CORTEX D. Povinelli and T. Preuss - Evolution of theory of mind Theory of mind: evolutionary cognitive specialization history of a Daniel J. Povinelli and Todd M. Preuss Traditional analyses of the evolution ity among species. However, particular kind of intelligence intentions and beliefs. Data indicate as those of others, might Behavioral studies in children in the developmental patterns in the same manner. mind, forever Trends Neurosci. altering (1995) I that be the their research is related result TN.7 Vol. 18, No. 9, 1995 that to reflect of evolutionary reveal Humans might commonality might mental have specialized states in the both similarities capacities. a cognitive in a such as desires, states, as well prefrontal and striking Humans but it is too early to be certain have evolved and continu- on one’s own mental changes that lead to theory-of-mind of social behavior, humans to understanding that the ability cortex. differences and great apes if they interpret specialization in theory of view of the social universe. 18, 418-424 N CULTURES around the world, humans exhibit an automatic and pervasive folk psychology that interprets the behavior of the self and others in terms of inferences about unobservable mental states such as desires, intentions and beliefs’. Premack and Woodruff2 coined the term ‘theory of mind’ to refer to this system of inferences, noting that an ability to represent and infer mental states from the self and others ‘may properly be viewed as a theory because such states are not directly observable, and the system can be used to make predictions about the behaviors of others’. This folk psychology encapsulates much of what it means to be human: our commitments to the self and others as mental agents replete with emotions, desires, beliefs, personalities and personal histories. Are humans alone in this thoroughly mentalistic interpretation of the self and others? Or did theory of mind evolve before the emergence of humans? Surely this is one of the most interesting and significant questions that we can ask about the evolution of human nature, yet it is one that has escaped empirical attention until quite recently. Daniel J. Povinelli This review explores the hypothesis that theory of is at the Laboratory mind is a recent evolutionary innovation. A strong of Comparative version of this hypothesis is that the ability to Behavioral Biology, attribute mental states, together with the specific New Iberia neural organization that makes such attributions Research Center, possible, is a unique specialization of the human New Iberia, LA species. This would imply that theory of mind 70560, USA, and emerged during the past several million years, after the Dept of the divergence of humans and apes. An alternative Psychology, Tulane hypothesis is that at least some aspects of theory of University, New mind were already in place before the emergence of Orleans, LA, USA, the human species, and might, therefore, be present to and Todd M. Preuss some extent in other primates, and most likely the is at the Dept of great apes. Nonetheless, if this view is correct, it would Psychology, mean that although for hundreds of millions of years Vanderbilt animals have been operating on the basis of implicit University, ‘intentions’ and ‘knowledge’ stored in their neural Nashville, 7N circuitry, it was not until relatively recently that evo37240, USA. lution produced animals with brains capable of rep418 have emphasized suggests and chimpanzees pathways share many ancient them recent of intelligence resenting those mental concepts explicitly. In other words, although many organisms routinely form neural states instantiating desire, intention and belief, it is possible that only a few species (and possibly only humans) have the ability to reflect upon these states. The evolution of a psychological system that is capable of representing such mental states would not require a precise understanding of the neural architecture that encodes them. After all, even an approximation of the workings of such mental states - a theory of mind might allow organisms to predict the behavior of their conspecifics with considerable accuracy. The purpose of this article is twofold. First, we establish a neural and psychological framework from which to consider the evolution of theory of mind. Second, we review the evidence that concerns the development of theory of mind in human children, and what is known about its development (or lack thereof) in great apes and other non-human primates. This review makes a preliminary assessment of alternative hypotheses of the timing of the evolution of various aspects of theory of mind and, in particular, the hypothesis that theory of mind, at least in an elaborated form, is a psychological specialization of the human species. Human brain specialization Figure 1 depicts the evolutionary relationships among humans