitsky, K., & Medvec, V. H. (1998). The illusion of transparency: Biased assessments of others’ ability to read one’s emotional states. Journal of Personality and Social Psychology, 75, 332–346. Goldberg, L. R. (1990). An alternative ‘‘description of personality”: The Big-Five factor structure. Journal of Personality and Social Psychology, 59, 1216–1229.

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

139

Goldman, A. (2006). Simulating minds: The philosophy, psychology, and neuroscience of mind reading. New York: Oxford University Press. Gopnik, A., & Wellman, H. (1992). Why the child’s theory of mind really is just a theory. Mind and Language, 7, 145–171. Gordon, R. M. (1992). The simulation theory: Objections and misconceptions. Mind and Language, 7, 11–34. Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision of personal history. American Psychologist, 35, 603–618. Greenwald, A. G., & Pratkanis, A. R. (1984). The self. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cognition. Hillsdale, NJ: Erlbaum. Hamilton, A. (2009). Research review: Goals, intentions and mental states: Challenges for theories of autism. Journal of Child Psychology and Psychiatry, 50, 881–892. Hampton, A. N., Bossaerts, P., & O’Doherty, J. P. (2008). Neural correlates of mentalizing-related computations during strategic interactions in humans. Proceedings of the National Academy of Sciences USA, 105, 6741–6746. Happe, F. G. (1995). The role of age and verbal ability in the theory of mind task performance of subjects with autism. Child Development, 66, 843–855. Happe, F. (2003). Theory of mind and the self. Annals of the New York Academy of Sciences, 1001, 134–144. Heal, J. (1986). Simulation, theory, and content. In P. Carruthers & P. K. Smith (Eds.), Theories of theories of mind (pp. 75–89). Cambridge, UK: Cambridge University Press. Hill, E., Berthoz, S., & Frith, U. (2004). Brief report: Cognitive processing of own emotions in individuals with autistic spectrum disorder and in their relatives. Journal of Autism and Developmental Disorders, 34, 229–235. Hobson, R. P. (1990). On the origins of self and the case of autism. Development and Psychopathology, 2, 163–181. Hobson, P. R., Chidambi, G., Lee, A., & Meyer, J. (2006). Foundations for self-awareness: An exploration through autism. Monographs of the Society for Research in Child Development, 71, vii–166. Hobson, R. P., & Meyer, J. A. (2005). Foundations for self and other: A study in autism. Developmental Science, 8, 481–491. Humphrey, N. (1984). Consciousness regained. Oxford: Oxford University Press. Hurlburt, R. T., Happe, F., & Frith, U. (1994). Sampling the form of inner experience in three adults with Asperger syndrome. Psychological Medicine, 24, 385–395. Hurley, S. (2008). The shared circuits model (SCM): How control, mirroring, and simulation can enable imitation, deliberation, and mindreading. Behavioral and Brain Sciences, 31, 1–22. Iacoboni, M., & Dapretto, M. (2006). The mirror neuron system and the consequences of its dysfunction. Nature Reviews Neuroscience, 7, 942–951. James, W. (1890). The principles of psychology. New York: Holt. Kanner, L. (1943). Autistic disturbance of affective contact. Nervous Child, 2, 217–250. Kelley, W. M., Macrae, C. N., Wyland, C. L., Caglar, S., Inati, S., & Heatherton, T. F. (2002). Finding the self? An event-related fMRI study. Journal of Cognitive Neuroscience, 14, 785–794. Kennedy, D. P., & Courchesne, E. (2008). Functional abnormalities of the default network during self- and other-reﬂection in autism. Social Cognitive and Affective Neuroscience, 3, 177–190. Kenny, D. A. (1994). Interpersonal perception: A social relations analysis. New York: Guilford Press. Kenny, D. A., & La Voie, L. (1984). The social relations model. Advances in Experimental Social Psychology, 18, 141–182. Keysar, B., Lin, S., & Barr, D. J. (2003). Limits on theory of mind use in adults. Cognition, 89, 25–41. Keysers, C., & Gazzola, V. (2007). Integrating simulation and theory of mind: From self to social cognition. Trends in Cognitive Sciences, 11, 194–196. King-Casas, B., Tomlin, D., Anen, C., Camerer, C. F., Quartz, S. R., & Montague, P. R. (2006). Getting to know you: Reputation and trust in a two-person economic exchange. Science, 308, 78–83. Klin, A. (2000). Attributing social meaning to ambiguous visual stimuli in higher-functioning autism and Asperger syndrome: The social attribution task. Journal of Child Psychology and Psychiatry, 41, 831–846. Krueger, J. I. (2003). Return of the ego-self-referent information as a ﬁlter for social prediction: Comment on Karniol (2003). Psychological Review, 110, 585–590. Krueger, J. I., & Clement, R. W. (1994). The truly false consensus effect: An ineradicable and egocentric bias in social perception. Journal of Personality and Social Psychology, 67, 596–610. Lee, A., & Hobson, P. R. (2006). Drawing self and others: How do children with autism differ from those with learning difﬁculties. British Journal of Developmental Psychology, 24, 547–565. Lee, A., Hobson, R. P., & Chiat, S. (1994). I, you, me, and autism: An experimental study. Journal of Autism and Developmental Disorders, 24, 155–176. Leslie, A. M. (1994). ToMM, ToBy, and Agency: Core architecture and domain speciﬁcity. In L. A. Hirschfeld & S. Gelman (Eds.), Mapping the mind: Domain speciﬁcity in cognition and culture (pp. 119–148). New York: Cambridge University Press. Leslie, A. M., & Thaiss, L. (1992). Domain speciﬁcity in conceptual development: Neuropsychological evidence from autism. Cognition, 43, 225–251. Lombardo, M. V., Barnes, J. L., Wheelwright, S. J., & Baron-Cohen, S. (2007). Self-referential cognition and empathy in autism. PLoS ONE, 2, e883. Lombardo, M. V., & Baron-Cohen, S. (2010). Unraveling the paradox of the autistic self. WIREs Cognitive Science, 1, 393–403. Lombardo, M. V., Chakrabarti, B., Bullmore, E. T., Sadek, S. A., Wheelwright, S. J., Pasco, G., et al (2010a). Atypical neural self-representation in autism. Brain, 133, 611–624. Lombardo, M. V., Chakrabarti, B., Bullmore, E. T., Wheelwright, S. J., Sadek, S. A., Suckling, J., et al (2010b). Shared neural circuits for mentalizing about the self and others. Journal of Cognitive Neuroscience, 22, 1623–1635. Lombardo, M. V., Chakrabarti, B., Bullmore, E. T., MRC AIMS Consortium, & Baron-Cohen, S. (in preparation). Specialization of right temporo-parietal junction for mentalizing and its relation to social impairments in autism. Mead, G. H. (1934). Mind, self and society. Chicago: University of Chicago Press. Meltzoff, A. N. (2007a). ‘Like me’: A foundation for social cognition. Developmental Science, 10, 126–134. Meltzoff, A. N. (2007b). The ‘like me’ framework for recognizing and becoming an intentional agent. Acta Psychologica, 124, 26–43. Meltzoff, A. N., & Brooks, R. (2008). Self-experience as a mechanism for learning about others: A training study in social cognition. Developmental Psychology, 44, 1257–1265. Minio-Paluello, I., Baron-Cohen, S., Avenanti, A., Walsh, V., & Aglioti, S. M. (2009). Absence of embodied empathy during pain observation in Asperger syndrome. Biological Psychiatry, 65, 55–62. Minio-Paluello, I., Lombardo, M. V., Chakrabarti, B., Wheelwright, S., & Baron-Cohen, S. (2009). Response to Smith’s letter to the editor ‘Emotional Empathy in Autism Spectrum Conditions: Weak, Intact, or Heightened?’. Journal of Autism and Developmental Disorders, 39, 1749–1754. Mitchell, J. P., Banaji, M. R., & Macrae, C. N. (2005). The link between social cognition and self-referential thought in the medial prefrontal cortex. Journal of Cognitive Neuroscience, 17, 1306–1315. Mitchell, J. P., Macrae, C. N., & Banaji, M. R. (2006). Dissociable medial prefrontal contributions to judgments of similar and dissimilar others. Neuron, 50, 655–663. Mitchell, P., & O’Keefe, K. (2008). Brief report: Do individuals with autism spectrum disorder think they know their own minds? Journal of Autism and Developmental Disorders, 38, 1591–1597. Moran, J. M., Heatherton, T. F., & Kelley, W. M. (2009). Modulation of cortical midline structures by implicit and explicit self-relevance evaluation. Social Neuroscience, 4, 197–211. Moran, J. M., Macrae, C. N., Heatherton, T. F., Wyland, C. L., & Kelley, W. M. (2006). Neuroanatomical evidence for distinct cognitive and affective components of self. Journal of Cognitive Neuroscience, 18, 1586–1594. Nichols, S., & Stich, S. (2003). Mindreading: An integrated account of pretence, self-awareness, and understanding other minds. New York: Oxford University Press. Nickerson, R. S. (1999). How we know – and sometimes misjudge – what others know: Imputing one’s own knowledge to others. Psychological Bulletin, 125, 737–759.

140

M.V. Lombardo, S. Baron-Cohen / Consciousness and Cognition 20 (2011) 130–140

Niedenthal, P. M. (2007). Embodying emotion. Science, 316, 1002–1005. Niedenthal, P. M., Barsalou, L. W., Winkielman, P., Krauth-Gruber, S., & Ric, F. (2005). Embodiment in attitudes, social perception, and emotion. Personality and Social Psychology Review, 9, 184–211. Olsson, A., & Ochsner, K. N. (2008). The role of social cognition in emotion. Trends in Cognitive Sciences, 12, 65–71. 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, 688–700. Pronin, E. (2008). How we see ourselves and how we see others. Science, 320, 1177–1180. Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Reviews of Neuroscience, 27, 169–192. Ross, L., Greene, D., & House, P. (1977). The false consensus effect: An egocentric bias in social perception and attribution processes. Journal of Experimental Social Psychology, 13, 279–301. Saxe, R., & Kanwisher, N. (2003). People thinking about thinking people. The role of the temporo-parietal junction in ‘‘theory of mind”. Neuroimage, 19, 1835–1842. Saxe, R., & Powell, L. J. (2006). It’s the thought that counts: Speciﬁc brain regions for one component of theory of mind. Psychological Science, 17, 692–699. Senju, A., Southgate, V., Miura, Y., Matsui, T., Hasegawa, T., Tojo, Y., et al (2010). Absence of spontaneous action anticipation by false belief attribution in children with autism spectrum disorder. Development and Psychopathology, 22, 353–360. Senju, A., Southgate, V., White, S., & Frith, U. (2009). Mindblind eyes: An absence of spontaneous theory of mind in Asperger syndrome. Science, 325, 883–885. Silani, G., Bird, G., Brindley, R., Singer, T., Frith, C., & Frith, U. (2008). Levels of emotional awareness and autism: An fMRI study. Social Neuroscience, 3, 97–110. Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R. J., & Frith, C. D. (2004). Empathy for pain involves the affective but not sensory components of pain. Science, 303, 1157–1162. Slovic, P., & Lichtenstein, S. (1971). Comparison of Bayesian and regression approaches to the study of information processing in judgment. Organizational Behavior and Human Performance, 6, 649–744. Srivastava, S., & Beer, J. S. (2005). How self-evaluations relate to being liked by others: Integrating sociometer and attachment perspectives. Journal of Personality and Social Psychology, 89, 966–977. Srivastava, S., Guglielmo, S., & Beer, J. S. (2010). Perceiving others’ personalities: Examining the dimensionality, assumed similarity to the self, and stability of perceiver effects. Journal of Personality and Social Psychology, 98, 520–534. Srull, T. K., & Gaelick, L. (1983). General principles and individual differences in the self as a habitual reference point: An examination of self-other judgments of similarity. Social Cognition, 2, 108–121. Toichi, M., Kamio, Y., Okada, T., Sakihama, M., Youngstrom, E. A., Findling, R. L., et al (2002). A lack of self-consciousness in autism. American Journal of Psychiatry, 159, 1422–1424. Tomlin, D., Kayali, M. A., King-Casas, B., Anen, C., Camerer, C. F., Quartz, S. R., et al (2006). Agent-speciﬁc responses in the cingulate cortex during economic exchanges. Science, 312, 1047–1050. Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5, 207–232. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185, 1124–1130. Wells, G. L., & Petty, R. E. (1980). The effects of overt head movements on persuasion: Compatibility and incompatibility of responses. Basic and Applied Social Psychology, 1, 219–230. Wicker, B., Keysers, C., Plailly, J., Royet, J. P., Gallese, V., & Rizzolatti, G. (2003). Both of us disgusted in my insula: The common neural basis of seeing and feeling disgust. Neuron, 40, 655–664. Williams, D. M. (2010). Theory of own mind in autism: Evidence for a speciﬁc deﬁcit in self-awareness. Autism. Williams, D. M., & Happe, F. (2009). What did I say? Versus what did I think? Attributing false beliefs to self amongst children with and without autism. Journal of Autism and Developmental Disorders, 39, 865–873. Williams, D., & Happe, F. (2010). Recognising ‘social’ and ‘non-social’ emotions in self and others: A study of autism. Autism, 14, 285–304.

Consciousness and Cognition 20 (2011) 141–148

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Faces and ascriptions: Mapping measures of the self q Dan Zahavi a,⇑, Andreas Roepstorff b a b

Center for Subjectivity Research, Department of Media, Cognition and Communication, University of Copenhagen, Denmark Department of Social Anthropology & Center for Functionally Integrative Neuroscience, Aarhus University, Denmark

a r t i c l e

i n f o

Article history: Available online 12 November 2010 Keywords: Self Brain imaging Adjective ascription task Mirror self-recognition Experimental psychology History of psychology (science) Philosophy Interdisciplinary research Face recognition Self-awareness

a b s t r a c t The ‘self’ is increasingly used as a variable in cognitive experiments and correlated with activity in particular areas in the brain. At ﬁrst glance, this seems to transform the self from an ephemeral theoretical entity to something concrete and measurable. However, the transformation is by no means unproblematic. We trace the development of two important experimental paradigms in the study of the self, self-face recognition and the adjective self ascription task. We show how the experimental instrumentalization has gone hand in hand with a simpliﬁcation of the self-concept, and how more conceptual and theoretical reﬂections on the structure, function and nature of self have either disappeared altogether or receded into the background. We argue that this development impedes and complicates the interdisciplinary study of self. Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction: conﬂicting perspectives on self The nature, structure and reality of self have been discussed by philosophers for millennia. In the current debate, a form of self-skepticism is not uncommon. As Thomas Metzinger puts it at the very beginning of his 2003 book Being No One: ‘‘...no such things as selves exist in the world. Nobody ever was or had a self’’ (Metzinger, 2003, p. 1). In defending this view, Metzinger, however, endorses a rather reiﬁed notion of self. If it exists, the self must be a non-physical soul-substance. Metzinger denies the existence of such an unchangeable and ontologically independent entity, and therefore argues that the self is illusory (Metzinger, 2003, pp. 370, 385, 390). Metzinger’s conclusion, however, is only warranted if his deﬁnition of self is the only one available, but that is hardly the case. Quite to the contrary, in fact, since most empirical researchers who currently investigate the development, structure, function and pathology of self, usually employ and operate with quite different notions of self. If one thought that only philosophers would be interested in investigating the nature and existence of self and that scientists would stay away from a topic as elusive as this, one is consequently bound to be surprised. If anything, recent years have witnessed a dramatic increase of interest in self in disciplines as various as cognitive science, developmental psychology, sociology, neuropsychology and psychiatry. Consider, for instance, Neisser’s already classical distinction between the ecological, the interpersonal, the extended, private and conceptual self (Neisser, 1988). Consider the interest in self found in emotion research. As a recent text book puts it: ‘‘One cannot study self-conscious emotions without trying to conceptualize the self and its many levels and its role in the generation of emotions’’ (Campos, 2007, p. xi). Consider, for a ﬁnal example among many, how current work on pathologies as diverse as schizophrenia and dementia refer to and discuss

q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap Special Issue: T. Feinberg.

⇑ Corresponding author. Address: Center for Subjectivity Research, University of Copenhagen, Njalsgade 140-142, 5th ﬂoor, DK-2300 Copenhagen S, Denmark. Fax: +45 35328681. E-mail address: [email protected] (D. Zahavi). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.10.011

142

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

the topic of self. As Seeley and Miller wrote in 2005: ‘‘Though once relegated to philosophers and mystics, the structure of the self may soon become mandatory reading for neurology, psychiatry, and neuroscience trainees. For the dementia specialist the need for this evolution is transparent, as shattered selves – of one form or another – remain a daily part of clinical practice’’ (Seeley & Miller, 2005, p. 160). One might wonder how Seeley and Miller would react to Metzinger’s claim that there never has existed anybody who was or had a self. But what should one conclude from this discrepancy between the perspectives on self found in contemporary philosophy and empirical research? If we reconsider Seeley and Miller’s assertion it can obviously be interpreted in two very different ways. Is the idea that empirical researchers should become familiar with philosophical discussions of self, since the latter are of relevance for the empirical research, or is the idea rather that empirical researchers should take on the task of analyzing and explaining the self themselves? It is certainly not difﬁcult to ﬁnd vocal representatives of the latter view. Crick, for instance, has argued that it is hopeless to try to solve the problems of self and consciousness by general philosophical arguments. In his view, what we need are suggestions for new experiments that might clarify and ultimately solve these problems (Crick, 1995, p. 19). Indeed, on Crick’s view, ‘‘the study of consciousness is a scientiﬁc problem. [. . .] There is no justiﬁcation for the view that only philosophers can deal with it’’ (Crick, 1995, p. 258). Quite on the contrary, in fact, since philosophers ‘‘have had such a poor record over the last two thousand years that they would do better to show a certain modesty rather than the lofty superiority that they usually display’’ (Crick, 1995, p. 258). Crick’s somewhat polemical assertions do raise a fundamental question. What is the right way to conceive of the relationship between philosophical analysis and empirical investigation when it comes to the study of self? One tempting reply might be that the task left to philosophy is to pose the questions and that empirical science can then provide the answers. This cannot quite be right, however, if only because the alleged empirical answers, as we shall soon see, are often in urgent need of conceptual clariﬁcation. Not surprisingly, one of the main challenges to a neuroscientiﬁc investigation of self has been to identify and locate the neural correlate of self. In a survey article entitled ‘‘Is self special: a critical review of evidence from experimental psychology and cognitive neuroscience’’ which was published in Psychological Bulletin in 2005, Gillihan and Farah, two neuroscientists, discussed the different suggestions that neuroscience had recently been offering. Their conclusion was somewhat discouraging in that different researchers had pointed to quite different areas in the brain. What is the reason for this lack of consensus? There are several different ones. But let us focus on a single, which is of particular relevance in this context. When one reads research on self written by neuroscientists much effort is usually spent on explaining the experimental setup and on discussing and interpreting the experimental results. Much less time is devoted to discussing and clarifying the very notion of self at work. For one example among many, consider a 2005 piece by BaronCohen, where he writes: ‘‘I do not tackle the thorny question of how to deﬁne the self [. . .]. Rather, I accept that this word refers to something we recognize and instead raise the question: are people with autism trapped – for neurological reasons – to be totally self-focused?’’ (Baron-Cohen, 2005, p. 166). But does it really make sense to discuss whether autism involves a disturbed focus on self, if one does not spend any time discussing and deﬁning the concept of self at play? Perhaps one could object that the notion of self is so univocal and obvious that it is superﬂuous with a more thematic demarcation and clariﬁcation, but that retort is easy to dismiss. The current discussion of self is quite diversiﬁed, to put it mildly. Given this situation, it also does not make much sense – apropos the survey article of Gillihan and Farah – to discuss where the neural correlate of self is located if one does not at the same time make it clear, what concept of self one is operating with, as well as make it clear why one takes one’s point of departure in precisely this concept rather than in another concept. Indeed, part of the lack of consensus documented by Gillihan and Farah might precisely be due to the fact that various experimentalists operate with different notions of self. This is per se not a bad thing – indeed on our view, the self is so multifaceted a phenomenon that various complementary accounts must be integrated if we are to do justice to its complexity. But needless to say, this complexity necessitates conceptual clariﬁcation, since a lack of clarity in the concepts used will lead to a lack of clarity in the questions posed, and thus also to a lack of clarity in the design of the experiments supposed to provide an answer to the questions. In the following, we will exemplify this lack of conceptual clarity, but we will also show that as different notions of self become instrumentalized in particular experimental designs, there is a risk that the initial precision and care with which the notions of self were introduced and deﬁned becomes discursively buried under particular developments in experimental design. We shall demonstrate that this allows for a slip, over time as research progresses, from the experiment as a model of a particular notion of the self to the experiment as a model for Self (Geertz, 1966). This shift does not only constitute a category mistake, it also further complicates necessary interdisciplinary work on these issues. To quickly outline the structure of our contribution: We ﬁrst intend to look more carefully at two highly inﬂuential paradigms in the neuroscientiﬁc investigation of self – the study of facial self-recognition and the study of adjectival self-attribution. Tracing the history of these paradigms clariﬁes how they have developed from highly particular, and highly different, theoretical notions of what the self might be. However, as these approaches have become increasingly instrumentalized in concrete experiments, the premises, which came with the original positions, have increasingly become implicit. Following these trajectories of scientiﬁc practices may thus make explicit tacit assumptions and limitations that characterize each paradigm. This will not only show how cautious one ought to be when making claims about neuroscientiﬁc solutions to the problem of self, but also exemplify why more theoretical reﬂections will remain of paramount importance in this ongoing enterprise.

