ARTICLE IN PRESS Consciousness and Cognition xxx (2008) xxx–xxx

Contents lists available at ScienceDirect

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

Commentary

Is my body the sum of online and offline body-representations? q Manos Tsakiris a,*, Aikaterini Fotopoulou b a b

Department of Psychology, Royal Holloway, University of London, London, UK Institute of Psychiatry, King’s College London, UK

1. Introduction The recent paper by Glenn Carruthers (2008) aptly addresses the question of embodiment by employing a distinction between two types of body-representations. Carruthers argues that embodiment, primarily used for self-recognition, is underpinned by online and offline body-representations that differ in terms of their origins and content. Online representations are generated directly by current sensory input and represent ‘‘what the body is currently like”, allowing thus an online control and monitoring of body-configurations. Offline representations are constructed by online representations and represent ‘‘what the body is usually like”. Offline representations, therefore, code for more permanent and diachronic aspects of the body, such as its structure and possibly its usual appearance and motor repertoire. Carruthers emphasizes that only offline body-representations underlie the sense of embodiment. On a strong version of the argument, individuals lacking offline representations, should feel—and possible be—disembodied. Carruthers addresses several body-related neurological and psychiatric conditions, such as anosognosia for hemiplegia and body integrity identity disorder, to illustrate his point. The crux of the argument is that the sense of embodiment is underpinned by the integration of various types of offline representations with visual and emotional information, while online representations are neither necessary nor sufficient for embodiment. The distinction between offline and online body-representations seems a useful one, especially because it is orthogonal to the frequently used and perhaps misused distinction between body-image and body-schema, and it can be valuable in generating testable empirical hypotheses. However, Carruthers’ proposal does not go as far as to fully explain what is actually meant by offline representations and how these interact with online representations to give us a sense of embodiment. 2. What is embodiment for? Carruthers argues that embodiment is used primarily for self-recognition An example to illustrate the role of embodiment for self-recognition is the recent research on explicit recognition of one’s own movements that investigates the necessary and sufficient conditions under which a body-part will be experienced as infallibly ‘‘me” (for reviews see Jeannerod, 2003; Tsakiris, 2008). Carruthers refers to the study by van den Bos and Jeannerod that investigated the contributions of ‘‘a sense of body” and ‘‘a sense of action” for self-recognition. According to Jeannerod (2003); see also van den Bos & Jeannerod, 2002), one main conclusion of the study is that ‘action cues’ are used when distinctive movements are made (e.g. in the different movement condition), and that afferent signals (i.e. vision and proprioception) are used when action cues are ambiguous (e.g. in the same movement condition). Importantly, for Carruthers, this experiment shows that ‘‘self-recognition depends in part of one’s sense of what is one’s body”, but he fails to see that what is manipulated in this experiment is not the identity or form of one’s body-parts, but instead, it is the postural configuration of a body-part; thus it is not a question of ‘‘what” the body is, as much as ‘‘where” the body is. Interestingly, the detection and recognition of postural configuration of our hands depends on both online and offline representations. Online multisensory representations are necessary for updating the representation of hand location in space, while offline representations would be useful in detecting that a left hand presented at an incongruent posture with respect q Commentary on Carruthers, G. (2008). Types of body representation and the sense of embodiment. doi:10.1016/j.concog.2008.02.001. This article is part of a special issue of this journal on (xxx). * Corresponding author. E-mail addresses: [email protected], [email protected] (M. Tsakiris).

1053-8100/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2008.06.012

Please cite this article in press as: Tsakiris, M., & Fotopoulou, A. Is my body the sum of online and offline body-representations? Consciousness and Cognition (2008), doi:10.1016/j.concog.2008.06.012