and our closest living relatives, the great apes. Evolutionary reconstructions of the features that were likely present in the common ancestor of great apes and humans suggest that each of its living representatives has departed in important ways from the ancestor, which was probably an arboreal ape of about 30 kg (Ref. 3). Orang-utans, for example, have become extreme arboreal specialists with important modifications that occur in the wrist, shoulder and hip to allow for more extreme suspensory postures. The African apes (chimpanzees and gorillas) have developed specializations for a distinctive form of terrestrial quadrupedalism called ‘knuckle-walking’, in 0 1995, Elsevier Science Ltd D. Povinelli which the weight of the upper body is borne on the middle phalanges of the hands. Humans have departed in an even more bizarre manner from the ancestor of the great-ape-human group, with a radical remodeling of the pelvis and lower limb to support habitual bipedalism4f5. Just as in musculoskeletal structure, the brain too has undergone dramatic evolution following the divergence of the human lineage from that of the African apes, -5-8 million years ago. The brain increased in volume by approximately threefold, and the fossil evidence indicates clearly that most of the increase occurred during the past 2 million years, long after the evolution of bipedality’. Not only did the volume of the brain increase during human evolution but there was also a disproportionate enlargement of the association cortex, and especially the prefrontal cortex, which occupies -24% of the cerebral mantle in humans, compared with -14% in great apes6. The expansion of the human prefrontal region might be related to the evolution of specialized neural systems that subserve distinctive human cognitive capacities’, including theory of mind. In a number of theoretical accounts, prefrontal cortex has been accorded a supervisory or executive role in the cognitive system, serving as the substrate by which mental representations of the state of the world exert their influence on other brain systems that control attention, memory and action’-“. Moreover, what is regarded as the peculiarly adaptive or flexible character of human behavior, and the insightful, self-reflective character of human thinking, has been attributed to the action of frontallobe systems’. The ability to regulate behavior through the use of representations of the mental states of oneself and others might be a human specialization of the representational capacities present in the prefrontal cortex of other primates (compare Ref. 9). In fact, studies of autistic individuals suggest that human theory-of-mind capacities involve frontal cortex. Autistic individuals show little understanding of their own mental states, or those of others and, thus, perform very poorly on formal theory-of-mind tasks, relative to mental-age-matched controls”. Autistics’ lack of insight and introspection is reminiscent of deficits that result from frontal lesions, as is their stereotyped, perseverative behavior, prompting the suggestion that frontal-lobe dysfunction is an important element in autism’z-‘4. This view is supported by a recent demonstration of activation in orbital prefrontal cortex during performance of a theory-of-mind taskls. If theory of mind is an evolutionary specialization of human cognition, it should be expected that there are corresponding specializations at the level of neural systems. Unfortunately, apart from the fact that the human brain is unusually large, and the prefrontal region disproportionately so, very little is currently known about how human cerebral organization differs from that of our closest relatives. Thus, our current understanding of possible human specialization in theory of mind is based largely upon behavioral studies. Developmental psychologists have undertaken substantial research to characterize the development of theory of mind in children. Likewise, comparative psychologists are beginning to explore whether other non-human primates possess a similar ‘folk psychology’. and T. Preuss - Evolution of theory CORTEX of mind Fig. 1. The great apes and humans share a common ancestor approximately 12-15 million years ago. Although the exact relationships among gorillas, chimpanzees and humans are still uncertain, it is wide/y acknowledged that the other living great ape, the orangutan, diverged considerably earlier. Theory of mind in humans and other primates Knowledge and belief Large portions of the adult human theory of mind are established firmly in children between three and five years of age16-18.Perhaps the quintessential achievement of the four- to five-year-old is the clear understanding that beliefs are mental representations - internal states that are formed as people interact with the world. The standard test of thii achievement involves determining at what age children