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

143

2. The face of the self Let us begin with the neuroscientiﬁc study of facial self-recognition. The assumption has typically been that if there are areas in the brain that show more pronounced activity when one recognized one’s own face (compared to what happens when one recognizes other familiar faces) then the respective areas in the brain must constitute or at least be a central part of the neural correlate of self (Gillihan & Farah, 2005). We wish to probe and highlight two main problems and limitations with this approach. The ﬁrst question concerns whether the recognition of a visual representation of one’s own face should count as a paradigmatic and fundamental instance of self-experience. The second question is whether the form of selfexperience that facial self-recognition arguably does exemplify is as socially and culturally impoverished as it has often been made out to be. In articles and books such as ‘‘Where in the brain is the self?’’, ‘‘Where am I? The neurological correlates of self and other’’, and The Face in the Mirror: How we know who we are Keenan’s (and colleagues’) search for the neural location of self has led to the study of facial self-recognition. One of the recurrent ﬁndings reported by Keenan is right frontal lateralized activation for self-face recognition – there is more than twice the activity for self-faces compared to familiar faces (Feinberg & Keenan, 2005, p. 673) – and Keenan has claimed that this empirical evidence provides support for the right hemisphere model of self-awareness (Platek, Keenan, Gallup, & Mohamed, 2004, p. 119). When reading the various publications, one is immediately struck by the almost complete absence of an actual working deﬁnition of both self and self-awareness. In The Face in the Mirror, however, it is acknowledged that a deﬁnition might be in place, and it is proposed that self-awareness amounts to higher-order consciousness and metacognition (Keenan, Gallup, & Falk, 2003, pp. xi–xi, 54, 57). At the same, however, it is also stated that consciousness might be used as a synonym for self-awareness (Keenan et al., 2003, pp. xix, xxi). As we will see in a moment, these initial deﬁnitions are far from innocuous. One question to ask though is why self-face recognition is considered particular relevant and important? Why does it tell us something important about self? It is not difﬁcult to see that Keenan’s investigation of facial self-recognition is indebted to an older and still highly inﬂuential paradigm in developmental psychology and comparative psychology, namely the attempt to subject children, chimpanzees, elephants, dolphins and most, recently, magpies to the mirror self-recognition task in order to test for the presence of self-awareness. Indeed this debt is explicitly admitted by Keenan, who points to Gallup’s classical work, in particular his 1970 and 1982 articles, for providing the theoretical framework (Platek et al., 2004, p. 114). Why did Gallup originally attribute such importance to the passing of the mirror self-recognition task? The standard answer is that the exhibition of self-directed behavior toward a mark surreptitiously put on the face and discovered in the mirror provides empirical and operational evidence for the presence of conceptual self-awareness (Gallup, 1977, p. 337). On closer look, however, it turns out that Gallup was not merely stressing the link between mirror self-recognition and self-awareness. He also took the passing of the mirror task to be a litmus test for the possession of consciousness. Thus, on Gallup’s view, consciousness is bidirectional. It allows one to attend outwardly to things in the world, but also to attend inwardly and to monitor one’s own mental states (Gallup, 1982, p. 242). To that extent, consciousness covers and includes both awareness and self-awareness. In continuation of this line of thought, Gallup also claimed that conscious experience necessarily presupposes self-awareness and that creatures that lacked the ability to monitor their own mental states were mindless. Either one is aware of being aware, or one is unaware of being aware, and the latter amounts to being unconscious (Gallup, 1982, pp. 243, 245; Gallup, 1985, p. 638). Following this line of reasoning, Gallup concluded that although most organisms behave as if they are conscious and minded, prior to the emergence of self-awareness as evidenced from their ability to pass the mirror self-recognition task, they are mindless. They lack conscious experience, and only possess unconscious sensations, pains, etc. (Gallup, 1982, p. 242; Gallup, 1985, p. 638). Although Keenan does not explicitly endorse such a view, his own deﬁnitions seem to point in similar directions. When deﬁning self-awareness in terms of reﬂective metacognition, and when saying that consciousness is a synonym for such selfawareness, it follows that creatures incapable of such reﬂective metacognition also lack consciousness. To put it differently, Keenan’s initial deﬁnition ultimately commits him to a highly controversial higher-order representational account of consciousness according to which a creature only enjoys conscious experiences if it has the ability to reﬂect upon its own mental states. There are many good reasons for resisting such a conclusion, which has rather dramatic implications not only for our ascription of conscious experiences, i.e., mental episodes with phenomenal feels, to infants, but also to all those animals which lack a cognitive capacity for reﬂection. In fact, higher-order representational accounts of consciousness have come under increasing attack in recent years (cf. Kriegel & Williford, 2006; Zahavi, 2004, 2005), and perhaps Keenan himself would refrain from such a view. At least, they are not implications he explicitly draw and endorse. So let us return to his central claim. Keenan repeatedly claims that the ability to pass the mirror self-recognition test demonstrates the capacity for self-awareness. But how strong is this correlation supposed to be? There is hardly any doubt that Keenan opts for a strong claim, which would also match Gallup’s view on the matter. This is why Keenan claims that the passing of the mirror test is highly correlated with every indicator of self-awareness and that the absence of mirror self-recognition is correlated with an absence of other self-aware behaviors (Keenan et al., 2003, p. 22). It is not entirely clear, however, that Keenan’s own ﬁndings really support these statements. Take the case of people suffering from the delusional misidentiﬁcation symptom known as mirror sign, which Keenan also discusses. Presumably they are incapable of passing the mirror self-recognition test – and the reason they have this incapacity is speciﬁcally related to problems with self-recognition, and is not simply due to some form

144

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

of prosopagnosia. But if one were to take Keenan seriously, people such as these, would also lack self-awareness and the capacity for self-reﬂection. But there is no evidence that this happens to be the case. More generally speaking, it is not difﬁcult to come up with quite general objections to the idea that our ability to identify a visual representation of our own face should constitute a particular central or fundamental form of self-awareness. Although facial self-recognition might testify to the existence of a form of self-awareness, the failure to recognize one’s own face certainly does not prove the absence of every form of self-awareness. To put it differently, the absence of facial self-recognition might be perfectly compatible with the presence of other forms of self-awareness. Indeed, as we see it, one decisive problem facing Keenan’s interpretation of his own ﬁndings is that he underestimates how complex and varied self-experience is. Not only does he disregard the possibility that phenomenal consciousness involves self-awareness in the weak sense that there is something it is like for the subject to have the experience, i.e., that the distinct ﬁrst-personal character of phenomenal consciousness amounts to a low-level form of self-awareness (cf. Flanagan, 1992; Zahavi, 1999, 2003, 2005), but he also seems to ignore the possibility that infants might have a sense of their own bodies as organized and environmentally embedded entities long before they are able to pass any mirror self-recognition tasks, and, hence, an early embodied sense of themselves in perception and action. Thus, as many developmental psychologists have pointed out, already from around 3 months of age, infants discriminate what pertains to the self and what pertains to someone else interacting with them. In the footsteps of Neisser and Gibson, one could call this early sense of self the infant’s ecological self (Rochat, 2001, pp. 30–31, 41). Would we really be able to recognize our own mirror-image, which presumably relies on a detection of the perfect cross-modal match between our own bodily movements and the movements of the mirror-image, if we were not already proprioceptively aware of our own bodily movements? To put it differently, would one really be able to recognize oneself in the mirror, if one lacked bodily self-awareness? There are even those who claim that this bodily selfawareness constitutes the fundamental requirement. This is for instance the claim made by Mitchell in a number of papers. He has argued that mirror self-recognition merely requires a kinesthetic sense of own body (subjective self-awareness), a capacity for kinesthetic-visual matching and an understanding of mirror-correspondence (Mitchell, 1997a, p. 31; Mitchell, 1997b, p. 41). If so, it would obviously invalidate Gallup and Keenan’s claim that mirror self-recognition requires both introspection and the possession of a self-concept or an internal self-model (cf. Keenan et al., 2003, p. 11). Indeed, as Mitchell has pointed out, it remains quite unclear what mental state a creature is supposed to attend to in recognizing itself in the mirror (Mitchell, 1997a, p. 23). All of this is not to say that mirror self-recognition is insigniﬁcant, the question though is whether Keenan and before him Gallup have realized its proper signiﬁcance. In some of his early writings, Gallup took mirror self-recognition to testify to the perfect match between the observer and the observed. As he put it ‘‘The unique feature of mirror-image stimulation is that the identity of the observer and his reﬂection in a mirror are necessarily one and the same’’ (Gallup, 1977, p. 334). In addition, he repeatedly emphasized the distinction between social responsiveness and self-directed mirror behavior and claimed that one in recognizing one’s mirror-image ceased to respond socially to it (Gallup, 1970, p. 86). But is this really correct? Consider that a visual representation of one’s own face provides one with information about oneself that is very different from what one hitherto has been in possession of. Consider that the face we see reﬂected in the mirror is also the face others see when we interact with them. Indeed one of the reasons why people spend so much time before a mirror engaged in impression management is precisely because of the high social valence of one’s face. When seeing myself in the mirror, I am confronted with the appearance I present to others. To see oneself in the mirror (or in a photo) is to become a spectator of oneself. It is to adopt a perspective or viewpoint on oneself that equals what others can adopt. Contrary to what Gallup is claiming, to recognize oneself in the mirror does not simply involve an identiﬁcation of the felt me which is here, and the perceived me which is there. There is more at stake than a simple afﬁrmation of a pre-existing identity attached to it, there is also what Merleau-Ponty described as the unsettling experience of realizing that the felt me has an exterior dimension that can be witnessed by others (cf. Merleau-Ponty, 1964, pp. 129, 136, 140). There is an awareness of one’s public appearance. In fact, not only am I seeing myself as others sees me, I am also seeing myself as if I was an other, i.e., I am adopting an alienating perspective on myself. To put it differently, the mirror and the photo afford quite new possibilities of adopting an objectiﬁed stance towards ourselves. To recognize an image of oneself is to appropriate an objectiﬁcation of oneself. More needs to be said in defending this claim. In particular, one has to consider the relation between successful mirror self-recognition in humans and in non-human animals. It is, however, not at all obvious that the two can be equated and that mirror self-recognition in chimpanzees or magpies corresponds to the cognitive and affective self-consciousness manifested in children passing the test (cf. Rochat & Zahavi, in press). For now, however, all we wish to deny is that the recognition of a visual representation of one’s own face counts as the paradigmatic and fundamental instance of self-experience and that it is as socially and culturally impoverished as it has often been made out to be. To recognize one’s own mirror-image is deﬁnitely not to cease responding socially to it. To put it differently, to test facial self-recognition is not to test the self per se, but to test and probe a quite speciﬁc dimension of self, e.g., the self as social object. This angle and limitation is something that has not been sufﬁciently considered in recent neuroscience. One possible objection to this criticism might run as follows: In both Gallup and Keenan one ﬁnds occasional positive references to the work of Mead and Cooley (cf. Gallup, 1977, p. 335; Keenan et al., 2003, p. 41), who both explicitly and persistently discussed the self as social object. It is consequently important to notice that whereas Keenan’s work is indebted to Gallup’s, Gallup’s own account of self-recognition is precisely inﬂuenced by ideas found much earlier in Cooley and Mead (cf. Gallup, 1975, 1983; Gallup, McClure, Hill, & Bundy, 1971). However, on closer examination it turns out that Gallup in the process of developing his own account gradually altered and ultimately inversed these ideas (for a careful analysis of Gallup’s

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

145

puzzling reliance and use of these authors, see Mitchell, 1997a). Whereas Cooley and Mead argued that self-knowledge presupposes knowledge of others and that the self-concept derives from one’s taking the perspective of another toward oneself (Cooley, 1912, 246; Mead, 1962, 138), Gallup and Keenan both defend the view that knowledge of others presupposes knowledge of self and a developed self-concept. Gallup writes that the tendency to impute mental states to others presupposes the capacity to monitor such states on the part of the individual making the imputation (Gallup, 1982, 243), and Keenan argues that it is because I know my own thoughts that I can predict or infer another person’s mental state (Keenan et al., 2003, p. 78). As should have become clear by now, the neuroscientiﬁc investigation of facial self-recognition relies on a theoretical framework with a long (and partly forgotten) history. When it comes to its more principled reﬂections on and analyses of central concepts such as consciousness, self-awareness and self – which shaped the design of the experiment and continue to inﬂuence the interpretation of the empirical ﬁndings – these are clearly inadequate and in many ways committed to rather controversial theses. 3. Ascribing myself In recent brain imaging literature, arguably the most inﬂuential experimental paradigm for identifying putative neural correlates of the self has been the adjective ascription task. Brieﬂy, the task involves placing experimental subjects in a suitable brain scanner and presenting them with a list of adjectives. While their brains are being scanned, subjects have to evaluate the list of adjectives in different ways, one of which involves evaluating whether the word appropriately describes themselves. In this way, it is possible to keep the visual input constant between the different sessions, while the experimental contrast is provided by the script, which speciﬁes what the experimental subject should do with the word (Jack & Roepstorff, 2002). The task was used both in an early PET study by Craik and colleagues (1999) and in an early fMRI study by Kelley and colleagues (2002). Both papers have been highly inﬂuential (with 346 citations and 535 citations respectively according to Google scholar at the time of writing). While the title of the Craik et al., study In Search of the Self: A Positron emission tomography study suggested an open ended exploration, by the time of the Kelley paper Finding the Self? An event related fMRI study only the question mark, highly unusual for the title of a neuroscience paper, suggested some uncertainty. Effectively, the paradigm today provides a ready-made technology to identify a neural correlate of the self; you take a scanner, add an adjective ascription task, and compare self ascription to e.g. other ascription or mother ascription (Ray et al., 2010; Vanderwal, Hunyadi, Grupe, Connors, & Schultz, 2008). Then one does the trick that brain scanners can do (Roepstorff, 2002), and out comes a set of data that may be transformed to an image (Roepstorff, 2007). Just as with the mirror-recognition task, one may ask the question why is adjective ascription considered a particularly relevant task to identify neural correlates of the self? To the extent that the self, at least analytically, effectively becomes what the experiment measures, it may be instructive to follow in some detail the construction of this paradigm for eliciting the self. The origin of the self ascription task is to be found more than 30 years ago, in a memory study that investigated whether self-reference could serve a function in the processing of certain kinds of information (Rogers, Kuiper, & Kirker, 1977). Their experiment combined two lines of experimentation. One of them was the already then classic idea that schemata or prototypes are key in organizing memory (Bartlett, 1932). A number of researchers in social and personality psychology had recently suggested that personality traits in general could be conceived of as such schemata. This had been tested e.g. in how subjects would remember trait adjectives about characters in a narrative (see e.g. Cantor & Mischel, 1977). Rogers et al. extrapolated this idea to the self, and suggested that self-reference could involve a schema of trait-like features, abstracted over the life history of the individual, which could be seen as a feature list. Hence, ‘‘[w]hen self-reference is involved, it should provide a useful device for encoding or interpreting incoming information by virtue of accessing the extensive past experience abstracted in the self’’ (Rogers et al., 1977, 678–9). They then combined this idea of the self with recent experimental work on memory retention, which found that words, which a subject had been exposed to, would be remembered differently, depending on how deep the processing had been. Craik and Tulving (1975) had recently demonstrated that words presented as part of a semantic task, which probed their meaning, were remembered better than words presented as part of a phonemic task, which probed the letters in the words, and these words were, again, recalled better than words presented in a structural task where subjects had to decide if the words had been written in large or in small letters. These data had been interpreted as support for the position that the strength of the memory trace is ‘‘a positive function of ‘depth’ of processing, where depth refers to greater degrees of semantic involvement’’ (Craik & Tulving, 1975 p. 268, cited from Rogers et al., 1977). With these two lines of experimentation in hand, Rogers et al. had an easy way to test the hypothesis that if the self was a prominent prototype in organizing experiences, then relating words to ‘the self’ would be a very efﬁcient way to remember them. In their own words: ‘‘If the self is an active agent in the encoding of personal data, we predicted that the self-reference rating would produce good incidental recall in this depth-of-processing paradigm. If incidental recall of the self-reference words is superior to that for semantic words, the hypothesis that the self serves an active and powerful role in processing personal data would be supported’’ (Rogers et al., 1977, p. 680). The design of the study was quite straight forward. It took the design developed by Craik and Tulving (1975) and added a ‘self’ related component. That is, subjects saw lists of adjectives on four tasks designed to force varying kinds of encoding: structural, phonemic, semantic, and self-reference, and they should consider whether each word ﬁttingly described them.

146

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

After this encoding phase, subjects incidental recall of the rated words were examined. This indicated, in accordance with the stated hypothesis, that adjectives rated under the self-reference task were recalled better than semantically, phonetically and structurally encoded words. The authors hence concluded that ‘‘the present data indicate unequivocally that words rated under the self-reference task show superior recall. This indicates that self-reference represents a powerful and rich encoding device’’ (Rogers et al., 1977, p. 685). Further, ‘‘In order for self-reference to be such a useful encoding process, the self must be a uniform, well-structured concept. During the recall phase of the study, subjects probably use the self as a retrieval cue [. . .]. In order for this to be functional, the self must be a consistent and uniform schema [. . .]. In sum, the self contains a set of ordered features [. . .]. The ordering appears to be from general to speciﬁc, with the general terms (e.g., traits) ordered by a combination of salience and extremity. The general terms can serve as schemata when studied independently of a person’s idiographic view of self’’ (op. cit., p. 686). The article is a masterpiece in experimental psychology, and it demonstrates quite clearly a particular style of producing facts (Fleck, 1979). Through a combination of different pieces of experimental work, that each solves a particular problem, bit-by-bit a new argument is constructed, almost like constructing out of Lego pieces a particular building. In the same manner, this adjective ascription task, ﬁrst used to test a particular version of how relating to the self may allow a particular ‘deep’ level of processing, has now become used as a proxy to identify in the brain ‘‘the neuronal correlate of the self’’ (Craik et al., 1999; Kelley et al., 2002). These and other studies identiﬁed a number of regions, but in particular there seemed to be more activity in the so-called medial prefrontal cortex (MPFC) during the ‘self’ part of the paradigm across a number of adjective ascription experiments. Once such a putative ‘self’ area was found, differential activity could be, and has been, used to study putative cultural (Zhu, Zhang, Fan, & Han, 2007) or religious (Han et al., 2008) aspects of the self, in relation to differences in self-construal within a given culture (Ray et al., 2010), between different generations (e.g. Feyers, Collette, D’Argembeau, & Majerus, 2010; Gutchess, Kensinger, & Schacter, 2010), between individuals with autism and a control group (Kennedy & Courchesne, 2008) or persons with depression and a control group (Lemogne et al., 2009) or as dynamic modulations of culturally mediated perceptions of the self (Chiao et al., 2010). A striking development happened in this transformation from a memory task to a technology to identify different aspects of the self between different groups. This occurred over the course of more than 30 years and through a set of subsequent experimental steps that all involve ampliﬁcations and reductions of speciﬁc features (Latour, 1999; Roepstorff, 2004). The ﬁrst experiments began with the assumption that personal traits could be seen as prototypes. This led to a ‘self as traits’ approach, where the self appeared to allow for deep processing. Once taken to the scanner environment, the data were probed to see if any parts of the brain could be indicative of a ‘deep’ processing, indexed by more activity in these regions. The next step was the assumption that as these regions appeared to be more involved in self processing, activity in these regions could provide a ‘neural signature’ of the self and they could be seen as ‘self related’ areas. If that were the case, activity in these regions could be interpreted as an index of self-related activities. Therefore, differences in activity during self related tasks in these areas could be a sign of differences in self processing between groups. Hence, differences in ‘self’ could be a trait for that group, and that trait could, perhaps, be considered prototypical. This progression has been a very powerful fact making strategy, but it also appears to have an inherent circularity that may be unfolded in the following way: personal trait as prototypes ? self as traits ? self as deep processing ? deep-processing as neuronally located ? neural location as index of self ? neuronal level of activation in groups as index of self ? self as group trait ? group trait as prototype for the individual self In other words, what began as a particular conceptualization of the self as a schema, which was handy because it could easily be implemented in an experiment, ended with subsequent generations of experiments purporting to provide a model for where the self is in the brain, and ultimately what the self is. This shift from a model of the self to a model for the self demonstrates, we believe, that the paradigms remain dependent upon the underlying assumptions, i.e., they point to and rely on a particular understanding of the self. This is very clearly seen in the early papers, where the paradigms were developed. The authors were acutely aware that the argument was, in the end, conceptual. However, as the paradigms moved across disciplines and techniques, this understanding became buried under new shifts in perspectives, new problems to solve. Whereas in the beginning the aim had been to design experiments that could allow one to get a handle on what a self is, in the end the self ended up being deﬁned in terms of what the experiment could handle. As also pointed out by Gillihan and Farah (2005) in their very careful review, the observation of apparent differential activity in the medial prefrontal cortex during a self ascription task cannot be considered a ‘proof’ that the ‘self’ is located in these areas. Remember the origin of the task, it began as a memory experiment that investigated whether adjectives were remembered better if they were related to the self. This was based on the argument that ‘the self’ formed a strong schema for categorizing and remembering incoming information. Given that the knowledge of oneself is probably better than the knowledge of abstract others such as the former US president (Kelley et al., 2002) or the Danish Queen (Lou et al., 2004) it cannot be ruled out that the differential activity in MPFC is a memory effect which reﬂects a difference in familiarity. This may explain, as Gillihan and Farah (2005) also suggests, that in some experimental contexts, close others and the self both appear to activate this region (see e.g. Zhu et al., 2007 & Vanderwal et al., 2008 for ‘mother related’ case). In this alternative interpretation, which closely follows the original rationale behind the self ascription task, close others, like oneself, may be used as powerful schema for organizing and remembering personal adjectives. It is obviously interesting if between individuals or groups, no matter whether these are ‘clinical’ categories or cultural’ categories, different types of persons form such