ARTICLE IN PRESS 2

M. Tsakiris, A. Fotopoulou / Consciousness and Cognition xxx (2008) xxx–xxx

to one’s body or at the location of one’s right hand could not possibly be one’s hand because that would violate the anatomical configuration of the body as represented offline. But is the congruence between online (e.g. vision and proprioception) and offline body-representations sufficient for infallible self-recognition? The paradigm of the Rubber Hand Illusion (RHI, see Botvinick & Cohen, 1998; Tsakiris & Haggard, 2005) suggests that if only afferent information were present or used for selfrecognition, then the viewed hand would always be attributed to the self, provided that there were no conflicts between sensory modalities (e.g. vision and touch in RHI, vision and proprioception in self-recognition tasks). Thus, it may be hypothesized that for infallible self-recognition, intersensory congruence or a congruency between online sensory representations and offline representations may not be sufficient. In fact, it has been shown that when participants are asked to recognize their movements in the absence of motor information while a congruence between visual and proprioceptive afferent feedback is present, their self-recognition performance drops below chance levels (Tsakiris, Haggard, Franck, Mainy, & Sirigu, 2005). Thus, self-recognition cannot be solely depending on afferent signals and body-representations. Self-recognition, in the sense of correctly recognizing a visual object (e.g. body-part) or event (e.g. movement) as ‘‘me’’ or ‘‘mine’’ seems to depend largely on efference and agency, and this dimension is only marginally considered in Carruthers’s model of embodiment. A more interesting proposal for the role of embodiment comes from Carruthers’s suggestion that embodiment underpins the feeling of self as being distinct from other objects and persons. This is a rather different question from self-recognition in the sense that a distinction between self and others, or between my body and other people’s bodies seems to be a prerequisite for correct recognition of myself and/or other. Diverse cognitive tasks such as attention, multisensory integration, agency, perspective-taking, mentalizing and understanding other people’s minds seem to rely on the efficient distinction between events that may or may not relate to oneself. In other words, these cognitive abilities seem to require a prior representation of a self sufficiently distinct from other subjects or objects to enter into relations with them. The mere fact of embodiment could be the first level at which this distinction between oneself and other subjects or objects takes place. In terms of information processing, we would expect that a model of one’s body could be used to test whether perceived sensations and events may or may not relate to one’s body. This model describes diachronic anatomical, postural and visual features of the body, acting as a reference against which current sensory inputs can be compared. Several studies (Costantini & Haggard, 2007; Lenggenhager, Tadi, Metzinger, & Blanke, 2007; Longo, Schüür, Kammers, Tsakiris, & Haggard, 2008; Tsakiris & Haggard, 2005; Tsakiris, Hesse, Boy, Haggard, & Fink, 2007) suggest that this model may have an important role in the sense of body-ownership because of its modulatory influence on multisensory perception. The effect of multisensory stimulation on body-ownership is not simply a passive stimulus-driven process, but rather seems to depend on the modulatory influence of an abstract body-model that contains visual, anatomical and postural representations of the body (de Vignemont, Tsakiris, & Haggard, 2006; Graziano & Botvinik, 2001). Current sensory stimuli are processed and finally tested-for-fit against this cognitive (i.e. non-sensorial) body-model that maintains a coherent sense of one’s own body (Tsakiris, Costantini, & Haggard, 2008). This basic mechanism involves filtering sensory events for their importance or attribution to the self. This process would be central to the functional role of offline body-representations, and it can be highly relevant to Carruthers’ point about the role of embodiment in the distinction between the body and the external world. 3. On the origins and content of offline representations Carruthers argues that online representations are constructed directly from current sensory input and that some of these online representations are eventually taken offline. His proposal raises two points of concern. First, the very term ‘‘directly” needs to be validated. Second, the possibility of hardwired offline representations and their modulatory role must be considered. We suppose that what is meant by ‘‘directly” is the primary, raw and immediate nature of online sensory representations. However, a review of the relevant literature suggests that these online representations are neither raw, nor immediate. The online representation of the multisensory body in the brain cannot be merely reduced to a registration of peripheral inputs. Multisensory perception involves the interpretation of sensory inputs in the context of a rich internal, possibly offline, multisensory model of the body, and the use of these inputs for an online representation of the body in space (Graziano & Botvinik, 2001). Moreover, the interactions between unimodal and multimodal brain areas are both horizontal and vertical (for a review see Driver & Spence, 2000). Thus, the relationship between online and offline representations may not be serial as Carruthers suggests. If, as Carruthers suggests, offline representations are serially constructed by online representations, then we can only think of offline representations as the reservoir of accumulated representations of bodily experiences to which we have been repeatedly exposed during ontogenetic development. According to this account, it is not possible to form an offline bodyrepresentation unless we had a prior experience that allowed the formation of an online representation of the body-part or sensation that is eventually represented offline. If that was the case, then we would not be able to account for the experience of phantom limbs in patients who congenitally lack limbs or for early demonstrations of hand-mouth coordination in utero and early infancy (Gallagher, Butterworth, Lew, & Cole, 1998) or, arguably, for the infants’ sensitivity to left-right reversal of their own legs shown on a screen (Rochat, 1998). In addition, Carruthers’ account implicitly suggests that it is not possible for an offline representation to modulate or alter an online representation. If that was the case, we would not be able to account for the phenomenon of phantom limb at all. In

Please cite this article in press as: Tsakiris, M., & Fotopoulou, A. Is my body the sum of online and offline body-representations? Consciousness and Cognition (2008), doi:10.1016/j.concog.2008.06.012

ARTICLE IN PRESS M. Tsakiris, A. Fotopoulou / Consciousness and Cognition xxx (2008) xxx–xxx