understand that the mind can represent incorrectly a true state of affairs in the world; in other words, when they understand the notion of false belief”. Understanding that beliefs can be false is an excellent diagnostic of theory of mind because, in this case, an organism must keep track of the divergence between the mind and the world, thus demonstrating clearly that it understands the distinction between the two. Research on children’s understanding of false belief suggests that between three and five years children solidify their understanding of the mind as a mechanism that creates mental representations (and misrepresentations) of the external world”. Four-year-olds also understand some of the ways in which belief and knowledge states arise. For example, they understand that perceptual contact with a situation or event is a sufficient (and in some cases necessary) condition for one to possess a privileged state of knowledge as compared with someone who has not had similar perceptual contact. In the case of visual perception, a four-year-old who observes someone peering into a box automatically attributes to that person knowledge of the box’s contents; likewise they attribute ignorance to someone who merely touches the box”. Strikingly, younger children appear unaware of this crucial aspect of acquisition of knowledge, failing to grasp that the person who did not look into the box has no way of knowing what was inside. Indeed, young three-year-olds apparently do not even know how they themselves come to know certain TINS Vol. 18, No. 9, 1995 419 CORTEX D. Povinelli and T. Preuss - Evolution of theory of mind Povinelli and his co11eagues25,26, has explored what (if anything) chimpanzees understand about the connection between seeing and knowing. For example, in the procedure shown in Fig. 2, if chimpanzees understand the seeingknowing relationship, they ought to choose the ‘advice’ offered by the trainer who hid the food (or the person who was carefully watching while it was being hidden), as opposed to someone else who could not see the hiding event. In general, these tasks have yet to produce definitive evidence that chimpanzees, or other non-human primates, such as macaques27,28, understand the perception-knowledge relationship. Finally, there are a number of observational reports of apparent deception in nonhuman primates (especially chimpanzees”), and it is tempting to interpret them as evidence that non-human primates understand that other agents have beliefs that can be manipulated. Very little experimental work has investigated non-human primates’ understanding of false belief 24.However, if these animals do not understand how beliefs arise (as the experimental data suggest), it seems unlikely that they could make the distinction between true and false beliefs. Emotions and desires The four-year-old’s discovery that mental states such as belief are internal representations of the external world clearly unites their emerging theory of mind with our Fig. 2. By the age of four years, young children understand how the act of seeing creates internal knowlown. However, children younger edge states but it is not clear if chimpanzees ever understand this relation. The experimental procedure that than four years might already underis used to investigate chimpanzees’ and children’s understanding of the perception-knowledge re/ationshipz6: stand other, nonrepresentational a chimpanzee watches as one experimenter (guesser; A) leaves the room and a second experimenter (knower; B) hides food under one of several cups. (C) Both experimenters later point to different cups. (D) Transfer promental states, and they might even cedure in which logical epistemoiogical relationships between seeing and knowing are held constant but the envipossessa nonrepresentational underronmental variables that define the role of the knower and guesser are altered. With sufficient experience, chimstanding of belieP0j31.Two-year-olds panzees can learn to form a discrimination between the two roles but, in their initial tria/s, they choose random/y. talk extensively about certain mental These findings raise the possibility that they solve the task by relying on learned behavioral strategies (for examstates, such as emotions and desires, ple, ‘pick the person who stayed in the room’), rather than on mentalistic inferences about the connection that are not representations of the between seeing and knowing. By contrast, when this same nonverbal task is administered to four-year-old chilexternal world but are nonetheless dren, most choose the correct person from the first trial forward; three-year-olds respond random/y23. thoroughly mentalistic concepts. For example, the subtle ways in facts (did you see what was in the box or did I tell which young children use mental verbs such as ‘want’, you?), even in cases where their knowledge was ‘hope’, and ‘wish’, suggest strongly that they know that they, along with others, are ‘repositories’ of unobservable acquired just seconds earlier21-23.This is not to say that three-year-olds lack the concept of knowledge alto- mental states, even if they have not yet grasped how the gether; they might possess an earlier understanding of mind can also serve as a mechanism to copy or represent knowledge, perhaps one in which knowledge is con- the state of affairs in the world3’. Furthermore, experimental research has revealed consistently and conflated with desire (T.D. Lyon, PhD Thesis, Stanford University, 1993). vincingly that three-year-olds understand that people Do non-human primates represent the beliefs of act on the basis of their desires, that they distinguish others? To date, most research on non-human pri- between someone thinking about something compared mates’ understanding of belief has focused on with someone having it, and when someone’s beliefs whether they understand the perceptual basis of are stipulated they can even make correct predictions acquisition of knowledge. Premackz4, as well as about how that person will behave”. In general, 420 TINS vol. 18, NO. 9, 1995 D. Povinelli and T. Preurs - Evolution of theory CORTEX of mind Fig. 3. Chimpanzees and orang-utans (but not other primates) show clear evidence of recognizing themselves in mirrors. Like most animals, chimpanzees react initially to mirrors as if confronted by another member of their species, and typically engage in a variety of social behaviors. However, in as little as S-30 min, many chimpanzees begin to engage in self-exploratory behaviors in which they use the mirror to gain access to previously unavailable information about their appearance. They use mirrors to make exaggerated facial displays (A and B), as we// as groom their body, face and teeth (C-F). Many of the chimpanzees that display the types of se/f-exploratory behaviors that are shown in C-F also will use a mirror to locate and inspect red marks p/aced surreptitiously on their ears and eyebrow ridge47s49,50. Organisms such as Old World monkeys, which do not display these spontaneous episodes of using the mirrors to explore their bodies, also fail to explore marks that were placed on their face by experimenters4648. younger children’s relative ease in understanding desires, as opposed to thoughts and beliefs, has led a number of researchers to propose the existence of a specific developmental pathway in which children come to grasp nonrepresentational aspects of the mind before understanding the representational aspects, such as those tapped in false belief and knowledgeattribution tasks31,33. Given that children have a better understanding of desires before beliefs, it is reasonable to ask if chimpanzees or other non-human primates display a better understanding of desires than they do of beliefs. Premack and Woodruff’s2 original investigation of theory of mind tested whether a chimpanzee could attribute desires or goals to actors who were struggling to solve staged problems. After observing videotapes of these problems, the chimpanzee correctly selected photographs that depicted solutions to the actors’ implicit desires or goals, as opposed to those that depicted associated events. In a different study, Povinelli and his colleagues trained chimpanzees to co-operate with a human partner on a task that required each to perform a different action (role), in order to achieve a common goa134.The crucial finding TINS vol. 18,No.9, 1995 421 CORTEX D. Povinelli and T. Preuss - Evolution of theory of mind Fig. 4. Humans and chimpanzees share portions of the developmental pathway that controls the expression of gaze-following. By 18 months, human infants are capable of responding to the shift in orientation of the gaze or head posture, or both, of another by following their line of sight, including tracking this line into the space behind them 43. One possible interpretation of this type of behavior is that it reflects a type of ‘joint-visual attention’ in which the infant and adult appreciate the attentional focus of each other. That is, not on/y do they have a mechanism that serves joint attention43, they might also appreciate that they share the same subjective attentional experience44. (A-C) Research with chimpanzees reveals that they a/so display this phenomenon [as, indeed, might other social mammals) 57. However, additional research with chimpanzees (see Fig, 5) indicates that they might not represent the subjective attentional state of the other oneJ6. This raises the possibility of both a developmental and evolutionary dissociation between joint-visual attention and the presence of a folk psychological interpretation of visual perception as the mental state of attention. There might even be earlier manifestations of human infants’ developing awareness of the subjective side of behavior. In the period