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

147

schema. This suggests different patterns in how self-other relations are constituted, and examining how such relations are changed, e.g. by priming, may point to important contextual dynamics both of the self and of the brain (Ng, Han, Mao, & Lai, 2010). Such a line of research provides a potentially powerful challenge to universalizing tendencies in the cognitive sciences (Roepstorff, in press), and it obviously differs considerable from the claim that the self is localized somewhere in the brain and that activity in this region is a sign of a particular form of selfhood. 4. Conclusion In various publications Keenan and colleagues have claimed that the search for the localization of the self in the brain has been the goal of consciousness research for centuries (Feinberg & Keenan, 2005, p. 661), and that this problem remains one of the great mysteries of science, philosophy and psychology (Keenan et al., 2003, p. 99). In the article entitled ‘‘Where in the brain is the self?’’ which Keenan co-authored with Feinberg, the authors are careful in modifying the initial claim by conceding not only that modules of the brain do not exist in isolation, and that one has to view the brain in its entirety (Feinberg & Keenan, 2005, p. 673), but also that it might be more appropriate and correct to opt for the more modest claim that the right hemisphere is dominant for certain aspects of self, than to make the stronger claim that the self resides in the right hemisphere (Feinberg & Keenan, 2005, p. 675). We could not agree more. It is indeed far better to label the search for the neural correlates of self, a search for those neural structures and mechanisms that enable self-recognition and self-experience, than to describe it as an attempt to locate the self in the brain. The latter claim is in our view (and here we would side with so otherwise antagonistic philosophers as Dennett and Hacker) tantamount to a category mistake (Bennett & Hacker, 2003; Dennett, 1992). For that very reason, the answer to the question, where in the brain is the self, can only be nowhere. To say that the self is nowhere in the brain, is, however, not to say that ‘‘nobody ever was or had a self’’ (Metzinger, 2003, p. 1). We are not self-skeptics. We do not deny the reality or question the ontological status of the self (cf. Siderits, Thompson, & Zahavi, 2010, 2005), we are simply denying that the self is a kind of thing that can be found in the brain. Again, this is by no means to question the value of some of the existing paradigms. In fact, a task like the adjective ascription task is a simple, effective and easily implementable method that has in the past taught us much about how memories are stored and processed, and used in combination with brain scanners may tell a lot about how probing knowledge, affects and expectations about oneself and about others may tie into particular networks of the brain. Equally, differential activities in these networks across groups, contexts and primings may give important hints both to the workings of the brain and of the mind. However, we are not convinced that this translates into the self being ‘located’ in those parts of the brain that are activated by a self-directed adjective ascription task. Moreover, as our analysis of facial self-recognition ought to have shown, neuroscientiﬁc research on the neural correlates self-recognition and self-experience necessitates rather than obviates the need for a careful conceptual analysis. Even recognizing something as apparently simple as an image of oneself is more complicated than just contrasting self and other. It involves the appropriation of an objectiﬁcation of oneself, and thus entails a critical tension between experiencing oneself as object and experiencing oneself as subject (Legrand, 2007). This is nothing but a special case of a much more complicated pattern of how people in interaction with others constitute and develop themselves, and others (Zahavi, 2009, 2010). Such intersubjective interplay can be followed even in concrete interactions during a simple cognitive experiment (Roepstorff, 2001). The self is complex and multidimensional. This complexity necessitates interdisciplinary collaboration; collaboration across the divide between theoretical analysis and empirical investigation. To think that a single discipline, be it philosophy or neuroscience, should have a monopoly on the investigation of self is merely an expression of both arrogance and ignorance. Acknowledgments The authors acknowledge the ﬁnancial support from the Danish National Research Foundation and from the Danish Council for Independent Research: Humanities. References Baron-Cohen, S. (2005). Autism-‘Autos’: Literally, a total focus on the self? In T. E. Feinberg & J. P. Keenan (Eds.), The lost self: Pathologies of brain and identity (pp. 166–180). Oxford: Oxford University Press. Bartlett, F. C. (1932). Remembering. A study in experimental and social psychology. Cambridge: Cambridge University Press. Bennett, M. R., & Hacker, P. M. S. (2003). Philosophical foundations of neuroscience. Oxford: Blackwell. Campos, J.J. (2007). Foreword. In J. L. Tracy, R. W. Robins, & J. P. Tangney (Eds.), The self-conscious emotions: Theory and research (pp. ix–xii.). New York: The Guilford Press. Cantor, N., & Mischel, W. (1977). Traits as prototypes: Effects on recognition memory. Journal of Personal and Social Psychology, 35, 38–48. Chiao, J., 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 Neuroscience, 22, 1–11. Cooley, C. H. (1912). Human nature and the social order. New Brunswick, NJ: Transaction Books. Craik, F., Moroz, T., Moscovitch, M., Stuss, D., Winocur, G., Tulving, E., et al (1999). In search of the self: A positron emission tomography study. Psychological Science, 10, 26–34. Craik, F., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268–294. Crick, F. (1995). The astonishing hypothesis. London: Touchstone.

148

D. Zahavi, A. Roepstorff / Consciousness and Cognition 20 (2011) 141–148

Dennett, D. C. (1992). The self as the center of narrative gravity. In Frank S. Kessel, Pamela M. Cole, & Dale L. Johnson (Eds.), Self and consciousness: Multiple perspectives (pp. 103–115). Hillsdale, NJ: Erlbaum. Feinberg, T. E., & Keenan, J. P. (2005). Where in the brain is the self? Consciousness and Cognition, 14, 661–678. Feyers, D., Collette, F., D’Argembeau, A., & Majerus, S. (2010). Neural networks involved in self-judgement in young and elderly adults. Neuroimage, 53, 341–347. Flanagan, O. (1992). Consciousness reconsidered. Cambridge, MA: MIT Press. Fleck, L. (1979). Genesis and development of a scientiﬁc fact. Chicago: Chicago University Press. Gallup, G. G. (1970). Chimpanzees: Self-recognition. Science, 167, 86–87. Gallup, G. G. (1975). Towards an operational deﬁnition of self-awareness. In R. H. Tuttle (Ed.), Socioecology and psychology of primates (pp. 309–341). Paris: Mouton & Co. Gallup, G. G. (1977). Self-recognition in primates: A comparative approach to the bidirectional properties of consciousness. American Psychologist, 32, 329–338. Gallup, G. G. (1982). Self-awareness and the emergence of mind in primates. American Journal of Primatology, 2(3), 237–248. Gallup, G. G. (1983). Toward a comparative psychology of mind. In R. L. Mellgren (Ed.), Animal cognition and behavior (pp. 473–510). Amsterdam: NorthHolland Publisher. Gallup, G. G. (1985). Do minds exist in species other than our own? Neuroscience and Biobehavioral Reviews, 9(4), 631–641. Gallup, G. G., McClure, M. K., Hill, S. D., & Bundy, R. A. (1971). Capacity for self-recognition in differentially reared chimpanzees. Psychological Record, 21, 69–74. Geertz, C. (1966). Religion as a cultural system. In the interpretation of cultures (pp. 87–125). New York: Basic Books. Gillihan, S. J., & Farah, M. J. (2005). Is self special? A critical review of evidence from experimental psychology and cognitive neuroscience. Psychological Bulletin, 131(1), 76–97. Gutchess, A., Kensinger, E., & Schacter, D. (2010). Functional neuroimaging of self-referential encoding with age. Neuropsychologia, 48, 211–219. Han, S., Mao, L., Gu, X., Xhu, Y., Ge, J., & Ma, Y. (2008). Neural consequences of religious belief on self-referential processing. Social Neuroscience, 3, 1–15. Jack, A. I., & Roepstorff, A. (2002). Introspection and cognitive brain mapping: From stimulus-response to script-report. Trends in Cognitive Sciences, 6, 333–339. Keenan, J. P., Gallup, G. G., & Falk, D. (2003). The face in the mirror: How we know who we are. New York: HarperCollins. Kelley, W., Macrae, C., Wyland, C., Caglar, S., Inati, S., & Heatherton, T. (2002). Finding the self? An event-related fMRI study. Journal of Cognitive Neuroscience, 14, 785–794. Kriegel, U., & Williford, K. (Eds.). (2006). Consciousness and self reference. Cambridge, Mass.: MIT Press. Kennedy, D., & Courchesne, E. (2008). Functional abnormalities of the default network during self-and other-reﬂection in autism. Social Cognitive and Affective Neuroscience, 3, 177. Latour, B. (1999). Pandora’s hope: Essays on the reality of science studies. Cambridge, Mass.: Harvard University Press. Legrand, D. (2007). Pre-reﬂective self-as-subject from experiential and empirical perspectives. Consciousness and Cognition, 16(3), 583–599. Lemogne, C., Le Bastard, G., Mayberg, H., Volle, E., Bergouignan, L., Lehéricy, S., et al (2009). In search of the depressive self: Extended medial prefrontal network during self-referential processing in major depression. Social Cognitive and Affective Neuroscience, 4, 305–312. Lou, H., Luber, B., Crupain, M., Keenan, J., Nowak, M., Kjaer, T., et al (2004). Parietal cortex and representation of the mental self. Proceedings of the National Academy of Sciences, 101(17), 6827. Mead, G. H. (1962). Mind, self and society. from the standpoint of a social behaviorist. Chicago: University of Chicago Press. Merleau-Ponty, M. (1964). The child’s relations with others. In J. Edie (Ed.), The primacy of perception (pp. 96–155). Evanston, Ill: Northwestern University Press. Metzinger, T. (2003). Being no one. Cambridge, MA: MIT Press. Mitchell, R. W. (1997a). Kinesthetic-visual matching and the self-concept as explanations of mirror-self-recognition. Journal for the Theory of Social Behavior, 27(1), 17–39. Mitchell, R. W. (1997b). A comparison of the self-awareness and kinesthetic-visual matching theories of self-recognition: Autistic children and others. Annals of the New York Academy of Sciences, 818, 39–62. Neisser, U. (1988). Five kinds of self-knowledge. Philosophical Psychology, 1(1), 35–59. Ng, S., Han, S., Mao, L., & Lai, J. (2010). Dynamic bicultural brains: fMRI Study of their ﬂexible neural representation of self and signiﬁcant others in response to culture primes. Asian Journal of Social Psychology, 13, 83–91. Platek, S. M., Keenan, J. P., Gallup, G. G., & Mohamed, F. B. (2004). Where am I? The neurological correlates of self and other. Cognitive Brain Research, 19, 114–122. Ray, R., Shelton, A., Hollon, N., Matsumoto, D., Frankel, C., Gross, J., et al (2010). Interdependent self-construal and neural representations of self and mother. Social Cognitive and Affective Neuroscience, 5(2–3), 318–323. Rochat, P. (2001). The infant’s world. Cambridge, MA: Harvard University Press. Rochat, Ph. & Zahavi, D. (in press). The uncanny mirror: A re-framing of mirror self-experience. Consciousness and Cognition, doi:10.1016/ j.concog.2010.06.007. Roepstorff, A. (2001). Brains in Scanners: An umwelt of cognitive neuroscience. Semiotica, 134, 747–765. Roepstorff, A. (2002). Transforming subjects into objectivity. An ethnography of knowledge in a brain imaging laboratory. FOLK, Journal of the Danish Ethnographic Society, 44, 145–170. Roepstorff, A. (2004). Mapping brain mappers: An ethnographic coda. In R. Frackowiak et al. (Eds.), Human brain function, 2nd ed (pp. 1105–1117). London: Elsevier. Roepstorff, A. (2007). Navigating the brainscape: When knowing becomes Seeing. In C. Grasseni (Ed.), Skilled Vision. Between apprenticeship and standards (pp. 191–206). Oxford: Berghahn Press. Roepstorff, A. (in press). Culture: A site of relativist energy in the cognitive sciences. Common Knowledge. Rogers, T., Kuiper, N., & Kirker, W. (1977). Self-reference and the encoding of personal information. Journal of Personal and Social Psychology, 35, 677–688. Seeley, W. W., & Miller, B. L. (2005). Disorders of the self in dementia. In T. E. Feinberg & J. P. Keenan (Eds.), The lost self: Pathologies of brain and identity (pp. 147–165). Oxford: Oxford University Press. Siderits, M., Thompson, E., & Zahavi, D. (Eds.). (2010). Self, no self? Perspectives from analytical, phenomenological and Indian traditions. Oxford: Oxford University Press. Vanderwal, T., Hunyadi, E., Grupe, D., Connors, C., & Schultz, R. (2008). Self, mother and abstract other: An fMRI study of reﬂective social processing. Neuroimage, 41, 1437–1446. Zahavi, D. (1999). Self-awareness and alterity: A phenomenological investigation. Evanston: Northwestern University Press. Zahavi, D. (2003). Phenomenology of self. In T. Kircher & A. David (Eds.), The self in neuroscience and psychiatry (pp. 56–75). Cambridge University Press. Zahavi, D. (2004). Back to brentano? Journal of Consciousness Studies, 11(10–11), 66–87. Zahavi, D. (2005). Subjectivity and selfhood: Investigating the ﬁrst-person perspective. Cambridge, MA: The MIT Press. Zahavi, D. (2009). Is the self a social construct? Inquiry, 52(6), 551–573. Zahavi, D. (2010). Shame and the exposed self. In J. Webber (Ed.), Reading Sartre: On phenomenology and existentialism (pp. 211–226). London: Routledge. Zhu, Y., Zhang, L., Fan, J., & Han, S. (2007). Neural basis of cultural inﬂuence on self-representation. Neuroimage, 34, 1310–1316.

Consciousness and Cognition 20 (2011) 149–155

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

Dissociation in self-narrative q Shaun Gallagher a,b,⇑, Jonathan Cole c a b c

Philosophy and Cognitive Sciences, Institute of Simulation and Training, University of Central Florida, United States School of Humanities, University of Hertfordshire, United Kingdom Poole Hospital and Centre for Postgraduate Medical Research and Education, University of Bournemouth, United Kingdom

a r t i c l e

i n f o

Article history: Received 2 October 2010 Available online 5 November 2010 Keywords: Narrative Narrative distance Dissociation Psychopathology Syntax

a b s t r a c t We review different analytic approaches to narratives by those with psychopathological conditions, and we suggest that the interpretation of such narratives are complicated by a variety of phenomenological and hermeneutical considerations. We summarize an empirical study of narrative distance in narratives by non-pathological subjects, and discuss how the results can be interpreted in two different ways with regard to the issue of dissociation. Ó 2010 Elsevier Inc. All rights reserved.

1. Narratives by those with psychopathological conditions What is the relation between our thoughts and feelings, and the varying degrees of their embodied expression in posture, gesture facial expression (in relation to emotions particularly), and linguistic expression (both talking to oneself and to others)? How one presents oneself and one’s feelings and thoughts to another is extraordinarily complex, and depends on many factors; age, sex, mood, power relations, and culture to name but a few. What language one uses is constrained by these factors, but also allows for the possibility of attempting straightforward veracity, or dissemblance and deception, as well as the use of various linguistic devices such as irony, humor, and sarcasm. Mood and affect are also communicated through gesture, posture and prosody of voice. Of course much interpersonal expression is ephemeral; body language and prosody may be potent communication channels but without video or ﬁlm are lost as moments past. In contrast the written language, as seen in narrative, endures. Within medicine and especially neurology there is a classic tradition of examining narrative accounts of living with conditions, from Weir Mitchell to Luria and Sacks. In these writers’ work are found extraordinarily rich narratives of brain damage, autism, aphasia, and blindness, for example. Yet rarely are these accounts, taken from the patients themselves, questioned in regard to their narrative structure; rather – where relevant – this is taken into account as required, say in aphasia in order to understand it and portray it to others. It seems clear, however, that certain conditions, of themselves, may affect the range of narrative possible to a given patient. If we, simplifying for the moment, distinguish the various parts which make up an expressive narrative then it might include the expressed thoughts, feelings, and experiential world, altered by a given condition, and the linguistic and embodied expressive channels through which experience is communicated. In some narratives of neurological patients the latter is

q

This article is part of a special issue of this journal on Brain and Self: Bridging the Gap.

⇑ Corresponding author at: Philosophy and Cognitive Sciences, Institute of Simulation and Training, University of Central Florida, USA. Fax: +1 407 823 6658. E-mail addresses: [email protected] (S. Gallagher), [email protected] (J. Cole). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.10.003

150

S. Gallagher, J. Cole / Consciousness and Cognition 20 (2011) 149–155

the focus and the former is presumed to function fairly well intact. Thus in pure aphasia one sees curtailment of language (and gesture) but we might presume that the experiential and feeling states are largely intact, though affected by the person’s new loss. Alternatively, in adult onset blindness linguistic skills may be unaffected, and there have been very powerful ﬁrst person accounts of going blind. In other cases there can be differences in narrative style and vocabulary. Someone living with autism may use language differently, reﬂecting in part the landscape of experience of the condition, as revealed in Temple Grandin’s ﬁrm expressive prose. In depression one might see a profound change in mood with no overt change in linguistic skills, though the person’s use of language and of various words may change too. In other conditions an alteration in one’s experience of one’s body, through neurological loss, or after a central brain problem, may not only change thoughts and feelings but may have effects on communication and in particular on that person’s use of narrative. In fact some conditions motivate us to ask if these two sides – the experiential and the expressive – are so interlinked as a continuum that our simple binary consideration above has little utility. In this paper we consider the following question: Can narratives provide a forensic measure, a linguistic ﬁngerprint, of psychopathology? In other words, in the self-narratives of psychopathogical conditions, might the narratives themselves be considered ‘pathological’? This question, or more precisely, this strategy has been pursued in a number of studies that have examined the narratives of people with psychopathological conditions. For example, Pennebaker, Mehl, and Niederhoffer (2003) reported that the increased use of the ﬁrst-person pronoun predicts both depression and mania. Stirman and Pennebaker (2001) analyzed works of poets who committed suicide compared to poets who did not; they found a relatively higher frequency of self-reference words (e.g., I, me, my) as compared to other-reference words (e.g., we, them, they). Junghaenel, Smyth, and Santner (2008) found a lower frequency of cognitive words (e.g., cause, know, ought) in narratives produced by psychiatric patients. Focusing on schizophrenia, Caixeta, Chaves, Caixeta, and Reis (1999) showed that schizophrenic patients who narrate emotionallyloaded facts (vs. neutral facts) showed higher degree of dyslogia or wrong or novel words. In a study by Raffard et al. (2010) schizophrenics’ narratives were less coherent and elaborate than controls, suggesting problems connecting with the self and with memory. Lysaker, Clements, Plascak-Hallberg, Knipschure, and Wright (2002), and Lysaker et al. (2005) have performed a number of studies of narrative, metacognition and self-esteem and have developed the Scale to Assess Narrative Development (STAND) that focuses on the narrative self. Finally, Gruber and Kring (2008) showed that for schizophrenics, negative emotion narratives were less grammatically clear than positive ones, and positive emotion narratives were more likely to involve other people than negative narratives. These studies suggest that a variety of data can be collected from the analysis of narratives from people with psychopathological conditions. We want to suggest, however, that accurate interpretation of such data is not always easy. Interpretation of narrative, in terms of its semantic and syntactic structure, is constrained both by psychopathological theory, and hermeneutical-narrative theory. Consider, for example, three different views on the nature of schizophrenia and how they impinge on one’s interpretation of schizophrenic narratives. First, Graham and Stephens (1994) argue that in schizophrenia a certain disruption in the sense of agency may be the result of problems at a higher-order cognitive level which is reﬂected in ‘‘our proclivity for constructing self-referential narratives.” [normally] the subject’s sense of agency regarding her thoughts . . . depends on her belief that these mental episodes are expressions of her intentional states. That is, whether the subject regards an episode of thinking occurring in her psychological history as something she does, as her mental action, depends on whether she ﬁnds its occurrence explicable in terms of her theory or story of her own underlying intentional states (Graham & Stephens, 1994, p. 102). In contrast to the previously mentioned studies which seem designed to capture something about the tacit nature of emotive and other aspects of experience expressed in narrative, Graham and Stephens’ account seems overly intellectualistic. They assume that the subject is in reﬂective control of her narrative, and that she simply discovers some inconsistency and, for that reason, disowns a particular action as genuinely hers. ‘‘On our account, what is critical is that the subject ﬁnd her thoughts inexplicable in terms of beliefs about her intentional states” (1994, p. 105). A second view (Bovet & Parnas, 1993) takes a more bottom-up approach, and points to impairments in self-temporalization at the level of pre-reﬂective ipseity, which then gets reﬂected in narrative. Such disruptions are clear in self-narratives that focus on the patient’s experience. Consider the following bits of narrative. [. . .] there . . .. In the middle of time I was coming from the past toward myself. . . . Before there was a before and after. Yet it isn’t there now. . . . (Cited by Minkowski (1933), p. 286). I get all mixed up so that I don’t know myself. I feel like more than one person when this happens. I’m falling apart into bits . . ..” (Cited by Sass (1995), p. 70). Then I, through this combination of myself projecting into the other person, and the other person in itself, am monitored to react as expected . . . and that happens so rapidly that I, even if I had wanted to, am unable to stop myself (Cited by Sass (1999), p. 334). Consistent with this phenomenological view, Phillips (2003) suggests that in schizophrenia, self-narratives become impoverished and fragmented, or overly focused on the present and what is happening to the subject now – and are often speciﬁcally about their experience of their condition.