3

phantom limb, an amputated arm or leg continues to be experienced as present and occupies its former location in space. This phenomenon is thought to involve continued input to areas of cortex formerly responsible for representing the position of the missing limb (for a review see Ramachandran & Hirstein, 1998). The fact that this input is translated into a detailed limb representation indicates that peripheral information (e.g. current sensory input) is interpreted with reference to a centrally-maintained model of the body’s structure (e.g. offline) that continues to exist even in the absence of the limb that is eventually represented online. Overall, unimodal or multimodal sensory online representations are neither raw nor immediate. Instead, they seem to be processed and constructed with reference to an abstract, possibly offline, body-model that guarantees the spatial coherence of the body. At least some of the offline body-representations seem to be hardwired and innate, challenging the idea that offline representations derive only from online representations. This suggestion has also important implications for the content of offline representations. How many types of offline representations are needed to represent the usual state of the body? No answer is given to this question. The idea of innate representations suggests that at least some anatomical features of the body (e.g. number of limbs, left vs. right limbs) are a priori represented, but the actual features of body-parts (e.g. morphology, length of a limb) are constructed and updated throughout development. How these different types of offline representations are integrated? Other than mentioning the need for integration, Carruthers does not say how this integration takes place. 4. The case of Anosognosia for Hemiplegia Carruthers uses the case of AHP as a confirmatory example for the existence of distinct online and offline bodily representations. He claims patients with AHP have intact online representations of their bodily states but an impaired (not-updated) offline representation. This leads to the patients’ general false belief they are not paralysed, despite their occasional and circumstantial awareness of their paralysis (Marcel, Tegner, & Nimmo-Smith, 2004). The idea that AHP stems out of a deficit in body state updating is not new (e.g. Berti et al., 2005; Damasio, 1994; Frith, Blakemore, & Wolpert, 2000; Ramachandran, 1995). The critical questions are what is the nature and extent of this deficit, and how does it relate to normal models of motor awareness. According to Carruthers, patients with AHP have an impaired offline representation of their body because they are mistaken about its generic state. We feel AHP is not a straightforward choice for this argument. As Carruthers admits, these patients do not experience disembodiment. However, he still claims their sense of the state of the body is mistaken. This argument seems to lack specificity and appears self-contradictory. If offline representations are a necessary precondition for embodiment and AHP patients have impaired offline body-representations then they should experience feelings of disembodiment. However, unlike patients with personal neglect, asomatognosia, or, somatoparaphrenia, AHP patients do not experience alterations of body-ownership or self-identity. Pathologies such as asomatognosia and somatoparaphrenia seem to be more straightforward examples of body state misrepresentation and they have been reported to doubly dissociate with AHP. Furthermore, as Carruthers describes, during temporary remission of AHP via vestibular stimulation, patients are not just able to accurately describe their paralysis but they can also determine for how long they have been paralysed. This suggests a generic updating (albeit temporary) of their body representation, rather than merely intact online representation. In fact, Carruthers admits that such generic updating must have had taken place all along but remained unavailable to consciousness. It thus appears that the critical difficulty in AHP is the non-accessibility to consciousness of different kinds of body representations, rather than dissociation between two different types of body representation, one of which is impaired. Finally, while the more general notion of a non-updated body representation is clearly a useful one for understanding AHP, Carruthers’ proposal does not offer an explanation for the lack of updating other than stating that ‘‘For some reason new information from the body has not updated it since the patient was paralysed”. Assuming that some other type of body representation (online in this case) is intact does not seem to offer any theoretical advantage. In conclusion, the binary distinction at question does not seem to account for the complexity of AHP presentations and there is little theoretical advantage in adopting it. It would perhaps be more appropriate to first apply and test this distinction in cases that do involve altered states of embodiment per se, such as autotopagnosia, somatoparaphrenia and asomatognosia. 5. Conclusion Carruthers argues that the integration of offline body-representations with vision and emotion should be sufficient for embodiment. In our commentary we tried to highlight three key points that demand further clarification at the theoretical and empirical level. First, embodiment is not primarily for self-recognition. The suggestion that embodiment allows for a clear distinction between self and the external world is more interesting. Second, we argued that the serial model of online/offline relationships does not take into account the more complex patterns of interaction between current sensory inflow, online and offline body-representations. Finally, we highlighted that AHP may not be the most suitable candidate for testing the dichotomy between online and offline representations, at least not without considering motor awareness and action representations in addition to body representations. Carruthers clearly stated his intention to sidestep agency in this paper, but it is precisely for that reason that his answer seems incomplete. Converging evidence suggests that agency

Please cite this article in press as: Tsakiris, M., & Fotopoulou, A. Is my body the sum of online and offline body-representations? Consciousness and Cognition (2008), doi:10.1016/j.concog.2008.06.012