leading up to 18-24 months, children begin to develop sophisticated behaviors that suggest at least an implicit awareness of the mental state of attention in themselves and others. For example, proto-declarative pointing (in which infants point to objects or events solely to Early-emerging knowledge of se/f and others draw the other’s attention to them) is consolidated by In humans, the emergence of conceptual knowlabout 12 months4’, and between 12 and 18 months edge of the self and others manifests itself at around infants develop the ability to track another person’s 18-24months in the transitional period between line of sight into space outside of their own immediinfancy and early childhood. Indeed, this period ate perceptual field 43. Some researchers have suggested introduces such dramatic achievements by the child that these ‘joint-attention’ behaviors reflect44, or at (not all limited to an understanding of mental states) least lay the perceptual foundations for45, an initial that it is tempting to speculate (as did Piaget) that a understanding of the subjective attentional states of fundamentally new conceptual system is emerging. both the self and others. In an ingenious series of Although even the very young infant has access to studies, Dare Baldwin and her colleagues have provided proprioceptive and kinesthetic information that evidence that, by at least 18 months, infants underenables it to distinguish self from the environment stand that when another person looks or points at an (self-perception), evidence for an objective underobject or event, that person becomes connected to it standing of the self does not begin to emerge clearly subjectively through the mental state of attention4’j. until -1Smonths. At this point, children recognize Do non-human primates share even these limited themselves in mirrors, engage in symbolic play, exhib- glimpses into the mind? Like humans, many chimit simple acts of altruism, participate in reciprocal co- panzees and orang-utans display clear evidence of recoperative actions with others, produce linguistic com- ognizing themselves in mirrors, whereas no other ments about the failure of self-generated plans, display non-human primates have done so, despite extensive mastery smiles upon completion of a task, and use research47,48(Fig. 3). The ability of these apes to recogmental state terminology that refers to desires32,36-41. nize themselves in mirrors has been replicated many Each of these behaviors suggests at least a limited times, and there is little doubt that they display the understanding of the self and other as possessing same phenomenon of self-recognition as do 18-24agency, desires or internal emotional states, or both. month-old children49,50. Gallup’s initial discovery of was that the chimpanzees were able to reverse roles immediately, assuming their partner’s role without explicit training. In contrast, rhesus monkeys failed the same role-reversal task3’. Collectively, these data suggest that chimpanzees might possess some understanding of goals or desires, although both of these studies have enough methodological limitations to warrant further research before firm conclusions are drawn. 422 TINS vol. 18, No. p. 1995 D. Povinelli and T. Preuss - Evolution of theory CORTEX of mind Fig. 5. Although young chimpanzees are very sensitive to the eyes of others, they do not appear to understand them as ‘portals’ through which the mental state of attention emanates. (A and B) Chimpanzees can be trained to use their natural begging gesture to request food from an experimenter. If an experimenter stands on the left (A and B), the chimpanzee will gesture through the hole on the left side of the partition; if she stands on the right, the subject will gesture there. However, this is no guarantee that the subjects realize that this person is subjective/y linked to them via the mental state of attention. (C) When two experimenters are used, one who can see the chimpanzees and one who cannot, the chimpanzees typically respond randomly j6 . They apparently fail to understand lhat on/y one of the experimenters is connected lo them through the mental state of attention, perhaps because they simply cannot represent such states. Alternatively, they might simply fail to understand the specific role that eyes play in deploying attention. Young children that are tested similarly reveal an understanding of the subjective aspects of visual attention by about two and a half years of agej6. self-recognition in chimpanzees led him to speculate that chimpanzees might possess a self-concept. Adopting a long-standing epistemological argument, he further hypothesized that perhaps their knowledge of self might enable them to make inferences about such mental states as desires and beliefs in other?. Thus, he has speculated that in organisms that are capable of self-recognition, there might be a causal