S. Gallagher, J. Cole / Consciousness and Cognition 20 (2011) 149–155

151

A third view is represented by Lysaker and Lysaker (2002) who suggest that narrative structure derives from an internal self-dialogue. ‘‘The self is inherently ‘dialogical’, or the product of ongoing conversations both within the individual and between the individual and others” (p. 210). In some cases of schizophrenia, they suggest, narrative becomes monological, or the self-dialogue becomes distorted. Following these differing interpretations there seem to be three approaches to narratives in schizophrenia. 1. We control our narratives; the narrative reﬂects a higher-order (meta-) cognitive mistake (Graham and Stephens) 2. Something at a level more basic than narrative (e.g., at a pre-reﬂective level of ipseity) goes wrong and is simply reﬂected at the narrative level (Bovet and Parnas). 3. Something in the linguistic-narrative-dialogical practice itself goes wrong (Lysaker and Lysaker). In (1) and (2), narrative is a neutral vehicle for either a higher-order or a lower-order disturbance. In (3), it seems, that narrative is complicit in the disturbance, that is, that the narrative is itself pathological. One might think that it should be easy to distinguish between neutral or complicit narratives; or that it would be easy to distinguish between (2) and (3) – a distinction that might parallel a distinction between minimal self, which involves the immediate experience of one’s present embodied situation (ipseity), and narrative self, a temporally extended sense of self involved in a series of events (see Gallagher, 2000). We want to suggest that these distinctions may not be so easy. Accordingly, if one followed the general research strategy of identifying features of narrative structure that might be shifted around or disrupted as the result of living with a given condition, one would still have to work out how to distinguish between these alternative interpretations. This strategy, searching for the ﬁngerprints of pathology in narrative, may not, on its own, distinguish between these possible interpretations. Deciding which is most appropriate may depend on how one views the relation between the minimal self and the narrative self in the narrative itself – and this involves, at least in part, the hermeneutical concept of narrative distance. 2. Narrative distance The concept of narrative distance goes back to Aristotle’s Poetics, and has been used in narrative theory to indicate how far removed the narrator is from the narrated events (Andringa, 1996; Lothe, 2000). It involves a number of different aspects: perspectival distance, evaluative/affective distance, temporal distance, and hermeneutical distance. Perspectival distance: For example, there is less distance between the narrator and the narrated events if the narration is done in the ﬁrst person vs. third person perspective. This suggests less narrative distance in autobiographical narratives where the narrator is using the ﬁrst person. Evaluative/affective distance: Narrative distance measured in terms of the extent and the valence of the narrator’s evaluation of the events, or how the narrator feels about the events. This affective or emotional valence reﬂects how pleasurable or painful the narrated experience was for the narrator. Temporal distance: A third kind of narrative distance, which can be characterized as the distance between the time when the narrator narrates and the time represented by the narrated events. This has to do with both content and structure. With respect to content, if my narrative of events is based on episodic memory, then the limitations imposed by my memory may introduce important discrepancies between what I narrate and what actually happened – a difference between narrated events and historical events. With respect to structure, one ought to consider both internal and external timeframe. Internal timeframe reﬂects serial order within the narrative. One can think of this in terms of McTaggart’s (1908) notion of the B-series, cast in terms of earlier and later. Plot depends on this, as Ricoeur (1992) explains in terms of conﬁgurations of discordance (where a novel event is introduced) and concordance (where an event is integrated with the plot). Even without plot, some ﬁxed order of narrated events are necessary. In contrast, external timeframe refers to the perspectival temporal sense (McTaggart’s A-series, cast in relative terms of past–present–future). Speciﬁcally, from the narrator’s viewpoint, events are either past, future, or present relative to the narrator’s own present time. The narrator has a perspective on the narrated; in self-narrative, the narrator has a perspective on the narrated self. Hermeneutic distance: These various kinds of differences – in perspective, valuation/emotion, and temporality – are factors that can introduce limitations and biases into the recounting of the events. Hermeneutics, in addition, suggests that narrative distance is unavoidable and that all narrative recounting is an interpretation, the veracity of which is measurable in degree, but is never 100%, due to factors such as the narrator’s interest, the intended audience, and of course the ever present sheer impossibility of any narrative description conveying everything in a given scene or event. In every narrative there is selection, even in Joyce’s famously ending Ulysses in a stream of consciousness. Temporal distance, of course is not necessarily a weakness. It is not only in descriptive power and relative eloquence that good biography, like good ﬁction, resides, but in the selection of events and of the pertinent, signal aspects of them. In autobiographical (or self-) narrative one may, speciﬁcally, ask about the distance between the self who narrates and the self who is narrated. Within the context of autobiographical narrative, the narrative self is composed of both the narrating self (the narrator) who in the act of narration, ‘‘here and now,” is telling the story and is being affected by it, and the narrated self, the object (the protagonist) of the narrative. When the narrator says, for example, ‘‘I had a great time in college,” the ‘I’ points

152

S. Gallagher, J. Cole / Consciousness and Cognition 20 (2011) 149–155

in two directions. It points back to the person who is telling the story, the narrator, and signiﬁes that the narrator means to say something about herself; but it also points to the person or character who the narrator was at some point in the past. The narrator thus expresses an identity between herself as-subject and the person she is talking about, but there is always some degree of difference or narrative distance involved. We might want to equate this with a non-pathological psychological dissociation. Even in a ﬁrst-person autobiographical narrative, the person the narrator is today is not necessarily identical to the person she was. In narrative theory, ipseity, as a form of identity, is a concept that allows for a difference between the self I am describing at an earlier (or later) time, and myself as narrator in the present, despite numerical (idem) identity. A person might say, ‘‘I did this in college, but I certainly wouldn’t engage in that activity today. I’ve changed quite a bit.” Furthermore, it’s also quite possible that the narrator doesn’t have it right when she states ‘‘I had a great time in college.” To some extent this will depend on the veracity or selectivity of her memory (perhaps she can only remember the few good things that happened). It will also depend on certain hermeneutical biases generated, for example, by the kind of things she is interested in reporting, or by the kind of effect that she would like her narrative to have, or the kind of audience she is addressing and whether her narrative aims for truthfulness; she may want to appear as someone who had a great time, even if she did not. It seems possible to regard narrative distance and its various aspects as reﬂecting two different kinds of processes involved in narrative practice. Both the perspectival and temporal aspects may be considered functions of the structural features of narrative itself which depend, in turn, on the language or the particular narrative stance taken by the narrator. Something like this correlates with the Lysaker and Lysaker view outlined in the previous section. In contrast, the evaluative/affective and hermeneutical aspects seem to reﬂect something closer to psychological factors that involve the narrator’s experiences, and her interpretation of them in general and how she wants to portray that, and may correlate with the phenomenological (e.g., Bovet and Parnas), or cognitive (Graham and Stephens) view. In an analysis of narrative, however, these different views may not be clearly distinguished, as we make clear in the next section. 3. An empirical measure of narrative distance In this section we set aside questions about the narratives of those with psychopathological conditions, though already there is a slight blur between their narratives and those of ‘neurotypicals’, and we focus on an experiment that compares deceptive and non-deceptive autobiographical narratives in order to clarify some of the interpretational questions. We return to questions about psychopathologies in the next section where we apply the lessons we take from this experiment. To investigate the identity dynamics of the self in narrative, Bedwell, Gallagher, Whitten, and Fiore (in press) examined syntactical structure in 176 short, oral, autobiographical narratives, half of which were intentionally deceptive, and half nondeceptive. In non-autobiographical, intentionally deceptive communication, Anolli, Balconi, and Ciceri (2003) have shown that there is less self-reference, more frequent references to third parties and to objective factors, and greater use of impersonal vocabulary than in truthful communication. Bedwell et al. thus hypothesized that there would be an increase of narrative distance in intentionally deceptive autobiographical narratives compared to non-deceptive autobiographical narratives. If so, then it raises the questions as to why narrative distance might increase and what kind of narrative distance is involved. The hypothesis was addressed using a comparative analysis of syntactic structure in the two kinds (deceptive vs. nondeceptive) of narrative. The study analyzed 176 videotaped oral autobiographical narrative samples obtained via two narrative prompts: (1) ‘‘Describe your immediate family during childhood, including the quality of your relationship with these individuals,” and (2) ‘‘Describe your personality during high school, including examples of situations that occurred during this time period that exempliﬁed your personality.” Participants (50 undergraduate students) were asked to make the deceptive narratives as believable as possible by being told that a blind rater would try to judge which stories were deceptive and which non-deceptive. The participants were given 4 min to prepare their narratives, using written notes as needed. These notes were not allowed during the actual oral narrative delivery. The oral narratives were then transcribed and imported into the software program Coh-Metrix for analysis. Coh-Metrix is a software program that automatically computes several linguistic features, or indices, of cohesion within a text (McNamara, Louwerse, Cai, & Graesser, 2005), e.g., word-level indices like familiarity, syntactic analysis, referential and semantic cohesion indices, and situation model indices (Graesser, McNamara, Louwerse, & Cai, 2004; Graesser, Jeon, Yan, & Cai, 2007). Bedwell et al. identiﬁed seven linguistic variables that showed a statistically signiﬁcant effect for deception vs. nondeception and no statistically signiﬁcant interaction with story type (family vs. high school). 1. Flesch Kincaid Grade Level: A traditional formula for calculating the grade level (0–12) of textbooks. The higher the number, the harder it is to read the text. Deceptive stories had a lower Flesch Kincaid Grade Level than the non-deceptive stories. 2. Number of sentences: Deceptive stories contained more sentences than the non-deceptive stories. Since there was no difference in the number of words used between the deceptive and non-deceptive conditions, deceptive studies contained more, but shorter, sentences. 3. Sentence syntax similarity: Comparison of syntactic tree structures of sentences. Coh-metrix returns the proportion of nodes in the two tree structures that are intersecting nodes. This is an indication of how similar sentences are in structural complexity. Deceptive stories had greater sentence syntax similarity than the non-deceptive stories.

S. Gallagher, J. Cole / Consciousness and Cognition 20 (2011) 149–155

153

4. Explicit action words (‘‘intentional content”): The number of actions involved estimated by counting the number of main verbs that imply intentional or goal-driven action. Examples: put, skip, eat, cut, give, and play. Deceptive stories contained a greater amount of action words/intentional content than the non-deceptive stories. 5. Log min in sentence for content words: Returns the log word frequency value for the rarest word in each sentence and computes the mean across sentences. Scores range from 0 to 6. Lower values relate to the general use of relatively rare words. Non-deceptive stories had a lower value than the deceptive stories, indicating more use of relatively rare words in the truthful stories. 6. Stem overlap: A form of referential cohesion, in which a noun in one sentence has the same or a similar morphological root as a word in another sentence. Non-deceptive stories contained a greater degree of stem overlap than the deceptive stories. 7. Anaphor reference: An anaphor is a word (usually a pronoun) that substitutes for a previously mentioned noun to avoid repetition. Coh-metrix calculates the proportion of anaphor references that refer back to its constituent within 5 sentences. Anaphor reference is another aspect of referential cohesion. Deceptive stories contained fewer anaphor references than the non-deceptive stories. Generally, the intentionally deceptive narratives (vs. non-deceptive narratives) showed: Less referential coherence (measured by anaphor reference and stem overlap). Less complexity (lower Flesch Kincaid Grade Level, higher number of simple short sentences) with similar syntax (sentence syntax similarity) and less rare words (Log Min. in sentence for content words). Signiﬁcantly higher number of action words, i.e., more descriptions of actions, than in the non-deceptive narratives. 4. Interpretations Bedwell et al. interpret the higher number of action words in terms of narrative distance. This interpretation, as they note, can go in two directions. First, on one view, there is a kind of psychological dissociation between the narrating self and the narrated self that could reﬂect dissociation at the level of pre-reﬂective ipseity. Returning to the interpretation of psychopathological narratives, this view would support the phenomenological account provided by Bovet and Parnas. Second, however, narrative distance might result from certain cognitive-linguistic requirements and stay at the linguistic-narrativedialogical level – a view more consistent with Lysaker and Lysaker. On the ﬁrst interpretation, narrative distance may signify a psychological dissociation. Then the higher number of action words (i.e., descriptions of self in terms of actions and external events) in deceptive narratives reﬂects a greater narrative distance (a higher degree of impersonality) between the narrator and the narrated self. One way to think of this is to consider the self described in the narrative as more like another person (not me, not the true me). That is, if in fact I am lying about myself or about what I have done in the past, then in some sense I am describing someone who is not me, and this distance exists between me and someone who is not me. In that case I (the narrator) am giving a description of someone (myself – the narrated self) as another, and from the outside. The easiest way to run a ﬁrst-person description of someone else is not in terms of the other’s mental states but in terms of their actions, as anyone comparing pulp ﬁction with, say, Proust or Henry James will know. So in describing others, actions become the main focus (Genette, 1983). According to this view, the prediction is that my narrations about people other than myself will reﬂect a higher number of action words. That is, they will offer a description more in terms of actions and external events than in terms of the other’s motives or reasons for acting. A study by Krackow (2010) makes a related point. Krackow compared narratives of remembered childhood events to imagined childhood events and found a higher number of mental state terms in the remembered events. Accordingly, the fact that deceptive self-narratives reﬂect a higher number of action words suggests that the narrating self is distancing (dissociating) herself from the narrated self. Bruner’s (1986) distinction between the landscape of action (LoA) and the landscape of consciousness (LoC) can be informative here. LoA narratives describe what happens in terms of a series of actions or external events. LoC narratives describe ‘‘what those involved in the action know, think, or feel, or do not know, think, or feel” (Bruner, 1986, p. 14). An example of LoA: ‘‘I walked to the store to get some candy.” An example of LoC: ‘‘I decided to walk to the store because I wanted to get some candy.” Feldman, Bruner, Renderer, and Spitzer (1990) show, in a study of narrative comprehension in adults, that people easily understand the gist of a story cast solely in the less complex LoA, and it is clearly possible to represent the major events in a drama without always being able to decipher or explain, with full clarity or perhaps at all, the reasons why a protagonist will have acted. The main idea, however, is that if in the case of deceptive autobiographical narrative, I am describing myself as if I were another, and from the outside, then, we would expect the narrative to be framed more in the LoA (more action words) than in LoC (more mental state words). This, however, is not the only interpretation available. An alternative account can be given in terms, not of psychological dissociation, but simply in terms of narrative reﬂecting certain cognitive-linguistic requirements related to less syntactical complexity in the deceptive narratives, which was also found by Bedwell et al. Less referential coherence, for example, may reﬂect the fact that the participants are constructing the deceptive narratives on a more ad hoc basis (they were given only 4 min to construct their narratives) than narratives based on memory of their own activities, which they may have described before. Despite the similarity in the total number of words in deceptive and non-deceptive narratives, the reduced

154

S. Gallagher, J. Cole / Consciousness and Cognition 20 (2011) 149–155

complexity in the deceptive narratives is consistent with previous research showing less detailed information delivered (Vrij 2001; Vrij & Heaven, 1999) and shorter sentences with simpler syntax for deceptive communication (Miller & Stiff, 1993). Even if deceptive narratives themselves are less complex, generating them is more cognitively complex than generating non-deceptive narratives. As Bedwell et al. point out, deceivers must use extra cognitive resources to maintain a coherent and consistent version of a false story. There’s a difference in cognitive complexity between generating autobiographical narratives on the basis of imagination (as in deceptive narratives) than on the basis of memory (as in non-deceptive narratives), presuming for the sake of argument that memories are fairly truthful. In the case of memory, where I have a direct memory of my experience of the past event, I do not require an extra step of identifying who the subject is (Shoemaker, 1959). I know that it was I who experienced X. In the case of imagining myself doing something, however, the structure involves: (1) imagining someone doing X, (2) knowing that I did not do X; but (3) nonetheless identifying that someone as myself. In other words, pretense enters into the deceit as an extra cognitive step. So the cognitive complexity involved in generating a false narrative does not automatically translate into complex narratives. That deceptive narrative may be expressed with a higher syntactical simplicity than the non-deceptive narrative may be viewed as a compensatory effect – a way of balancing out the cognitive complexity. In other words, in order to be able do all the cognitive work involved in deception, the narrator may keep the story as simple as possible. Likewise, if, as in non-deceptive narrative, the cognitive work is not as complex, the narrator may have more resources available to enhance the story or embellish the details. This same reasoning, however, may explain the higher number of action words in the deceptive narratives. Generating LoA narratives requires less cognitive work than LoC narratives, and may be preferred to balance out the higher cognitive demands of lying. This can be seen in the developmental literature. As children gain narrative competency, they begin with narratives that are dominated by actions, and only gradually learn to add intentional language (Hutto, 2008; Nelson, 2009). In addition, to explain one’s reasons for acting may be easier than to explain the motives or reasons of another person. Describing another person’s motives, emotions, intentions, etc. in narrative form seems more complex than describing the actions of others. So the key question may be whether the deceptive narrator uses LoA (more action words) and thereby creates more narrative distance because: 1. The narrator treats the narrated self as if it were someone other (psychological dissociation), or 2. It is a less complex cognitive/syntactic strategy that reduces the cognitive workload (a linguistic artifact). Returning to the narratives of those with psychopathological conditions, there are certainly different narrative dynamics involved. The differences discussed between deceptive and non-deceptive narratives are not the same as those involved in narratives of those with schizophrenia, for example. Yet, similar kinds of considerations about the interpretation of narrative distance can be raised in discussing how to interpret psychopathological narratives. Consider the following self-narrative by a schizophrenic patient. ‘I get all mixed up so that I don’t know myself. I feel like more than one person when this happens. I’m falling apart into bits . . .. I’m frightened to say a word in case everything goes ﬂeeting from me so that there’s nothing in my mind. . . . My head’s full of thoughts, fears, hates, jealousies. My head can’t grip them; I can’t hold onto them. I’m behind the bridge of my nose – I mean, my consciousness is there. They’re splitting open my head, oh, that’s schizophrenic, isn’t it? I don’t know whether I have these thoughts or not (Cited in Sass (1995), p. 70). Are narratives like this reﬂecting a real psychological dissociation that touches on ipseity (‘‘I feel like more than one person . . .”) – the Bovet–Parnas interpretation? Or might it reﬂect a cognitive/linguistic difﬁculty (‘‘I’m frightened to say a word in case everything goes ﬂeeting from me . . .”) – in this case a problem in dialogical structure – the Lysaker–Lysaker interpretation? Or might the quote suggest that expression of what is being felt may lead of itself to a dissolution of self, which is closer again to the Bovet–Parnas interpretation? Or is this either-or view too simplistic? Might we say that both issues – disruptions in both ipseity and dialogical structure – are involved? To distinguish these we may have to look to extra-narrative indications (e.g., empirical testing) or to more quantitative measures. This brings us back to the question of whether narrative itself can be neutral or complicit (pathological). Can one distinguish ‘pathological’ narrative, forms of language which can be shown to have abnormal structures, in verb use, in syntax, etc., from narrative that simply reﬂects the experience of living with unusual conditions. As we have suggested, it is certainly imaginable that someone with, say, depression, may have a perfectly normal use of language, and so use narrative to express their pathological condition. Alternatively, someone with a pure dysphasia may have profound changes in use of language isolated from psychological problems. In more complex and severe psychoses, like schizophrenia, the relations between narrative structure and content and between expression and experience are far more complex. In this paper we have tried to scratch the surface of some aspects of this in relation to a relatively simple empirical experiment on deceptive and nondeceptive narrative. To distinguish psychological dissociation from linguistic and cognitive difﬁculties, if indeed these can be separated in such complex conditions as schizophrenia, we suggest, may require the development and use of other measures which address both the experiential (e.g., Examination of Anomalous Self-experience – EASE (Parnas et al., 2005)), and the expressive (e.g., Narrative Coherence Rating Scale – NCRS (Lysaker, Clements, Plascak-Hallberg, Knipschure, & Wright, 2002; Lysaker et al., 2005)). These may have to be further compared with other measures of cognition, memory, affective