ARTICLE IN PRESS 4

M. Tsakiris, A. Fotopoulou / Consciousness and Cognition xxx (2008) xxx–xxx

produces important changes in the temporal and spatial experience of one’s body (for a review see Tsakiris, Schütz-Bosbach, & Gallagher, 2007). To the extent that my body is for the most part an acting body, rather than a merely sensing body, offline body-representations alone cannot provide a full account of embodiment. Acknowledgments Dr. Tsakiris and Dr. Fotopoulou were supported by the ‘‘European Platform for Life Sciences, Mind Sciences, and the Humanities” grant by the Volkswagen Stiftung for the ‘‘Body-Project: interdisciplinary investigations on bodily experiences”. References Berti, A., Bottini, G., Gandola, M., Pia, L., Smania, N., Stracciari, A., et al (2005). Shared cortical anatomy for motor awareness and motor control. Science, 309(5733), 488–491. Botvinick, M., & Cohen, J. (1998). Rubber hands ‘feel’ touch that eyes see. Nature, 391, 756. Carruthers, G. (2008). Types of body representation and the sense of embodiment. Consciousness and Cognition. doi:10.1016/j.concog.2008.02.001. Costantini, M., & Haggard, P. (2007). The rubber hand illusion: Sensitivity and reference frame for body ownership. Consciousness and Cognition, 16, 229–240. Damasio, A. R. (1994). Descarte’’ error: Emotion, reason and the human brain. Quill. de Vignemont, F., Tsakiris, M., & Haggard, P. (2006). Body mereology. In G. Knoblich, I. Thornton, M. Grosjean, & M. Shiffrar (Eds.), Perception of the human body. New York, NY: Oxford University Press. Driver, J., & Spence, C. (2000). Multisensory perception: Beyond modularity and convergence. Current Biology, 10, R731–R735. 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. Gallagher, S., Butterworth, G. E., Lew, A., & Cole, J. (1998). Hand-mouth coordination, congenital absence of limb, and evidence for innate body schemas. Brain & Cognition, 38, 53–65. Graziano, M., & Botvinik, M. (2001). How the brain represents the body: Insights from neurophysiology and psychology. In W. Prinz & B. Hommel (Eds.), Common mechanisms in perception and action. Attention and performance XIX. Oxford New York: Oxford University Press. Jeannerod, M. (2003). The mechanism of self-recognition in humans. Behavioural Brain Research, 142, 1–15. Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317, 1096–1099. Longo, M., Schüür, F., Kammers, M. P. M., Tsakiris, M., & Haggard, P. (2008). What is embodiment? A psychometric approach. Cognition, 107, 978–998. Marcel, A. J., Tegner, R., & Nimmo-Smith, I. (2004). Anosognosia for plegia: Specificity, extension, partiality and disunity of bodily unawareness. Cortex, 40, 19–40. Ramachandran, V. S. (1995). Anosognosia in parietal lobe syndrome. Consciousness and Cognition, 4, 22–51. Ramachandran, V. S., & Hirstein, W. (1998). The perception of phantom limbs. The D. O. Hebb lecture. Brain, 121(Pt. 9), 1603–1630. Rochat, P. (1998). Self perception and action in infancy. Experimental Brain Research, 123, 102–109. Tsakiris, M. (2008). The self-other distinction: Insights from self-recognition experiments. In F. Morganti, A. Carassa, & G. Riva (Eds.). Enacting intersubjectivity a cognitive and social perspective to the study of interactions. Amsterdam: IOP Press. Tsakiris, M., Costantini, M., & Haggard, P. (2008). The role of the right temporo-parietal junction in maintaining a coherent sense of one’s body, Neuropsychologia, in press, doi:10.1016/j.neuropsychologia.2008.06.004. Tsakiris, M., & Haggard, P. (2005). The rubber hand illusion revisited: Visuotactile integration and self-attribution. Journal of Experimental Psychology: Human Perception & Performance, 31, 80–91. Tsakiris, M., Haggard, P., Franck, N., Mainy, N., & Sirigu, A. (2005). A specific role for efferent information in self-recognition. Cognition, 96, 215–231. Tsakiris, M., Hesse, M., Boy, C., Haggard, P., & Fink, G. R. (2007). Neural correlates of body-ownership a sensory network for bodily self-consciousness. Cerebral Cortex, 17, 2235–2244. Tsakiris, M., Schütz-Bosbach, S., & Gallagher, S. (2007). On agency and body-ownership: Phenomenological and neurocognitive reflections. Consciousness & Cognition, 16, 645–660. van den Bos, E., & Jeannerod, M. (2002). Sense of body and sense of action both contribute to self-recognition. Cognition, 85, 177–187.

Please cite this article in press as: Tsakiris, M., & Fotopoulou, A. Is my body the sum of online and offline body-representations? Consciousness and Cognition (2008), doi:10.1016/j.concog.2008.06.012