S. Gallagher, J. Cole / Consciousness and Cognition 20 (2011) 149–155

155

valence etc. In schizophrenia in particular, such methods may be useful in the earlier stages of the condition when narrative skills are preserved and an experienced self is relatively coherent. To unravel the various aspects of thought, feeling, memory, lingustic skill and expression remains a large task but one where it may be useful to employ a battery of empirical methods. References Andringa, E. (1996). Effects of ‘narrative distance’ on readers’ emotional involvement and response. Poetics, 23, 431–452. Anolli, L., Balconi, M., & Ciceri, R. (2003). Linguistic styles in deceptive communication: Dubitative ambiguity and elliptic eluding in packaged lies. Social Behavior and Personality, 31(7), 687–710. Bedwell, J. S., Gallagher, S., Whitten, S. N., & Fiore, S. M. (in press). Linguistic correlates of self in deceptive oral autobiographical narratives. Consciousness and Cognition. doi:10.1016/j.concog.2010.10.001. Bovet, P., & Parnas, J. (1993). Schizophrenic delusions: A phenomenological approach. Schizophrenia Bulletin, 19, 579–597. Bruner, J. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press. Caixeta, M., Chaves, M., Caixeta, L., & Reis, O. (1999). Emotionally-loaded narrative increases dyslogia in schizophrenics. Arquivos de Neuro-Psiquiatria, 57(3A), 695–700. Feldman, C. F., Bruner, J., Renderer, B., & Spitzer, S. (1990). Narrative comprehension. In B. K. Britton & A. D. Pellegrini (Eds.), Narrative thought and narrative language (pp. 1–78). Hillsdale, NJ: Lawrence Erlbaum Associates. Gallagher, S. (2000). Philosophical conceptions of the self: Implications for cognitive science. Trends in Cognitive Sciences, 4(1), 14–21. Genette, G. (1983). Narrative discourse: An essay in method. Ithaca: Cornell University Press (trans. J. E. Lewin). Graesser, A. C., Jeon, M., Yan, Y., & Cai, C. (2007). Discourse cohesion in text and tutorial dialogue. Information Design Journal, 15, 199–213. Graesser, A. C., McNamara, D. S., Louwerse, M., & Cai, Z. (2004). Coh-Metrix: Analysis of text on cohesion and language. Behavioral Research Methods, Instruments, and Computers, 36, 193–202. Graham, G., & Stephens, G. L. (1994). Mind and mine. In G. Graham & G. L. Stephens (Eds.), Philosophical psychopathology (pp. 91–109). Cambridge, MA: MIT Press. Gruber, J., & Kring, A. M. (2008). Narrating emotional events in schizophrenia. Journal of Abnormal Psychology, 117(3), 520–533. Hutto, D. (2008). Folk-psychological narratives. Cambridge, MA: MIT Press. Junghaenel, D., Smyth, J. M., & Santner, L. (2008). Linguistic dimensions of psychopathology: A quantitative analysis. Journal of Social & Clinical Psychology, 27(1), 36–55. Krackow, E. (2010). Narratives distinguish experienced from imagined childhood events. American Journal of Psychology, 123(1), 71–80. Lothe, J. (2000). Narrative in Fiction and Film. Oxford: Oxford University Press. Lysaker, P. H., Clements, C. A., Plascak-Hallberg, C., Knipschure, S. J., & Wright, D. E. (2002). Insight and personal narratives of illness in schizophrenia. Psychiatry, 65, 197–206. Lysaker, P. H., Davis, L. W., Eckert, G. J., Strasburger, A. M., Hunter, N. L., & Buck, K. D. (2005). Changes in narrative structure and content in schizophrenia in long term individual psychotherapy: A single case study. Clinical Psychology and Psychotherapy, 12, 406–416. Lysaker, P. H., & Lysaker, J. T. (2002). Narrative structure in psychosis: Schizophrenia and disruptions in the dialogical self. Theory and Psychology, 12(2), 207–220. McNamara, D. S., Louwerse, M. M., Cai, Z., Graesser, A. C. (2005). Coh-Metrix version 1.4. software. Retrieved 08.02.07. McTaggart, J. M. E. (1908). The unreality of time. Mind 17 (New series, no. 68), pp. 457–474. Miller, G. R., Stiff, J. B. (1993). Deceptive communication. Sage Publications. Minkowski, E. (1933). Lived time: Phenomenological and psychological studies. Evanston: Northwestern University Press (trans. N. Metzel, 1970). Nelson, K. (2009). Narrative practices and folk psychology: A perspective from developmental psychology. Journal of Consciousness Studies, 16(6–8), 69–93. Parnas, J., Moeller, P., Kircher, T., Thalbitzer, J., Jansson, L., Handest, P., et al (2005). EASE-scale: Examination of anomalous self-experience. Psychopathology, 38(5), 236–258. Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G. (2003). Psychological aspects of natural language use: Our words, our selves. Annual Review of Psychology, 54, 547–577. Phillips, J. (2003). Schizophrenia and the narrative self. In A. S. David & T. Kircher (Eds.), The Self in Neuroscience and Psychiatry (pp. 319–335). Cambridge: Cambridge University Press. Raffard, S., D’Argembeau, A., Lardi, C., Bayard, S., Boulenger, J. P., & Van der Linden, M. (2010). Narrative identity in schizophrenia. Consciousness and Cognition, 19(1), 328–340. Ricoeur, P. (1992). Oneself as another. Chicago: University of Chicago Press (trans. K. Blamey). Sass, L. (1995). The Paradoxes of delusion: Wittgenstein, Schreber, and the Schizophrenic mind. Ithaca: Cornell University Press. Sass, L. (1999). Schizophrenia, self-consciousness, and the modern mind. In S. Gallagher & J. Shear (Eds.), Models of the self (pp. 319–341). Exeter: Imprint Academic. Shoemaker, S. (1959). Personal identity and memory. Journal of Philosophy, 56(22), 868–882. Stirman, S. W., & Pennebaker, J. W. (2001). Word use in the poetry of suicidal and nonsuicidal poets. Psychosomatic Medicine, 63(4), 517–522. Vrij, A. (2001). Detecting lies and deceit: The psychology of lying and implications for professional practice. Chichester: Wiley & Sons. Vrij, A., & Heaven, S. (1999). Vocal and verbal indicators of deception as a function of lie complexity. Psychology, Crime, and Law, 5, 203–215.

Consciousness and Cognition 20 (2011) 156–171

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The legal self: Executive processes and legal theory q William Hirstein ⇑, Katrina Sifferd Department of Philosophy, Cognitive Science Lab, Elmhurst College, 190 Prospect Ave, Box 113, Elmhurst, IL 60126, United States

a r t i c l e

i n f o

Article history: Available online 4 November 2010 Keywords: Executive processes Legal agency Criminal responsibility Psychopathy Intention Planning Hypnosis

a b s t r a c t When laws or legal principles mention mental states such as intentions to form a contract, knowledge of risk, or purposely causing a death, what parts of the brain are they speaking about? We argue here that these principles are tacitly directed at our prefrontal executive processes. Our current best theories of consciousness portray it as a workspace in which executive processes operate, but what is important to the law is what is done with the workspace content rather than the content itself. This makes executive processes more important to the law than consciousness, since they are responsible for channelling conscious decision-making into intentions and actions, or inhibiting action. We provide a summary of the current state of our knowledge about executive processes, which consists primarily of information about which portions of the prefrontal lobes perform which executive processes. Then we describe several examples in which legal principles can be understood as tacitly singling out executive processes, including principles regarding defendants’ intentions or plans to commit crimes and their awareness that certain facts are the case (for instance, that a gun is loaded), as well as excusatory principles which result in lesser responsibility for those who are juveniles, mentally ill, sleepwalking, hypnotized, or who suffer from psychopathy. Ó 2010 Elsevier Inc. All rights reserved.

1. Legal agency and the brain There is currently a huge gap between the understanding our legal system has of the human self as rational, responsible decision-maker and the understanding of the human brain offered by neuroscience. The civil and criminal legal systems refer to the mind and mental processes when they speak of peoples’ intentions, plans, motives, and beliefs. Not all mental states or processes are equally important to the law, however. Our contention in this article is that a close reading of legal principles shows that when they direct attention to the mind, they focus primarily on the brain’s prefrontal executive processes. Executive processes are the seat of a person’s decision-making, intention-forming, planning, and behavior-inhibiting processes, all of which are absolutely crucial to his legal and ethical being. Hence we call the set of executive processes ‘‘the legal self.’’ We will argue that neuroscience is an important tool for understanding the connection between mental processes and legal, or illegal, actions. We will also begin to establish correspondences between knowledge of the brain and established legal principles. This project can ultimately result in more just and appropriate verdicts, by clarifying the nature of human intention, planning, and decision-making and their disorders, especially those that are beyond the normal person’s ability to correct. The brain’s executive processes have been repeatedly singled out, albeit tacitly or non-explicitly, as the grounds of agency and responsibility by legal theorists and the law itself. We will argue that when referring to mental states and processes associated with illegal actions, the law is referring to these executive processes in the same way that a person who

q

This article is part of a special issue of this journal on ‘‘Brain and Self: Bridging the Gap’’.

⇑ Corresponding author. Address: Department of Philosophy, 190 Prospect Ave., Box 113, Elmhurst, IL 60126, USA. Fax: +1 630 617 4609. E-mail address: [email protected] (W. Hirstein). 1053-8100/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2010.10.007

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

157

refers to water is also referring H2O molecules, whether she knows it or not. Legal agency, including the capacity to make contracts and commit crimes, requires one to be able to form certain mental states and to act upon them. Similarly, intentions are vital to the formation of a contract: contracts are the legal crystallization of mutual intentions. Legal questions about agency, intent, and responsibility involve our conceptions of ourselves and of how our minds and our actions relate. The legal self is important, both for individuals who possess them, as well as the societies in which they are embedded. We will begin with an examination of the views of legal scholar Stephen Morse, according to whom the legal system focuses on consciousness and rationality. Our response to this is that the focus is more on our executive processes than on consciousness itself. Our current best theories of consciousness portray it as a workspace in which executive processes operate (Baars & Steven, 2005), but what is important to the law is what is done with the workspace content rather than the content itself. This makes executive processes more important to the law than consciousness itself, since they are responsible for channelling conscious decision-making into intentions, plans, and actions, or inhibiting those actions. On the topic of rationality, our views are closer to Morse, but we will argue that both our everyday and legal concepts of rationality can be captured by the concept of proper functioning of our executive processes. Section three details the current state of our knowledge about executive processes, which consists primarily of information about which portions of the prefrontal lobes perform which executive processes. Then our ﬁnal section, section four, contains several examples in which legal principles can be understood as singling out executive processes, including principles regarding defendants’ intentions or plans to commit crimes and their awareness that certain facts are the case (for instance, that a gun is loaded), as well as principles about the competence and/culpability of persons who are juveniles, mentally ill, sleepwalking, hypnotized, or who suffer from psychopathy.

2. The legal self as the rational self A survey of American jurisprudence over the last two centuries reveals a continuing articulation of legal agency in commonsense psychological terms (Blumenthal, 2007). Legal agency, including the capacity to make contracts, and the capacity to commit a crime, require that one be able to form certain mental states and act on them. For example, one must be capable of forming the intention to create a binding last testament or will, and then knowingly complete the legal procedures to formalize those intentions. Susan Blummenthal calls these minimal psychological requirements for legal agency the ‘default legal person’: ‘‘To be sure, the default legal person was a shifty character, fading into the background of many judicial opinions and appearing in different guises as he moved across doctrinal ﬁelds. . .. But all the while, and arguably to the present day, the default legal person has served the same basic function: establishing the relationship between mental capacity and legal responsibility in any given case’’ (Blumenthal, 2007)(1149). Blummenthal argues that the law tends to elucidate these psychological or mental capacities in terms of consciousness and reason. ‘‘Apparent rationality’’ is not sufﬁcient for legal capacity; in addition, an action must have been the product of the actor’s conscious choice (Blumenthal, 2007)(1175). ‘‘[T]he default legal person. . .was capable of understanding the nature and consequences of his actions, and freely determining how to proceed on the basis of this knowledge. An individual shown to be in possession of these basic mental attributes would be held accountable for his actions. . .’’ (Blumenthal, 2007)(1175). Morse – whose work focuses on the intersection of law and psychology – articulates a modern legal self that agrees with Blummenthal’s formulation. Morse argues that legal agency is primarily concerned with the ability to reason: that is, to be a legal agent one must be capable of recognizing and acting for good reasons (Morse, 2003, 2006a, 2006b). This capacity for reason is often visible, or apparent, to both the agent and the law via the agent’s consciousness. Most obviously, an agent reveals his or her conscious beliefs and desires through speech – verbal or written – and this speech is introduced to the court through witness testimony or written documents. In other cases, a court may be asked to extrapolate from behavioral evidence to a person’s conscious states; e.g., in cases where a jury is asked to attribute motive to a criminal defendant from evidence that he was observed purchasing the murder weapon. According to Morse we evolved conscious rationality precisely because we are social creatures who use this ability in addition to a set of rules to live cooperatively (Morse, 2003)(60). Morse states that ‘‘If the criminal law operates by guiding the conscious actions of persons capable of understanding the rules and rationally applying them, it would be unfair and thus unjustiﬁed to punish and to inﬂict pain intentionally on those who did not act intentionally or who were incapable of the minimum degree of rationality required for normatively acceptable cooperative interaction. People who lack the capacity for rational guidance are not morally responsible and should not be held criminally culpable’’ (Morse, 2003)(61). According to Morse, consciousness itself can be ‘‘integrated’’ or diminished, and can thus indicate a defendant’s psychological capacity or incapacity. ‘‘Law and morality agree that if an agent’s capacity for consciousness is non-culpably diminished, responsibility is likewise diminished’’ (Morse, 2003). At times, Morse appears to privilege consciousness: he claims consciousness can be diminished either because action without consciousness is not deemed to be action, or because diminished consciousness reduces the capacity for rationality.1 It is clear, however, that the link between consciousness and responsibility depends upon rationality. Where consciousness is diminished, moral and legal rules fail to provide an agent with reasons for action in the normal way.

1

Or, it might seem, the other way round: the capacity for rationality is reduced, and this is expressed via a diminished consciousness.

158

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

Thus, Morse and other legal scholars conceive of excusing conditions for legal liability as cases where an agent is not fully conscious (in the normal way) or capable of rational thought. Legal insanity is an obvious case. The M’Naughten rule states that a person is legally insane if he or she is ‘‘laboring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong’’ (Rex v. M’Naughten, 1843). If a defendant truly believes, for example, that her son is about to be abducted by the devil, and that throwing him out a 10th-story window was the only way to save him from eternal hellﬁre, then she is thought to lack a normal understanding of the act’s circumstances and consequences. In this way, she is incapable of rationally guiding her behavior. Morse acknowledges that the capacity for rationality is physically instantiated, most likely in the human brain. However, he is hesitant to use psychology or neuroscience as a means of evidence of this capacity (or its lack) in defendants. Although ‘‘there is accumulating evidence that various psychological processes have their biological substrates in localized regions in the brain,’’ Morse believes that this evidence doesn’t speak to the capacity for rationality or consciousness per se, because the way the brain enables the mind is still unknown (Morse, 2008)(28). Although Morse describes himself as ‘‘a thorough-going, matter-up materialist who believes that all mental and behavioral activity is the causal product of lawful physical events in the brain,’’ (Morse, 2006b) he is a non-reductive physicalist who believes that the way in which the brain enables consciousness remains ‘‘fundamentally mysterious.’’2 He thus argues that neuroscience, at least in its current state, should not be used in criminal trials. Instead, the court should stick to direct behavioral evidence to determine if a defendant lacks of the minimal consciousness or rationality required for responsibility. The writings of Morse tend to run both consciousness and rationality together: when he claims that a responsible agent has a ‘‘fully integrated consciousness;’’ what he means is that the person ‘‘is capable as acting for reasons’’ (Morse, 2007). He sometimes mentions rationality, and other times consciousness, as diminished or faulty in cases where an agent is not fully responsible. We agree that the formulation of the legal self as the rational self is in a sense correct, and we accept as evidence of this that the law has had much success in categorizing persons with regard to criminal responsibility or legal capacity. However, we disagree that consciousness and rationality are interchangeable, and that consciousness alone is as reliable an indicator of the legal self as rationality. In addition, by de-emphasizing the role of consciousness we can keep the vexing problems associated with it from preventing us from making progress toward a neuroscientiﬁc articulation of the substrates of the legal self. Our approach allows for a clear distinction between consciousness itself, and the executive processes that operate on its contents. One may wonder why a neuroscientiﬁc understanding of the legal self is necessary, as we have already indicated that the law’s use of a folk understanding of the self as a rational actor is at least adequate to describe what mental states and processes are relevant. As the recent increase in the use of scientiﬁc psychology in the law shows, the folk approach has its limitations that science can be used to correct. Commonsense psychology, the tool judges and juries use to attribute mental states (and thus rationality), operates via something close to cognitive heuristics whereby mental states are attributed based on theoretical assumptions regarding human behavior and outward behavioral cues.3 Thus, as trial attorneys know well, it can be manipulated. Criminal verdicts are subject to a variety of arguably unreliable inﬂuences and behavioral data that push and pull commonsense mental state attributions in different directions. For example, by not allowing juvenile offenders to shave, and by dressing them in clothes a size too small, criminal prosecutors may make their request to transfer the juvenile to adult court more convincing. Courts are increasingly turning to expert witness testimony and scientiﬁc psychology to shore up their commonsense notions and better classify defendants, especially those with diminished legal capacity. For example, in the 2005 Supreme Court case Roper v. Simmons, the US Supreme Court barred capital punishment of juveniles who killed while they were under the age of eighteen (2005). In the US most states use age together with the severity of the offense, in conjunction with other factors, when determining whether a juvenile defendant should be transferred up to adult court (Butts & Mitchell, 2000). Consideration of age appears to be a heuristic for mental capacity; but consideration of severity of offense is clearly not a good indicator of capacity, and appears to be used because states consider juveniles who commit particularly violent crimes more deserving of retribution, despite their mental capacity. Other factors used to transfer juveniles are also far-fromsatisfactory heuristic methods of determining mental capacity. For example, under California’s Welfare and Institutions Code 707(a), the juvenile judge must evaluate the degree of ‘criminal sophistication’ exhibited by the child. Amicus Curiae briefs ﬁled prior to the Roper decision strongly encouraged the court to utilize evidence from neuroscience in their assessment of juvenile culpability. The American Medical Association, the American Bar Association, the American Psychiatric Association, and the American Psychological Association all ﬁled or subscribed to amicus briefs urging abolition of the juvenile death penalty based in part on the neuroscientiﬁc ﬁndings (Morse, 2006). Morse, however, argued that the neuroscientiﬁc evidence offered by the amici in Roper added nothing to the behavioral evidence already available to the court, claiming that the scientiﬁc evidence just served to distract the court from the real problem of assessing juvenile capacity for rationality in commonsense terms (Morse, 2006). In the end, the Roper majority opinion cited much scientiﬁc psychology,

2 Morse states that ‘‘Most fundamentally, action and consciousness are scientiﬁc and conceptual mysteries. We do not know how the brain enables the mind, and we do not know how action is possible.’’ (2007, p. 2555). 3 See Sellars, W. (1956). Empiricism and the philosophy of mind. In H. Feigl & M. Scriven (Eds.), The foundations of science and the concepts of psychoanalysis, Minnesota studies in the philosophy of science (Vol. 1, pp. 127–196). Minneapolis, MN: University of Minnesota Press. Fodor, J. (1987). Psychosemantics: The problem of meaning in the philosophy of mind. Cambridge, MA: MIT Press. Dennett, D. (1991). Real patterns. Journal of Philosophy LXXXVIII, 27–51.

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

159

some of it based on neuroscience. Noting that juveniles are more subject to peer pressure and that they have less control over their environments, Kennedy then cited an article by Steinberg and Scott which noted that brain development in regions implicated in judgment and decision-making continue to develop well into late adolescence (Steinburg & Scott, 2003). Other cases where scientiﬁc psychology has been used to inform designations of legal capacity in criminal defendants include cases of mental retardation (Atkins v. Virginia, 2002); legal insanity (Panetti v. Quarterman, 2007); sleepwalking (R v. Parks, 1992); and psychopathy (People v. Dugan, 1985). As Nicole Vincent argues, ‘‘. . .To be a responsible moral agent one must have the right mental capacities, but since mental capacities are implemented in brain mechanisms (in brain ‘‘hardware’’), to be a responsible moral agent one must have the right brain mechanisms. . .’’ (Vincent, 2008). This trend in criminal law supports our project of describing the legal self in neuroscientiﬁc terms: while analysis of consciousness and rationality works well, in many cases it isn’t the best tool in our modern toolbox for categorizing legal agents. Below, we ground the trend toward the legal system’s use of scientiﬁc psychology by providing a theoretical framework for a neuroscientiﬁc description of the legal self. 3. An alternative approach At the moment, the sciences of the mind and brain (these include neuroscience, psychology, linguistics, philosophy, and portions of computer science), can be divided into two competing paradigms: the classical approach, centered around artiﬁcial intelligence and the computer model of the mind; and a new competitor, the neuroscience-based approach that employs anatomical and physiological descriptions, linked to behavior via cognitive neuropsychology. Despite some major differences between the two paradigms, there is an emerging consensus on the idea that cognition can be thought of as involving two primary components: representations, and capacities for manipulating those representations. A distinction between representations and processes performing functions on them is congenial to cognitivist approaches in psychology and cognitive neuroscience. For instance, Sternberg says that, ‘‘underlying most cognitive approaches to intelligence is the assumption that intelligence comprises mental representations (such as propositions or images) of information and processes that can operate on such representations’’ (Sternberg, 2010). It is much less widely appreciated, however, that neuroscience has lately been coming around to a very similar view. Neuroscientists speak of executive processes as those brain processes that act on and manipulate representations contained in a conscious workspace. This suggests a simple and intuitive way to think of the relation between consciousness and rationality. We are conscious of representations, but to be rational is to use your representations in certain ways. This distinction is crucial for the legal realm in that we hold people responsible for what they do with their thoughts and representations, rather than for them alone. 3.1. Consciousness and executive processes Baars has developed a cognitive theory of consciousness according to which ‘‘conscious experience in general can be viewed as information presented to prefrontal executive regions for interpretation, decision-making and voluntary control’’ (Baars, Ramsøy, & Laureys, 2003)(673). Baars and his co-authors say that, ‘‘conscious experience emerges from a nervous system in which multiple input processors compete for access to a broadcasting capability; the winning processor can disseminate its information globally throughout the brain’’ (Baars, 1993)(282). Representations are held in consciousness so that they can be further processed by any of a number of executive processes. Recently, neuroscientists have suggested that something of the sort Baars posits could be accomplished by the working memory areas located in the dorsolateral prefrontal cortex coupled with multimodal sensory integration areas in the posterior of the cortex (Dehaene et al., 2001). The executive processes exist in the dorsolateral prefrontal cortex, while the representations they manipulate exist in posterior sensory areas. If consciousness is a workspace where representations are held online and manipulated, what manipulates them? A great deal of what we normally call thinking, deciding, planning and remembering is accomplished primarily by the brain’s executive processes. There is an ongoing debate about how exactly to parse the set of executive functions, but the following tend to appear in most lists: attention, remembering, decision-making, planning, intending, and inhibiting. Gilbert and Burgess (2008)(10) say that, ‘‘executive functions are the high-level cognitive processes that facilitate new ways of behaving, and optimise one’s approach to unfamiliar circumstances.’’ Executive processes play a part in all non-routine actions. When we attempt something new, executive processes are required; they are needed when there are no effective learned input–output links. As we get better at the new task, processing moves to other brain areas that specialize in efﬁciently performing routine actions without conscious interruption. Routine actions, some of which exist in the brain’s procedural memory, do not require executive activity to operate. As Miller and Wallis pithily state it, ‘‘You do not need executive control to grab a beer, but you will need it to ﬁnish college’’ (Miller & Wallis, 2009). According to Gilbert and Burgess, ‘‘we particularly engage such processes when, for instance, we make a plan for the future, or voluntarily switch from one activity to another, or resist temptation; in other words, whenever we do many of the things that allow us to lead independent, purposeful lives’’ (Gilbert & Burgess, 2008). Most executive processes reside in the prefrontal lobes, including the dorsolateral frontal lobes on the side of the brain, the ventrolateral frontal lobes below them, the medial prefrontal lobes, on the inner surfaces of the two hemispheres, and the orbitofrontal lobes located on the brain’s undersurface just above the eye sockets (Moscovitch & Winocur, 2002). Our

160

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

information about executive processes comes from three separate data sources: First, the study of brain anatomy, especially its connectivity. Second, humans who have prefrontal lesions have been carefully studied to determine which mental functions have been compromised by the damage. Third, brain imaging. Subjects are given tasks requiring executive intervention, and their brains are imaged as they solve the tasks. One introspectively accessible measure of the amount of executive activity is our sense of mental effort. Increased mental effort correlates with increased usage of oxygen by executive areas, which is detectable by brain imaging. 3.2. Executive functions Our understanding of the brain’s cortex has progressed from back to front. The ﬁrst areas we were able to clearly assign function to were the unimodal perceptual areas, whose responses could be mapped to the sense organs themselves. We have now reached the front of the brain, which contains its highest-level mental operations, achieved by executive processes. If we read the reports of brain anatomists and physiologists who investigate executive processes, we can begin to see the familiar mental functions we know both via our own minds and via commonsense psychology. As we noted above, executive processes are required for non-routine cognition. A portion of the anterior cingulate cortex, in the medial prefrontal cortex located on the inner surface of the two hemispheres, is frequently found to be active during effortful processing. The anterior cingulate is thought to play a role in resolving conﬂicts between routine actions that are not relevant to the present task, and novel actions that are relevant. It is located just above the front of the corpus callosum, and connects to several parts of the prefrontal cortex in addition to the orbitofrontal cortex, including dorsolateral and medial areas. In their review of the anatomy and function of the anterior cingulate cortex, Devinsky et al. say that it ‘‘assesses the motivational content of internal and external stimuli and regulates context-dependent behaviors’’ (Devinsky, Morrell, & Vogt, 1995)(279). They divide the anterior cingulate into an affect region, which assesses ‘‘motivational content,’’ and a cognitive region, which goes into action during ‘‘cognitively demanding information processing,’’ and which also has dense connections to motor areas. Areas 32 and 24 are also important for error detection (Carter, Braver, et al., 1998; Garavan, Ross, Murphy, Roche, & Stein, 2002). They activate strongly when experimental subjects detect errors in their responses (Carter et al., 1998), and they then interact with behavioral inhibition areas in the dorsolateral prefrontal cortex (i.e., areas 44 and 9) to effect an inhibition of the erroneous response. 3.2.1. Attention We perceive the world across a large bandwidth, but we need to focus on certain parts of the incoming ﬂow in order to maximize the effectiveness of our behavior. There are at least two different attentional systems. The ﬁrst is ‘‘rapid, saliencydriven and bottom-up’’ whereas the second is ‘‘slower, volitionally controlled and top-down. Each form of attention can also be more diffuse or more focused’’ (Crick & Koch, 2003). Top-down attention is an executive process and has a focusperiphery structure that is most obvious in the visual modality, but is present in all of them. In contrast, bottom-up attention is more evenly distributed throughout the visual ﬁeld, for instance in our ability to detect moving objects (Koch, 2004), although as noted above, even this form of attention can be focused. The executive portion of the anterior cingulate cortex fulﬁlls two of the roles that an area embodying attention should perform. It receives input from both sensory and mnemonic systems and it also shows activity in brain imaging studies consistent with a role in attention. The posterior portions of the dorsolateral prefrontal cortex (i.e., areas 8 and rostral 6) also support top-down attentional executive processes. We direct attention not only to portions of the incoming perceptual ﬂow, but also to our own memories. The brain apparently uses some of the same tricks to do both. Barbas, Ghashghaei, Rempel-Clower, & Xiao, 2002 say that the role of the orbitofrontal cortex in selective attention to memories is ‘‘analogous to the functions of the frontal eye ﬁelds in directing [visual] attention and medial area 10 in directing auditory attention to currently relevant stimuli’’ (Barbas et al., 2002). 3.2.2. Recognition It is important to note that the executive processes can participate in the act of recognition itself, and that they normally have the power to overrule our initial perceptual identiﬁcations. For example, when you seem to hear someone in your house at night, executive processes can take control and manage the gathering of additional information – for instance, by eliciting a memory that a tree branch hits a certain window when the wind blows – and use this memory to negate the thought of burglars. This requires holding the memory online and directing attention back and forth between it and attention to the sound outside. We might thus distinguish between initial identiﬁcation and considered identiﬁcation. We all experience strange perceptions at times, but we are able to correct them using executive processes. I believe I see my friend during a trip to Nepal, but then I realize how improbable that is—this friend never travels, has no interest in Nepal, etc.—so I do not allow myself to recognize that person as my friend. 3.2.3. Memory When you remember, especially something from your personal past, you often provide a cue for your memory to work from. When you want to remember someone’s name, for instance, you picture her face. When you are trying to recall a speciﬁc event, you will typically recall related events, in order to better reconstruct the event you are interested in. We can also check memories against other memories, of both the autobiographical and semantic variety. You believe you remember seeing Bill Clinton give a campaign speech in August of 1990, but then a check against your semantic memory conﬂicts with this,

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

161

since the presidential campaign would not have commenced yet. It must have been August of 1991, you conclude. Executive processes used the existing memories to cue other memories, and managed the resolution of the inconsistency. Correctly reporting events from memory does not simply involve reading off what the memory record contains. It is a reconstructive process, involving interaction between executive processes and the memory store. Johnson and her colleagues distinguish between heuristic checking of candidate memories, which is usually operating when we are remembering, and systematic checking which is ‘‘engaged selectively and deliberately.’’ Heuristic processing ‘‘uses readily available information (e.g., familiarity), including qualities (e.g., perceptual detail) and schemas (e.g., world knowledge, stereotypes) activated by a cue’’ (Johnson, Hayes, D’Esposito, & Raye, 2000)(362). Systematic processing involves several different executive functions. It requires selective attention: the person must explicitly attend to the candidate memory. 3.2.4. Decision making Studies support the idea that executive functions are essential to decision-making by showing that damage to certain executive systems can produce poor decision-making. For instance cognitive neuropsychologist Antonio Damasio’s patient E.V.R., who had ventromedial frontal damage due to a brain tumor, failed to use past experience and knowledge of what was likely to happen in the future when he made decisions, with disastrous results including failed marriages and businesses (Damasio, 1995). In general, E. V. R. failed to inhibit unwise actions. Shimamura notes that thinking of the function of the prefrontal lobes as ﬁltering, or inhibiting ‘‘may also be useful in explaining the kinds of social disinhibition observed in patients with orbital prefrontal lesions. . .. Such patients exhibit a failure to inhibit inappropriate social behaviors (e.g., telling crude jokes, emotional rages)’’ (811). Duncan and Owen argue that mid-dorsolateral (the upper surface of the frontal lobes), mid-ventrolateral (the lower surface), and dorsal anterior cingulate cortex form a system that accomplishes several executive tasks, including resolving response conﬂicts, managing working memory, and responding to perceptual tasks (Duncan & Owen, 2000). Areas 9 and 46 (jointly known as mid-dorsolateral prefrontal cortex) function to monitor and manipulate information held in working memory (Moscovitch & Winocur, 2002). Lesions to this area in the brains of monkeys impairs their ability to track and keep in mind multiple stimuli (Petrides, 2005)(5649). Owen and his colleagues (Owen, Evans, & Petrides, 1996) have theorized that the ventrolateral prefrontal cortex serves to activate and hold online information located in posterior cortices (D’Esposito, Postle, et al., 1999), while the dorsolateral prefrontal cortex monitors and manipulates this information. Area 47/12 of the ventrolateral prefrontal cortex subserves the conscious, explicit encoding and retrieval of information into and from memory. This involves ‘‘processing initiated under conscious effort by the subject and guided by the subject’s plans and intentions’’ according to Petrides (2005)(791). 3.2.5. Action and its planning The executive system has several output systems that it uses to create effective actions. ‘‘All areas of the prefrontal cortex have access to specialized motor control systems’’ (Barbas et al., 2002). Action planning is an important part of decisionmaking. Planning involves devising effective sub goals, and recalling task-relevant information, all of which is managed by executive processes. Another important feature of planning is that the sequence of actions must be carefully worked out. Miller and Wallis (2009) provide the example of a patient with a frontal lobe tumor who ﬁrst stirred her coffee, then added milk. Executive processes are required to disengage from one task, and engage in another. Task switching is currently one of the experimental paradigms of choice for scientists who study executive processes. The classic experimental test of this is called the Stroop test. The subject must say aloud which color ink each word in a list is printed in. The difﬁculty is that the words themselves are the names of colors, so that the word ‘‘blue,’’ for instance, is printed in yellow ink. Executive processes are required in order to inhibit ourselves from simply reading the word, and answering ‘‘blue.’’ This produces a slight, but measurable, delay in our responses, as well as a sense of increased mental effort (Miller & Wallis, 2009). The frontal pole is at the highest-level of processing and appears to play a role in keeping track of more complicated cognitive tasks, ensuring that the steps needed to accomplish them are executed in the right order. After reviewing several theories of the function of area 10, Ramnani and Owen (2004) conclude that what they all share is the idea that this area is responsible for coordinating information processing and transfer among different prefrontal areas (Ramnani & Owen, 2004). More speciﬁcally, they argue that area 10 becomes involved when two or more separate cognitive operations must be performed to accomplish a task. Area 10 keeps track of these operations, for instance by ensuring that they are performed in the correct order. Barbas and Zikopoulos also offer an account consistent with this: ‘‘Area 10 has a key role in complex working memory tasks that require juggling of multiple tasks. . ., such as interrupting reading a book to answer the telephone then remembering to resume reading at the place of departure’’ (Barbas & Zikopoulos, 2007)(538). Petrides says that, given its anatomical constitution and connectivity, area 10 ‘‘is in an ideal position to monitor the monitoring process in the middorsolateral prefrontal cortex, namely to engage in what might be called ‘hyper-monitoring’. This sort of hyper-monitoring would be ‘‘critical in multi-tasking and high-level planning’’ (Petrides, 2005)(790). Burgess suggests that area 10 working together with the right dorsolateral prefrontal cortex is involved in the ‘‘creation and maintenance of intentions’’ (Burgess, 2000)(470). When executive systems are damaged, actions seems to be generated bottom-up rather than top-down (Humphreys, Forde, & Francis, 2000). The orbitofrontal cortex is the highest-level area of the ventral cortical functional network, and one of the ultimate destinations for the ventral visual processing stream, the stream that functions primarily to allow us to detect, categorize, and associate reward values with objects, and people. Barbas and Zikopoulos describe area 13 as having ‘‘a role in evaluating the

162

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

signiﬁcance of stimuli’’ (Barbas & Zikopoulos, 2007)(540). We also noted above that orbitofrontal cortex plays a role in inhibiting imprudent or socially inappropriate actions. 4. Points of connection: Neuroscience and the law In this section we will describe several points at which legal principles are implicitly directed at the executive processing capacities of agents. These include the voluntary/involuntary distinction, the delineation of mental states required for criminal responsibility, the troublesome case of sleepwalking, and other possible justiﬁcations or excuses to criminal responsibility. 4.1. The voluntary act and mental state requirements To be found guilty of a crime in the United States a defendant must be found to meet two criteria: he must have committed an act deemed criminal, and he must possess a certain mental state with regard to this act. The ﬁrst of these criteria is called the actus reus (which literally means guilty act) requirement. The actus reus requirement is met only if a defendant is found to have committed a criminal act voluntarily. The second criterion is called the mens rea (which means guilty mind) requirement, which is met if the defendant has certain mental states with regard to the criminal act. The criminal law requires an act or omission that comprises the physical elements of a crime as stipulated by statute, this is the actus reus requirement. But while this requirement is primarily concerned with a defendant’s act, it also refers to the defendant’s mental states: an act is voluntary only if it has some minimal connection to the defendant’s goal states or desires. An example of an involuntary act is a case in which a person has an epileptic seizure while driving (with no history of such seizures, and thus no reason to suspect one would occur). If the seizure causes the driver to swerve off the road and kill a pedestrian he would not be guilty due to a failure to the meet the actus reus requirement. Similarly, if a malevolent person trips a jogger who then falls onto an elderly lady, causing her to break a hip, the jogger is not responsible due to the lack of a voluntary act. As prominent legal scholar HLA Hart has noted, an unconscious act or an act by reﬂex is not really a human action at all, and thus for an act to be culpable there must be a ‘‘minimum link between mind and body’’ (Hart, 1968). Legislators and the courts have largely deﬁned the voluntary act condition by exclusion. According to US Model Penal Code (MPC), a statutory text developed to assist state legislatures in updating and standardizing their criminal codes, the following cannot be considered voluntary acts: a reﬂex or convulsion, a bodily movement when unconscious or asleep, conduct during hypnosis, and movement not produced by effort of the actor (e.g., someone is forced to hold a gun and then her ﬁnger is pushed against the trigger by someone else’s hand).4 Other possibilities, depending on jurisdiction, include movements while ‘‘brainwashed,’’ during dissociation or automatism not during sleep (e.g. unconsciousness caused by severe bodily injury), and movements that are part of a habitual act. In cases of reﬂex, convulsion, or where the bodily movement is not produced by effort of the actor, there is no relationship between the bodily movement that caused harm and the actor’s desires. In the case of hypnosis, one might argue that the bodily movement that causes harm is actually linked to the hypnotist’s desire, not the actor’s. The second requirement for criminal culpability is the mental state, or mens rea, requirement. Even where it is determined that a criminal act has been committed voluntarily, in most cases (those not involving strict liability) there must be some speciﬁc relationship between a defendant’s desires and the criminal harm caused.5 Imagine stumbling over a person passed out in the park at midnight. It might be the case that you desired to put your foot exactly there, in that place where the man was, and that as a result of this desire you injured him. However, even though what you did was voluntary, you are not guilty of battery because you didn’t ‘‘intentionally apply unlawful force’’ – that is, you didn’t desire to cause him harm, even though you desired to put your foot there.6 In the case of murder, a person might pull the trigger of a gun and thus take a human life – fulﬁlling the actus reus requirement for murder – but not intend or desire to take a human life. Instead he might have, under extreme conditions, reasonably believed he was shooting a bear. In both cases, the act requirement seems to be met for the particular crime, but the mental state requirement is not. The mental state requirement also relates to a defendant’s beliefs or informational states. For the requirement to be met, in many cases a defendant must: (1) possess speciﬁc desires to perform an act, (2) possess those desires within certain circumstances of which he is aware, and (3) know that the act, once performed, will have certain consequences. For the most severe crimes like ﬁrst-degree homicide, there is often a further requirement that the defendant (4) desire that the postulated consequences will actually occur. For example, the act requirement for theft requires that the defendant take possession of property that does not belong to him. The mental state requirement for theft calls for two additional mental states: that the defendant know he is not entitled to the property, and that he have the desire to deprive the rightful owner of possession permanently.

4

(1985). Model Penal Code and Commentaries (Ofﬁcial Draft and Revised Comments). §2.01. Hereafter MPC. Strict liability offenses are those where, for policy reasons, the standard for proving culpability has been lowered so a mental state element is not required. A common example is statutory rape: a person can be found guilty of statutory rape even if he didn’t know the person he was having sex with was underage. As a matter of policy, the law asks persons to bear the burden of ﬁnding out the age of their partner before they have sex. 6 See MPC §2.02. 5

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

163

The two most important mental concepts that form a bridge from the legal concepts above to the language of neuroscience are intention and planning. As we will discuss below, the MPC standards requiring defendants to engage in acts purposefully and knowingly both involve having speciﬁc intentions to commit a particular act, and some planning to perform that act. Neuroscientists do not shy away from using either term, and their uses are largely consistent and coherent, in that they ﬁt our ordinary uses of the terms. If Burgess is correct in stating that area 10 in the polar prefrontal cortex, as well as the right dorsolateral prefrontal cortex, are involved in the ‘‘creation and maintenance of intentions’’ (Burgess, 2000)(470), legal theorists should be interested in these two areas. Planning also involves recalling items from memory, verifying them, and merging them with other representations and plan segments already in consciousness. This would require activation of executive processes to verify the recalled memories. It would involve the ventrolateral cortex to hold relevant information, contained in the posterior cortices, online while the dorsolateral cortex, under the direction of the polar cortex, manipulated that information. While all this is happening, executive sectors of the anterior cingulate are monitoring the process for errors or inconsistencies, capable of producing a strong alerting signal via the autonomic system when one is found. Disorders of this planning and decision-making process are relevant to a defendant’s guilt and degree of responsibility. Now the crucial questions are directed at the person’s ability to notice and correct for any of these disorders. 4.2. Purposely, knowingly, recklessly, negligently The Model Penal Code has classiﬁed the types of mental states required for criminal responsibility into four categories, which generally correspond to different levels of culpability. The categories refer speciﬁcally to the relationship between the defendant’s desires or beliefs and the harm caused by a criminal act. An act can be done purposely, knowingly, recklessly, or negligently.7 According to the MPC, ‘‘purposely’’ means that the defendant consciously desired that his act have a particular result.8 ‘‘Knowingly’’ means that the defendant was aware that his act would (very likely) have a particular result, even if he didn’t directly desire the outcome.9 This distinction may or may not be relevant with regard to a particular crime. For example, one might think that this distinction would matter for the purposes of determining level of culpability in a homicide, but it does not. A case where a defendant lit a house on ﬁre with the intention of killing the person sleeping inside, and a case where the defendant lit the house of ﬁre to scare the occupant (knowing there was a good chance he would not escape the ﬁre) are both likely to constitute ﬁrst-degree murder. One example of a crime that requires the defendant act with purpose is theft, mentioned above, which requires that the criminal act be done with the purpose of permanently depriving another person of his or her rightful property. In some cases, however, full culpability for a crime only requires proof of the less strict mental state of knowing. For example, to be convicted of the federal crime of importing an illegal drug into the US, one need only know one is doing so. Similarly, to be convicted of selling illegal drug paraphernalia, one need only know the paraphernalia can be used to administer illegal drugs. One need not in addition be selling the items with that explicit purpose. A defendant is deemed reckless when he is aware of a risk that his act may have a particular result, yet performs it anyway. A homicide committed recklessly is treated differently from one committed purposely or knowingly. For example, in many states unintentionally killing another person while recklessly engaged in an action that is likely to cause death or great bodily harm constitutes involuntary manslaughter, an offense that carries signiﬁcantly lesser penalties than ﬁrst-degree murder.10 An example might be offering drunk teens bungee cord jumps off a high bridge when the cord has been improperly tied or is worn. A negligent state of mind means that the defendant should have been aware of a risk that his act might cause a particular harmful result, even though he was not aware. With regard to recklessness and negligence, it is usually also required that the risk of harm be substantial and unjustiﬁable.11 The distinction between recklessness and negligence is a difference between what the defendant knows regarding possible consequences of his action (in other words, his level of foresight). In this case the bungee cord jump operator may have consistently failed to inspect his equipment, but not actually known the cord was very likely to break. A reckless defendant knew there was a chance of substantial harm; a negligent defendant did not know but should have. In some cases this failure to foresee harmful consequences can rise to the level of ‘‘willful blindness’’ and a ‘‘wanton disregard for human life’’ as is required for corporate manslaughter. What a person should have known is determined by the rational person standard: that is, would an ordinary person of normal cognitive faculties have known there was a risk of

7

See generally, MPC §2.02, and MPC §2.06–2.08. MPC §2.02(2)(a) Purposely. A person acts purposely with respect to a material element of an offense when: (i) if the element involves the nature of his conduct or a result thereof, it is his conscious object to engage in conduct of that nature or to cause such a result and (ii) if the element involves the attendant circumstances, he is aware of the existence of such circumstances or he believes or hopes that they exist. 9 MPC §2.02(2)(b) Knowingly. A person acts knowingly with respect to a material element of an offense when: (i) if the element involves the nature of his conduct or the attendant circumstances, he is aware that his conduct is of that nature or that such circumstances exist and (ii) if the element involves a result of his conduct, he is aware that it is practically certain that his conduct will cause such a result. 10 For example, see Illinois statute 720 ILCS 5/9-3(a), which states that ‘‘A person who unintentionally kills an individual without lawful justiﬁcation commits involuntary manslaughter if his acts whether lawful or unlawful which cause the death are such as are likely to cause death or great bodily harm to some individual, and he performs them recklessly, except in cases in which the cause of the death consists of the driving of a motor vehicle or operating a snowmobile, all-terrain vehicle, or watercraft, in which case the person commits reckless homicide.’’ 11 The MPC recognizes one further state of mind, ‘willful blindness’, where a defendant is aware that there is a high probability that he is committing a crime, but intentionally avoids ascertaining the facts. 8

164

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

harm? This standard usually stipulates average capacity, but it may be stricter in certain cases. For example, to prove medical malpractice, a physician must fail to exhibit the standard of care accepted for a person of his profession and expertise. With regard to the difference between a reckless and negligent state of mind, awareness will function as a crucial bridge concept between neuroscience and the law, but the concept of attention is also relevant. These two concepts can function together to capture the legal notion of an agent who should have been aware of something. We are only fully aware of things, people and events after we have directed our attention toward them. Numerous experiments in psychology and cognitive neuroscience attest to this (Mack & Rock, 1998; Rensink, O’Regan, et al., 2003). A typical experiment of this sort allows the subject to examine a photo of scene for several seconds. Then the subject is shown the same photo with something in it altered. No matter how obvious the alteration is, if the subject did not explicitly attend to that detail and consciously register it, a large percentage of them do not notice the change (Rensink, O’Regan, et al., 2003). If we accept this, one way to analyze what it means to say that someone should have been aware of something means that the person’s attention should have been directed toward it. This means that the law is saying that an executive process, attention, should operate in certain ways in certain situations. A possible objection here is that the relevant brain process is consciousness rather than executive processing, contrary to what we suggested above. Indeed, the awareness being spoken about is conscious awareness and ‘‘attention’’ of course refers to conscious attention. However, ‘‘attention’’ means more than just that an item is at the ‘center’ of consciousness. It is possible to stare directly at something while not being aware of it. A more speciﬁc and accurate concept of awareness is that one is aware of an item if it has been brought into contact with the brain’s executive centers, and which may now perform manipulations on it. Similarly, the deﬁnition in the Model Penal Code of a purposeful action entails that the agent consciously desires a certain outcome. The importance of consciousness in this deﬁnition is not, we suggest, due to the importance of consciousness itself, but rather because those representations brought into consciousness are those representations exposed to the agent’s executive processes. It is not the consciousness that is doing the crucial mental work, it is the executive processes. A desire that has been brought into consciousness has been exposed to the sort of action inhibiting processes based in ventromedial prefrontal cortex. Desires that are exposed to executive processes and not eliminated or prevented from causing behavior will then become that agent’s considered desires. The agent has had the opportunity to eliminate or check them, but has failed to do so. If she then commits a criminal act, information regarding these considered desires is relevant to her guilt. Note what is gained by focusing on the executive processes instead of consciousness: in a particular subject we can perform a detailed analysis of where the process of consideration or decision-making might have broken down, and whether this breakdown was due to the subject’s limited capacity, or due to the failure to apply his fully functioning capacities. In the former case, lesser responsibility might be appropriate due to an excuse (see below); in the latter, responsibility and punishment might be fully warranted, although it could hinge on the law’s desire to encourage people to use these capacities. In sum, it seems that where the law demands one commit an act purposely, knowingly, or recklessly, an offender must have engaged his executive functions with regard to that act, although he would have paid lesser attention to the consequences of the act in the case of an act committed recklessly. However, in the case of negligence, it seems executive functions are not engaged with regard to the criminal act – but in the eyes of the law, they should have been. 4.3. Excuses and justiﬁcations If the prosecution fails to convince a judge or jury that the defendant either meets the voluntary act or the mental state requirement the defendant cannot be found guilty of a crime.12 However, even if both elements of the crime have been proven, the defendant can still attempt to make a plausible argument as to why he shouldn’t suffer punishment. A defendant can attempt to justify his act by claiming that it advanced some social interest, or prevented a greater wrong. Or, he can attempt to excuse his act by arguing that he merits exculpation and should not be held blameworthy. Typically, justiﬁcation defenses include ‘choice of evils’ or duress arguments. The insanity defense and a defense of intoxication are typically considered defenses of excuse.13 If we are right that the legal self consists the agent’s set of executive processes, then we should be able to explain many excuses and justiﬁcations in terms of abnormality or malfunction in executive processing. While we cannot address every legal excuse and justiﬁcation here, we will discuss two excuses that have traditionally resulted in lesser culpability: the insanity defense and juvenile status. We will also discuss sleepwalking cases and the case of psychopaths, who some have argued should have lesser culpability. 4.4. Sleepwalking Although, as indicated above, acts performed while asleep are not considered voluntary by the Model Penal Code, sleepwalking has proven to be a particularly difﬁcult case for the courts. Movements produced while sleepwalking may be related

12 Failure of proof defenses include things such as the defense of mistake or accident, discussed above. A mistake defense is merely a way of explaining why the prosecution has failed to prove the mental intent requirement. 13 Legal doctrine can be somewhat confusing on this point, however, because many of the defenses discussed below, including insanity and intoxication, can also be considered to show the defendant lacked the requisite mental state, which would mean the defendant would not need to present a defense (as the prosecution failed to make a case against him).

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

165

to some desire or other of the person, even though those acts are not produced in the normal way. For this reason some courts have characterized cases where a defendant appears to be credibly sleepwalking as cases where the sleepwalker’s act is voluntary, but where the requisite mens rea is lacking, and others have categorized sleepwalking offenders as legally insane. Consider the well-known Parks case (R v. Parks, 1992). Parks was a 23-year-old Toronto man with a wife and infant daughter. He was suffering from severe insomnia caused by joblessness and gambling debts. Early one morning in 1987 he got out of bed, got into his car and drove 23 km to his in-laws’ home, where he stabbed his mother-in-law to death and assaulted his father-in-law, who survived. He then drove to a police station and said, ‘‘I think I have killed some people. . .my hands,’’ seeming to only realize at that moment that he had severely cut his own hands.14 There was evidence that Parks had a good relationship with his in-laws and he had no apparent motive for the crime. He also had a history of sleepwalking. His team of defense experts concluded that he was in a state of automatism at the time the crime was committed, and thus had not committed a voluntary act. Dr. R. Billings, a psychiatrist from the University of Toronto, testiﬁed as follows15: ‘‘Probably the most striking feature of what we know of what goes on in the mind during sleep is that it is very independent of waking mentation in terms of its objectives and so forth. There is a lack of control of directing our minds in sleep compared to wakefulness. In the waking state, of course, we often voluntarily plan things, what we call volition – that is, we decide to do this as opposed to that – and there is no evidence that this occurs during the sleepwalking episode.’’ In response to the question of whether the defendant, if he was sleepwalking at the time of the crime, would have had the capacity to intend, Dr. Billings answered ‘No’. Parks was acquitted of all charges. In a 1988 case of sleepwalking in the UK, however, the court took a different view of the excuse of sleepwalking. In this case the defendant attacked a friend who was sleeping on his couch, striking her with a bottle and attempting to strangle her before he awoke and called an ambulance. The court determined the defendant, Burgess, was suffering from a sleep disorder based on testimony that sleepwalking is ‘‘. . .an abnormality of brain function, so it would be regarded as a pathological condition (R v. Burgess, 1991).’’16 This expert also opined that persons suffering from the disorder should be detained in a hospital ‘‘. . .because it is a treatable condition.’’ In the end, the court held that Burgess was not guilty by reason of insanity. In normal human beings there is a mechanism functioning to keep the brain activity that occurs in deep sleep from initiating action. In REM Behavior Disorder, however, a lesion in the brain stem causes the loss of muscle atonia usually present during REM, and one may act out the dreams one is having in REM sleep. Patients with this disorder complain of vivid, repetitive, and sometimes violent dreams. While still asleep these people often yell, punch, and kick, sometimes causing injury. While most agree that RBD is a clear disorder of the bypass of the motor system that usually occurs during REM sleep, some argue because the violent nature of the dream-enacting behavior RBD is also a psychological disorder. Sleepwalking, however, also occurs during stage 2 or non-REM sleep (Guilleminault, 1995). The sleeper suddenly starts to perform motor and/or verbal activity but remains asleep, as demonstrated by EEG. The individual is not aware of the activity, and unless awakened fully by others during the episode, will have total amnesia of the event (Sharp & D’Cruz, 2002). REM sleep is marked by a pronounced absence of executive activity. Imaging studies show that while the brain’s emotional centers are active, including orbitfrontal and orbitomedial cortices, the executive centers housed primarily in the dorsolateral cortex are inactive (Maquet, 2000) in both REM and non-REM sleep. We can verify this via introspection simply by recalling how uncritical we are in dreams. We are especially uncritical of our ‘‘perceptions’’: in dreams, people turn into other people right before us and we happily accept that and dream on. Normally such strange perceptions would arouse executive activity as the agent struggles to interpret them. In the Parks case, when the testifying psychiatrist spoke of the lack of volition during sleepwalking, and the lack of ability to ‘‘voluntarily plan things’’ in the sleepwalking state, he was speaking about executive processes, we would submit. Associating a person’s legal agency with proper executive function would thus work in this sort of case to limit the sleepwalker’s culpability. He was in a state in which executive activity would not be expected to happen. We cannot say, ‘‘He should have realized what he was about to. . .,’’ etc., because sleeping people can’t do this. Thus it makes sense to excuse sleepwalkers from their crimes, because the mental state requirement could not be fulﬁlled: the sleepwalker cannot be ‘‘purposeful’’, ‘‘knowing’’ or ‘‘reckless’’, all of which require some executive activity, and he can’t be accused of negligence because he couldn’t have been expected to engage in executive activity. Our analysis reinforces the law’s general assumption that a person who commits an act while sleepwalking should not be held responsible for her acts. However, there is some question whether the MPC is correct to classify such acts as ‘‘involuntary.’’ These acts seem intentional in a way that acts committed during an epileptic seizure or acts ‘‘not caused by volition of the actor’’ are not: there is, using Hart’s deﬁnition, a ‘‘minimal connection between mind and body.’’ Sleepwalking cases also seem substantively different from cases of legal insanity. As we will discuss below, in many cases a person suffering from mental illness has executive processes that are engaged, but signiﬁcantly deﬁcient in their operations. It thus seems that sleepwalking should be considered a failure of proof case, where a defendant is deemed not guilty due to lack of the requisite mens rea.17

14 A summary of facts in this case can be found in Lawrence, Martin ‘Can Sleepwalking be a Murder Defense’, at www.mtsinai.org/pulmonary/sleep/sleepmurder.htm. 15 Testimony quoted from E Law – Murdoch University Electronic Journal of Law, Vol. 3, No. 1 (May 1996). 16 A summary of facts in this case can be found in Lawrence, Martin ‘Can Sleepwalking be a Murder Defense’, at www.mtsinai.org/pulmonary/sleep/sleepmurder.htm. 17 See Morse 1984 for an argument that cases of diminished mental capacity should be considered failure of proof cases.

166

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

There is an interesting similarity between cases of sleepwalking and hypnosis, another state in which acts produced are considered involuntary by the MPC. Subjects who are hypnotized seem to retain some connection between desires or goal states represented within the actor’s cognitive system and acts performed, and thus such acts seem voluntary. However, some have argued that in cases of hypnosis a person’s executive functions appear to be disengaged (Farvolden & Woody, 2004; Woody & Bowers, 1994). It has also been suggested that some persons who are highly suggestible in hypnosis can be characterized by deﬁcits in executive functioning (Terhune, Cardea, et al., 2010). A study by Gruzelier and Warren (1993) found decreased executive functioning during hypnosis consistent with what has been termed the dissociatedcontrol model, which argues that executive processes are disengaged in hypnosis, allowing the hypnotist to originate some degree of motor and information-processing activity in the person hypnotized. Other studies have emphasized the similarities between cases of hypnosis and some cases of frontal lobe damage, where patients also suffer from disengagement of the executive system (Aikens & Ray, 2001). One way to interpret these ﬁndings is that the hypnotist is acting as a stand-in for the disengaged executive functions of the hypnotized actor. If this is true, in the case of legitimate hypnosis, as in sleepwalking, it would seem that the person who commits the act lacks the requisite mens rea to be held responsible because his or her executive functions are suppressed or disengaged. 4.5. Insanity In some jurisdictions, if a defendant was insane at the time the crime was committed a determination of legal insanity may wholly exculpate him due to lack of the requisite mens rea. One example is a burglary case where the defendant was under the delusion that he owned the apartment and its contents (See People v. Wetmore 22 Cal.3d 318 (1978)). In the majority of jurisdictions, however, a not guilty by reason of insanity (NGRI) plea is not considered a failure to prove mens rea but instead a defense – an excuse for the act – which qualiﬁes the defendant for special status. According to this view defendants with a mental illness are not considered as morally responsible as other defendants, and thus they are generally sentenced to a hospital or a prison for the criminally insane instead of prison, where they are generally held until considered ‘cured’ of their illness. Courts have been largely unsuccessful in determining legal insanity using medical terminology. For example, an attempt by courts to use the Durham test, a test meant to focus on the medical evidence of mental disorder, failed. The test required that the criminal act be the ‘‘product of a mental disease or defect’’ (Durham v. United States, 1954). Most jurisdictions in the US use one of two standards to determine whether a defendant is insane: either the 1843 M’Naghten test, or the test promulgated in the MPC. The M’Naghten test focuses on the requirement that a defendant possess a ‘‘defect of reason’’ resulting in an inability to discern right from wrong, requiring that ‘‘at the time of the committing of the act, the party accused was laboring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing, or if he did know it that he did not know what he was doing was wrong’’ (Rex v. M’Naughten, 1843). Thus this test for insanity focuses upon cognitive impairment. Compared to the M’Naughten test, the MPC lowers the insanity standard from an absolute knowledge of right from wrong to a substantial incapacity to appreciate the difference between right and wrong. The MPC also broadens the insanity test to include a volitional or ‘‘irresistible impulse’’ component: to be determined insane a defendant must either be unable to appreciate the difference between right and wrong or must be unable to conform his conduct to the requirements of the law. Sixteen states currently have an insanity statute that allows a defendant to qualify based upon volitional defect alone, without requiring any cognitive impairment (Donohue, Arya, et al., 2008). The American Psychiatric Association has argued that volitional tests may be unnecessary because defendants with signiﬁcant volitional impairment usually suffer from signiﬁcant cognitive impairment as well (Donohue et al., 2008). However, those in favor of preserving a volitional test claim that debilitating but treatable psychiatric disorders, such as bipolar disorder, result in loss of volitional control, while cognitive knowledge and appreciation of wrongfulness/criminality remain intact (Donohue et al., 2008). The use of the insanity defense is rare. On average the insanity defense is raised in less than 1% of all felony cases, and is only successful in about 15–25% of the cases in which it is raised (Borum & Solomon, 1999). Those acquitted tend to be charged with more serious offenses, older, more physically attractive, more educated and have a less extensive criminal history (Rice and Harris, 1990). Between 60% and 77% of those deemed legally insane are diagnosed with a psychosis (Gurley & Marcus, 2008). As we will discuss below, evidence of psychopathy is much less likely to result in an acquittal (Gurley & Marcus, 2008). A person who is experiencing hallucinations is only mentally incompetent if she believes her hallucinations. As we noted above, the brain’s executive processes have the power to reject perceptions and memories as false. Executive processes, when functioning normally, seem able to correct for just about any defect in perception. We all experience strange thoughts on occasion, the feeling that someone is watching us; the odd notion that we had a causal inﬂuence on something where no reasonable physical explanation is available (e.g., I ﬂip a light switch and a car horn honks outside); the idea that others can read our thoughts. But we are able to reject these thoughts and not let them establish themselves as beliefs because we have the cognitive processes required to assess their plausibility. A person with delusions, however, is not mentally competent because she believes her delusions. This indicates an executive problem; in order for strange and/or implausible thoughts to become beliefs, there needs to be a second problem at the cognitive level (in addition to the ﬁrst factor, responsible for generating the delusional belief to begin with). A two-factor hypothesis of delusions ﬁnds its ﬁrst full explicit form in 2000 (Langdon & Coltheart, 2000), followed a year later by a more general version of the hypothesis (Davies, Coltheart, et al., 2001). More recently, the nature of the executive problems leading to the failure to reject anomalous beliefs has been

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

167

described in greater detail (McKay, Langdon et al., 2007). When the writers of the M’Naghton standard spoke of a ‘‘defect of reason’’ they were singling out executive processes, we suggest, speciﬁcally a failure to respond in a rational way to what is contained in one’s consciousness. Because of this, we feel that cases where a criminal defendant suffers from a hallucination or delusion, where the hallucination or delusion is directly related to the crime committed, are cases where the defendant should be legally excused from his act. Similarly, where a defendant, due to a serious mental illness, suffers from severe dysfunction in executive processes, or a lack of executive processing, the legal self is properly considered ‘‘insane’’ and thus excused from culpability. Persons with progressive schizophrenia, for example, often exhibit severe deﬁcits in tasks testing attentional set shifting, online use of working memory, and planning and strategy capacity (Hutton, Puri, et al., 1998). In such cases it seems clear that a lack of executive function due to mental illness ought to excuse a defendant due to cognitive impairment – and that evidence of volitional capacity is unnecessary. In keeping with the views of the American Bar and American Psychiatric Associations, our analysis also leads us to disagree with current shift allowing ‘‘guilty but mentally ill’’ (GBMI) verdicts. In 2000, at least 20 states had instituted GBMI provisions. GBMI statutes include a ﬁnding that the defendant was mentally ill at the time the crime was committed, but that he or she fails to meet the requirements for legal insanity. Such a category seems unnecessary. Either the defendant’s executive processes were severely compromised at the time the crime was committed, or they were not. If they were, the person lacks mental capacity and thus is not culpable. We might agree with the GBMI verdict if it offered lesser culpability for mentally ill defendants who suffered from substantial, but not severe, problems with executive functioning. But ﬁrst, using executive function as the gauge for legal insanity is something we are now suggesting; it is not a currently a common tactic for offering or assessing evidence of mental capacity. Second, the verdict is not used to apply lesser culpability to those with lesser levels of mental illness. A defendant who receives a GBMI verdict is still considered legally guilty of the crime in question, but since he is mentally ill, he is entitled to receive mental health treatment while in prison. If the defendant’s symptoms remit while in prison he is required to serve out the remainder of his sentence. (A defendant who was acquitted by reason of insanity, on the other hand, must be released from his treatment facility if it is determined he is no longer dangerous to himself or others.) Third and ﬁnally, many studies have determined that the inclusion of a GBMI provision has signiﬁcant impact on juror verdicts in that jurors are less likely to hold a defendant not guilty by reason of insanity if they have a GBMI option (Callahan, McGreevy, et al., 1992; Caton, Golding, et al., 1987). This means that the GBMI verdict may be allowing jurors to ignore the moral or legal consequences of relevant evidence of mental illness. Our contention is that in the future, evidence of mental illness will become more standardized, and easier to understand, if presented in terms of executive function. 4.6. Juvenile status Juveniles have traditionally been thought to have lesser criminal responsibility because they lack mental capacities important to decision-making and thus are more impulsive and easily manipulated (e.g., via peer pressure). Offenders below a certain age – traditionally in common law the age of 7 – were thought to be incapable of forming the mental intent required for any crime. In addition, the separate juvenile system of justice in the US, which applies less severe penalties than the adult system, can be seen as an outgrowth of the doctrine of diminished mental capacity. As mentioned above, most US states use some combination of age of offender and severity of offense, and possibly some other factors, to determine whether a young defendant should be transferred up to adult court, or should remain in the juvenile system (Butts & Mitchell, 2000). Age is often thought to be a proxy for level of mental capacity. Severity of offense, on the other hand, seems to identify defendants most deserving of retribution. Some legislatures require any juvenile, regardless of age, to be transferred to adult court for certain offenses. For example, in West Virginia, any child who commits a violent criminal act must be transferred. In addition, several states provide judicial discretion to send any juvenile regardless of age to criminal court based on the speciﬁc offense.18 As stated earlier, in the 2005 case of Roper v. Simmons the Supreme Court categorically excluded offenders under eighteen from the death penalty in part due to diminished capacity. In his majority opinion Justice Kennedy cited much scientiﬁc psychology from amici briefs ﬁled in the case. For example, noting that juveniles are more subject to peer pressure and that they have less control over their environments, Kennedy then cited an article by Steinburg and Scott (2003). Here is a section of the article Kennedy cited: What is most interesting is that studies of brain development during adolescence, and of differences in patterns of brain activation between adolescents and adults, indicate that the most important developments during adolescence occur in regions that are implicated in processes of long-term planning (Spear, 2000). . . .[P]atterns of development in the prefrontal cortex, which is active during the performance of complicated tasks involving long-term planning and judgment and decision-making, suggest that these higher order cognitive capacities may be immature well into late adolescence (Giedd, 1999). (Steinburg & Scott, 2003)

18 For a complete review of juvenile transfer provisions in the US, see the governmental website for the US Department of Justice’s Ofﬁce of Juvenile Delinquency and Prevention, located at http://www.ojjdp.ncjrs.gov/pubs/reform/contents.html.

168

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

In this quotation Steinberg and Scott are referring to the slow development of gray matter in the prefrontal cortex. A large part of the development of the frontal lobes occurs between ages 5 and 10 (Case, 1992), and they do not fully mature until the teenage years (Smith, Kates, & Vreizen, 1992). There are maturational improvements in frontal connectivity, both within the frontal cortex and with more distal regions (speciﬁcally the parietal cortex and basal ganglia, respectively) occurring into early adulthood (Baird & Fugelsang, 2004). This maturation closely coincides with age-related behavioral improvements in working memory, as well as action planning and inhibition. The improved fronto-parietal connectivity that comes with age has also been found to be highly correlated with improvements in working memory (Olesen, Westerberg, & Klingberg, 2004). Abigail Baird, whose research focuses upon juvenile decision-making, argues that because of these deﬁcits, ‘‘it may be physically impossible for adolescents to engage in counterfactual reasoning, and as a result of this are often unable to effectively foresee the possible consequences of their actions’’ (Baird & Fugelsang, 2004). Interestingly, before the cortical areas that embody executive processes tasked with managing and verifying memories have matured, children make unreliable witnesses. Johnson and her colleagues note that children exhibit some of the same patterns of memory problems that frontal patients show (Lindsay, Johnson, & Kwon, 1991). There are now several experiments showing that false memories can be induced in children by asking them leading questions (Ackil & Zaragoza, 1998). Ceci et al. were able to produce false memories in preschool children. Children were presented with a deck of cards, each of which described an event (Ceci, Huffman, & Smith, 1994). Some of the events had actually happened to the children, while others had not. When they were repeatedly asked whether the false events had happened to them, 58% of the children eventually agreed that they had, and many of them embellished the event with confabulated details. As all children below a certain age appear to suffer from global deﬁcits in executive function, we support a separate juvenile court system handling all criminal defendants below a certain age, regardless of the level of crime committed. This age should be determined by further research into the severity of deﬁcit in executive function during different points of cognitive development. A more exact alternative would be to undertake a case-by-case analysis of the cognitive maturity (in terms of executive function) of juvenile defendants. However, we understand this approach would be costly and thus is unlikely to be adopted. We also understand that retributive sentiment may support sending juveniles who commit very serious crime to adult prisons. However, the principle of retribution goes hand in hand with the principle of proportionality; and punishment ought to be proportional to the type of offender, not just to the level of crime committed. 4.7. Psychopaths As stated above, juveniles are both eligible to be excused from culpability or to have lesser culpability in accordance with the diminished mental capacity doctrine. In addition, there is conceptual overlap between the doctrines of legal insanity and diminished capacity because both can be used to indicate that a defendant was incapable of forming the requisite mental states for criminal guilt. The diminished capacity doctrine thus allows a criminal defendant to either negate a mental element of the crime charged, thereby exonerating the defendant of that charge, or to reduce the degree of the crime for which the defendant may be convicted, even if the defendant’s conduct satisﬁes all the elements of a higher offense (Morse, 1984). A fairly new and more controversial use of the diminished capacity doctrine is to claim that it qualiﬁes psychopaths for lesser culpability. Recently defense attorneys have introduced fMRI and PET scan evidence of psychopathy along with other evidence of abnormal decision-making and behavior, in an attempt to argue that psychopaths suffer similar cognitive deﬁciencies to defendants who are more commonly considered to have diminished capacity, such as juveniles and the mentally retarded.19 Similarly, philosophers have argued that, due to a lack of certain mental capacities, it may be inappropriate to hold psychopaths fully responsible for their criminal acts (Fine & Kennett, 2004). For example, at the sentencing phase of Brian Dugan’s capital trial for the 1983 kidnapping, rape and murder of a 10-yearold girl, neuroscientist Kent Kiehl, working for the defense, introduced fMRI evidence that Dugan had serious brain abnormalities similar to those who suffered from psychopathy. Dugan scored 37 out of a possible 40 on the Hare Psychopathy checklist, which placed him in the 99.5th percentile. The defense argued that Dugan had abnormal brain functioning typiﬁed by antisocial behavior, impulsivity, and an inability to feel sympathy or remorse. The defense used this evidence to make a general argument that Dugan was a psychopath, and that psychopaths suffer from diminished mental capacity, a mitigating circumstance in Illinois’ capital sentencing statute. However, the jury eventually returned a verdict of the death penalty, despite Kiehl’s testimony.20 Neuroscientist R. J. R Blair, among others, has argued that the functional contributions of the amygdala and ventromedial prefrontal cortex (vmPFC) are compromised in subjects with psychopathy (Blair, 2008a, 2008b). It appears that the amygdala may process emotional information, including information about the emotional states of those around us, sending this and other information to the vmPFC as information relevant to decision-making and action palnning. In an fMRI study of fearful expression processing, Marsh et al. (2008) reported reduced functional connectivity between the amygdala and the vmPFC in children with psychopathic tendencies relative to the comparison children. Moreover, Birbaumer, Veit, et al. (2005) reported reduced vmPFC activity as well as reduced amygdala activity in individuals with psychopathy during aversive conditioning (Birbaumer et al., 2005). Finally, psychopathic individuals show hippocampal and anterior cingulate impairment.

19 20

http://www.nytimes.com/2007/03/11/magazine/11Neurolaw.t.html. http://www.nature.com/news/2010/100317/full/464340a.html.

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

169

Both Adina Roskies and Adrian Raine and colleagues have argued that psychopaths appear to show normal moral reasoning ability, but appear to fail to apply the outcome of such reasoning processes due the lack of emotional input described above (Raine & Yang, 2006; Roskies, 2006). Roskies speciﬁcally argues that psychopaths have moral beliefs, but not the motivation to act upon them (Roskies, 2006). When asked to provide an answer to the famous ‘‘trolley’’ thought experiment – where subjects are asked to decide whether to intentionally kill one person to save ﬁve – patients with ventromedial damage are more likely to judge that intentionally killing the one man (by pushing him onto trolley tracks) is the right thing to do. Thus, it is thought that persons with ventromedial damage may be more likely to engage in antisocial or immoral behavior, precisely because they do not feel badly about such action. Given the basic separation between our representations and our executive processes, we have argued that the latter play a special role in legal agency and the legal self, while the former are much less important. Executive processes can correct for a wide variety of ﬂaws in our set of representations. For instance, if you have a bad memory, you can correct for this by taking great care in making any claims based on your memory. And, as discussed above, even if you are hallucinating, if you know this you can correct for it by ignoring implausible visual input. This ability to correct for ﬂawed perceptions or memories positions the set of executive processes as the ﬁnal line of defense to prevent ﬂawed representations from causing ﬂawed, i.e., illegal behavior. Psychopaths have ﬂawed perceptions in that they do not perceive other people as intrinsically valuable, at least in part because they fail to have the proper emotional input that would allow them to do so (Herpertz & Sass, 2000). They have had abundant time and feedback to correct these perceptions, yet they do not do so. Up to this point we have avoided answering the question of exactly what sort of executive damage, and what level of damage, is exculpatory. It seems inadvisable to make the weakness or total absence of a single executive process exculpatory, because many of the other executive processes are still capable of correcting for this weakness. Another reason for this is that we will all have stronger and weaker executive abilities, but we do not regard the presence of these weaknesses as exculpatory. If it is true, then, that psychopaths lack only a single executive process, this would not seem to be enough to relieve them of responsibility for their actions. 5. Conclusion One way to see that these executive processes constitute a sort of self is to observe that we use that special word ‘‘I’’ in describing them. We say ‘‘I decided to do it,’’ ‘‘I planned the event,’’ ‘‘I realized it that it was not a bear,’’ and even ‘‘I stopped myself from doing it.’’ We take ownership of these processes, and we take responsibility for what is accomplished with them. They are the natural locus of our sense of responsibility and guilt. Understanding exactly how the brain plans actions and forms intentions can help us begin to discern whether certain types of injury to the areas that accomplish these might be exculpatory. But there are some obvious limits here. All rational persons are, perhaps by deﬁnition, capable of acting intentionally, so we do not foresee a point where neuroscience could establish that a defendant was incapable of forming any intentions whatsoever. There are nevertheless particular conditions in which neuroscience might provide evidence for or against the presence of the relevant intentions. Examples might include actions while under hypnosis or during a seizure. Neuroscience might also be able to show that a defendant was incapable of constructing a complicated plan, by showing that he had damage to polar prefrontal cortex, for example. Despite the glacial pace at which change comes to the legal system, progress in the brain sciences is cause for optimism. A recent study found that its participants who acted as jurors were more likely to ﬁnd a defendant not guilty by reason of insanity if they heard testimony about brain injury or saw images of brain damage (Gurley & Marcus, 2008). The neuroimages used included MRI scans obtained from Anderson, Damasio, Tranel, and Damasio (2000) that showed extensive damage to the prefrontal cortex. The participants saw these scans and read testimony indicating that the prefrontal cortex ‘‘has been associated with impulse control and it is likely that because of the reduced volume in this area the defendant would have difﬁculty controlling his actions.’’ The study’s authors speculate that the reason for this is that neuroimages and brain injury testimony provide more tangible evidence than is typically presented in psychological or psychiatric testimony. Unlike medical disorders, psychiatric disorders cannot usually be seen through blood test results or X-rays; jurors have to rely on experts to make the diagnosis. The neuroimages of readily apparent brain damage give the jurors tangible proof of the disorder. The brain injury testimony may work in a different way: it provides the jurors with a clear cause and an effect. The defendant was normal prior to the brain injury and did not begin showing symptoms of a psychiatric disorder until after the brain injury, thus the brain injury must have caused the psychiatric disorder. (Gurley & Marcus, 2008)(94) Our set of executive processes is the most important, and ﬁnal way that we have for making sure our actions are not criminal. As Gurley and Marcus note, ‘‘Combined, the brain injury testimony may have led jurors to believe that the defendant was unable to control his actions, due to no fault of his own, and the neuroimages provided conclusive proof that the defendant had suffered brain damage.’’(Gurley & Marcus, 2008)(94) Explicitly (as opposed to implicitly) focusing on the brain’s executive processes can help open up the human planning and decision-making process to the view of the legal system. Understanding decisions and action in folk psychological terms – i.e., using terms such as ‘‘intention,’’ ‘‘knowledge,’’ and ‘‘purpose’’ – is certainly helpful, but these terms can also invite error by making us think we understand more about the psychology of defendants than we do. Connecting these principles to solid neuroscience may be able to correct for any biases or hasty generalizations used in the law.

170

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

Issues surrounding the notion of self are more commonly thought to involve questions about our identities, whereas the legal self primarily involves our responsibility for our actions. We have, however, exactly as much identity as we are able to claim responsibility for. Identity is not purely a social notion: there are personal consequences to identity choices as well. We have, to a carefully deliberated extent, the right to be who we are. But with identity comes responsibility. The legal self is that set of properties that simultaneously create this identity and this responsibility. It is that part of our psychology that is imperative for identity, and that the law attends to. It is vital to close the gap between our legal conceptions of self and the brain, so that our legal system can begin making ﬁner-grained distinctions in its judgments of guilt, and of competency. Acknowledgment We gratefully acknowledge the research assistance of Krista Wiley. References Ackil, J. K., & Zaragoza, M. S. (1998). Memorial consequences of forced confabulation: Age differences in susceptibility to false memories. Developmental Psychology, 34, 1358–1372. Anderson, S. W., Damasio, H., Tranel, D., & Damasio, A. R. (2000). Long-term sequelae of prefrontal cortex damage acquired in early childhood. Developmental Neuropsychology, 3(18), 281–296. Aikens, D., & Ray, W. J. (2001). Frontal lobe contributions to hypnotic susceptibility: A neuropsychological screening of executive functioning. The International Journal of Clinical and Experimental Hypnosis, 49(4). Atkins v. Virginia, 536 US 304 (2002). Baars, B. J. (Ed.). (1993). How does a stream of consciousness that is relatively simple, serial, and limited in capacity emerge from a brain that is largely unconscious, complex, and massively parallel? Ciba symposium on experimental and theoretical foundations of consciousness. London, UK: Wiley Interscience. Baars, B. J., Ramsøy, Thomas Z., & Laureys, Steven (2003). Brain, conscious experience and the observing self. Trends in Neurosciences, 26(12), 671–675. Baars, B. J., & Steven, L. (2005). Global workspace theory of consciousness: Toward a cognitive neuroscience of human experience. Progress in Brain Research, Elsevier, 150, 45–53. Baird, A. A., & Fugelsang, J. A. (2004). The emergence of consequential thought: Evidence from neuroscience. Philosophical Transactions of the Royal Society, 359(1451), 1797–1804. Barbas, H., Ghashghaei, H. T., Rempel-Clower, N. L., & Xiao, D. (2002). Anatomic basis of functional specialization in prefrontal cortices in primates. New York: Elsevier. Barbas, H., & Zikopoulos, B. (2007). The prefrontal cortex and ﬂexible behavior. The Neuroscientist, 13(5), 532–544. Birbaumer, N., Veit, R., et al (2005). Deﬁcient fear conditioning in psychopathy: A functional magnetic resonance imaging study. Archives of General Psychiatry, 62(7), 799–805. Blair, R. (2008a). The cognitive neuroscience of psychopathy and implications for judgments of responsibility. Neuroethics, 1(3), 149–157. Blair, R. J. R. (2008b). The amygdala and ventromedial prefrontal cortex: Functional contributions and dysfunction in psychopathy. Philosophical Transactions of the Royal Society B: Biological Sciences, 363(1503), 2557–2565. Blumenthal, S. L. (2007). The default legal person. UCLA Law Review, 54(5). Borum, R., & Solomon, M. F. (1999). Empirical research on the insanity defense and attempted reforms: Evidence toward informed policy. Law and Human Behavior, 23(1), 117–135. Burgess, P. W. (2000). Real-world multitasking from a cognitive neuroscience perspective. Cambridge, MA: MIT Press. Butts, J. A., & Mitchell, O. (2000). Brick by brick: Dismantling the border between juvenile and adult justice. Boundary changes in criminal justice organizations. NIJ, US Department of Justice, Ofﬁce of Justice Programs, 2, 167–213. Callahan, L. A., McGreevy, M. A., et al (1992). Measuring the effects of the guilty but mentally ill (GBMI) verdict: Georgia’s 1982 GBMI reform. Law and Human Behavior, 16(4), 447–462. Carter, C. S., Braver, T. S., et al (1998). Anterior cingulate cortex, error detection, and the online monitoring of performance. Science, 280(5364), 747–749. Case, R. (1992). The role of the frontal lobes in the regulation of cognitive development. Brain and Cognition, 20, 51–73. Caton, F. R., Golding, S. L., et al (1987). Implicit theories of criminal responsibility: Decision making and the insanity defense. Law and Human Behavior, 11(3), 207–232. Ceci, S. J., Huffman, M. L., & Smith, E. (1994). Repeatedly thinking about a non-event: Source misattributions among preschoolers. Consciousness and Cognition, 3, 388–407. Crick, F., & Koch, C. (2003). A framework for consciousness. Nature Neuroscience, 6(2), 119–126. Damasio, A. (1995). Descartes’ error: Emotion, reason, and the human brain. Harper Perennial. Davies, M., Coltheart, M., et al (2001). Monothematic delusions: Towards a two-factor account. Philosophy, Psychiatry, and Psychology, 8(2/3), 133–158. Dehaene, S., Naccache, L., Cohen, L., Bihan, D. L., Mangin, J. F., Poline, J. B., et al (2001). Cerebral mechanisms of word-masking and unconscious repetition priming. Nature Neuroscience, 4, 752–758. Dennett, D. (1991). Real patterns. Journal of Philosophy, LXXXVIII, 27–51. D’Esposito, M., Postle, B. R., et al (1999). Maintenance versus manipulation of information held in working memory: An event-related fMRI study. Brain and Cognition, 41(1), 66–86. Devinsky, O., Morrell, Martha. J., & Vogt, Brent. A. (1995). Contributions of anterior cingulate cortex to behaviour. Brain: A Journal of Neurology, 118(1), 279–306. Donohue, A., Arya, V., et al (2008). Legal insanity: An assessment of the inability to refrain. Psychiatry MMC March. Duncan, J., & Owen, A. (2000). Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends in Neurosciences, 23(10), 475–483. Durham v. United States, 214 F 2d 862 (1954). Farvolden, P., & Woody, E. Z. (2004). Hypnosis, memory, and frontal executive functioning. International Journal of Clinical and Experimental Hypnosis, 52(1), 3–26. Fine, C., & Kennett, J. (2004). Mental impairment, moral understanding and criminal responsibility: Psychopathy and the purposes of punishment. International Journal of Law and Psychiatry, 27(5), 425–443. Fodor, J. (1987). Psychosemantics: The problem of meaning in the philosophy of mind. Cambridge, MA: MIT Press. Garavan, H., Ross, T. J., Murphy, K., Roche, R. A. P., & Stein, E. A. (2002). Dissociable executive functions in the dynamic control of behavior. Neuroimage, 17, 1820–1829. Gilbert, S. J., & Burgess, P. W. (2008). Executive function. Current Biology, 18(3), 110–114. Gruzelier, J., & Warren, K. (1993). Neuropsychological evidence of reductions on left frontal tests with hypnosis. Psychological Medicine, 23(1), 93–101. Guilleminault, C. e. a. (1995). Forensic sleep medicine: Nocturnal wandering and violence. Sleep, 18, 740–748. Gurley, J. R., & Marcus, D. K. (2008). The effects of neuroimaging and brain injury on insanity defenses. Behavioral Sciences and the Law, 26(1), 85–97.

W. Hirstein, K. Sifferd / Consciousness and Cognition 20 (2011) 156–171

171

Hart, H. L. A. (1968). Punishment and responsibility: Essays in the philosophy of law. Oxford: Clarendon Pres. Herpertz, S. C., & Sass, H. (2000). Emotional deﬁciency and psychopathy. Behavioral Sciences and the Law, 18(5), 567–580. Humphreys, G., Forde, E. M. E., & Francis, Dawn. (Eds.). (2000). The organization of sequential actions. Control of cognitive processes. Cambridge, MA: MIT Press. Hutton, S. B., Puri, B. K., et al (1998). Executive function in ﬁrst-episode schizophrenia. Psychological Medicine, 28(02), 463–473. Johnson, M. K., Hayes, S. M., D’Esposito, M. D., & Raye, C. L. (Eds.). (2000). Confabulation. Handbook of neuropsychology. New York: Elsevier. Koch, C. (2004). The quest for consciousness: A neurobiological approach. Englewood, Colorado: Roberts and Company Publishers. Langdon, R., & Coltheart, M. (2000). The cognitive neuropsychology of delusions. Mind and Language, 15, 184–218. Lindsay, D. S., Johnson, M. K., & Kwon, P. (1991). Developmental changes in memory source monitoring. Journal of Experimental Child Psychology, 52, 297–318. Mack, A., & Rock, I. (1998). Inattentional blindness. Cambridge, MA: MIT Press. Maquet (2000). Functional neuroimaging of normal human sleep by positron emission tomography. Journal of Sleep Research, 9(3), 207–231. Marsh, A. A., Finger, E. C., Mitchell, D. G. V., Reid, M. E., Sims, C., Kosson, D. S., et al. (2008). Reduced amygdala response to fearful expressions in children and adolescents with callous-unemotional traits and disruptive behavior disorders. American Journal of Psychiatry, 165, 712–720. McKay, R., Langdon, R., et al (2007). Models of misbelief: Integrating motivational and deﬁcit theories of delusions. Consciousness and Cognition, 16(4), 932–941. Miller, E., & Wallis, J. (2009). Executive function and higher-order cognition: Deﬁnition and neural substrates. Encyclopedia of Neuroscience. Morse, S. J. (1984). Undiminished confusion in diminished capacity. The Journal of Criminal Law and Criminology, 75(1). Morse, S. J. (2003). Inevitable mens rea. Harvard Journal of Law and Public Policy, 27. Morse, S. J. (2006a). Moral and legal responsibility and the new neuroscience. Neuroethics: Deﬁning the issues in theory, practice, and policy, vol. 33. J. Illes. Oxford, New York: Oxford University Press. 50 p. Morse, S. J. (2006b). Brain overclaim syndrome and criminal responsibility: A diagnostic note. 3 Ohio St. J. Crim. L. Morse, S. J. (2007). Criminal responsibility and the disappearing person. Cordozo Law Review, 28, 2545. Morse, S. J. (2008). Determinism and the death of folk psychology: Two challenges to responsibility from neuroscience. Minnesota Journal of Law, Science and Technology, 9(1), 1–36. Moscovitch & Winocur (2002). The frontal cortex and working with memory. In D. T. Stuss & R. R. Knight (Eds.), Principles of Frontal Lobe Function (pp. 188–209). New York: Oxford University Press. MPC (1985). Model Penal Code and Commentaries (Ofﬁcial Draft and Revised Comments). Olesen, P. J., Westerberg, H., & Klingberg, T. (2004). Increased prefrontal and parietal activity after training of working memory. Nature Neuroscience, 7, 75–79. Owen, A. M., Evans, A. C., & Petrides, M. (1996). Evidence for a two-stage model of spatial working memory processing within the lateral prefrontal cortex. A positron emission tomography study. Cerebral Cortex, 6, 31–38. Panetti v. Quarterman, 551 US 930 (2007). People v. Dugan, 485 NE 2d 315 (1985). Petrides, M. (2005). Lateral prefrontal cortex: Architectonic and functional organization. Philosophical Transactions of the Royal Society Series B, 360, 781–795. R v. Burgess, 2 QB 92A (1991). R v. Parks, 2 SCR 871 (1992). Raine, A., & Yang, Y. (2006). Neural foundations to moral reasoning and antisocial behavior. SCAN(1), 203–213. Ramnani, N., & Owen, A. M. (2004). Anterior prefrontal cortex: Insights into function from anatomy and neuroimaging. Nature Reviews Neuroscience, 5(3), 184–194. Rensink, R. A., O’Regan, K., et al (2003). To see or not to see: The need for visual attention to perceive changes in scene. In B. J. Baars, W. P. Banks, & J. B. Newman (Eds.), Essential sources in the scientiﬁc study of consciousness. Cambridge, MA: MIT Press. Rex v. M’Naughten, 8 Eng. Rep. 718 (1843). Rice, M., & Harris, G. (1990). The predictors of insanity acquittal. International Journal of Law and Psychiatry, 13, 217–224. Roper v. Simmons, 543 US 551 (2005). Roskies, A. (2006). Patients with ventromedial frontal damage have moral beliefs. Philosophical Psychology, 19(5), 617–627. Sellars, W. (1956). Empiricism and the philosophy of mind. In H. Feigl & M. Scriven (Eds.). The foundations of science and the concepts of psychoanalysis, Minnesota studies in the philosophy of science (Vol. 1, pp. 127–196). Minneapolis, MN: University of Minnesota Press. Sharp, S., & D’Cruz, O. F. (2002). Somnambulism. Emedicine(January), 1–10. Smith, M. L., Kates, M. H., & Vreizen, E. R. (1992). The development of frontal lobe functions. In F. Boller & J. Grafman (Eds.), Handbook of Neuropsychology. New York: Elsevier. Steinburg, L., & Scott, E. S. (2003). Less guilty by reason of adolescence: Developmental immaturity, diminished responsibility, and the juvenile death penalty. American Psychologist, 58(12), 1009–1018. Sternberg, R. (2010). Human intelligence. Amsterdam: Elsevier. Terhune, D. B., Cardea, E., et al (2010). Dissociative tendencies and individual differences in high hypnotic suggestibility. Cognitive Neuropsychiatry. Vincent, N. A. (2008). Neuroimaging and responsibility assessments. Neuroethics. Woody, E. Z., & Bowers, K. S. (1994). A frontal assault on dissociated control. In S. J. Lynn & J. W. Rhue (Eds.), Dissociation: Clinical and theoretical perspectives. New York: Guilford.

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

Recommend Documents