connection between the onset of self-recognition and theory of mind, with the critical factor being selfawareness51~s2.Consistent with his predictions, some research has revealed correlations between the onset of self-recognition in infants and their capacity for altruism and the emergence of the self-conscious emotions53*54. By contrast, Gallup’s proposal would be in jeopardy if proto-declarative pointing and other joint-attention behaviors reflect a genuine understanding of the mental state of attention because, in humans, joint-attention behaviors typically emerge by 12 months or so, long before self-recognition in mirrors (18-24 months). At present, the paucity of research in this area (especially with non-human primates) makes a full evaluation of the proposition difficult”. Finally, there are some recent findings that concern joint attention in chimpanzees, and that highlight additional similarities and differences between humans and chimpanzees. On the one hand, chimpanzees do not naturally point to inform others about objects or events, nor is it clear that training them to point, or to respond to human pointing (see Fig. 2), produces an appreciation of pointing as a way of co-ordinating the mental states of self and others. On the other hand, chimpanzees do display gaze-following abilities that are as sophisticated as those of 18-month-old children56,57 (see Fig. 4). Although such gaze-following abilities suggest that chimpanzees might appreciate how the eyes connect someone’s internal states of attention to the world (as do children by 18-24 months), a recent extensive investigation of this question has indicated that they do nots6 (see Fig. 5). These somewhat contradictory findings raise a number of possibilities. On the one hand, it might be that limited aspects of theory of mind were present in the common ancestor of the great apes and humans, and this is reflected in chimpanzees’ limited joint-attention behaviors. On the other hand, these joint-attention behaviors might have evolved for other reasons, and might be controlled by psychological mechanisms that evolved before a human specialization in theory of mind (see Fig. 4), although in humans they now function in concert with each other. Reconstructing the evolution of theory of mind On the surface, much of the social behavior of the great apes resembles our own so closely that it is tempting to conclude that this behavioral similarity must reflect similarity in subjective experience”. But as developmental and comparative psychological research suggest, important differences in how humans, great apes and other animals interpret other organisms might lie behind these behavioral similarities”. It is too early to be certain, but current evidence does not yet exclude the possibility that, at some point during human evolution, elements of a new psychology were incorporated into existing neural systems - a psychology which, by its very nature, imbues ancient behavioral patterns that we share with apes with meanings they did not possess originally. Selected references 1 Povinelli, D.J. and Godfrey, L.R. (1993) in Evolutionary Ethics (Nitecki, M. and Nitecki, D., eds), pp. 277-324, SUNY Press 2 Premack, D. and Woodruff, G. (1978) Behav. Bruin Sci. 1,515-526 3 Harrison, T. (1991) in Or&e(s) de la bipMie chez les hominides (Coppens, Y. and Senut, B., eds), pp. 235-244, Editions du CNRS 4 Aiello, L. and Dean, C. (1990) Human Evolutionary Anatomy, Academic Press T~S vol. 18, No. 9,199s 423 CORTEX D. Povinelli Acknowledgements Supported by National Institutes of Health Grant No. RR-03583-05 to the USL-New Iberia Research Center, and National Science Foundation Young Investigator Award SBR-8458111 to DIP. Photographs by Donna T. Bierschwule. 5 McHenry, H.M. (1994) Proc. Nat1 Acad. Sci. USA 91, 6780-6786 6 Blinkov, S.M. and Glezer, 1.1. (1968) The Human Brain in Figures and Tables, Basic Books 7 Preuss, T.M. (1995) in The Cognitive Neurosciences (Gazzaniga, M.S., ed.), pp. 1227-1241, MIT Press 8 Stuss, D.T. and Benson, D.F. (1986) The Frontal Lobes, Raven Press 9 Goldman-Rakic, P.S. (1987) in Handbook of Physiology: The Nervous System, Vol. 5 (Mountcastle, V.B., Plum, F. and Geiger, S.R., eds), pp. 373-417, American Physiological Society 10 Shallice, T. (1988) From Neuropsychology To Men&I Structure, Cambridge University Press 11 Baron-Cohen, S., Leslie, A. and Frith, U. (1985) Cognition 21, 37-46 12 Frlth, U., Morton, J. and Leslie, A.M. (1991) Trends Neurosci. 14,433-438 13 Rogers, S.J. and Pennington, B.F. (1991) Dev. Psychopathol. 3, 137-162 14 Brothers, L. (1995) in The Cognitive Neurosciences (Gazzaniga, M.S., ed.), pp. 1107-1115, MIT Press 15 Baron-Cohen, S. et al. (1994) Br. 1. Psychiatry 165, 640-649 16 Flavell, J.H. (1988) in Developing Theories of Mind (Astington, J.W., Harris, P.L. and Olson, D.R., eds), pp. 244-267, Cambridge University Press 17 Wellman, H.M. (1990) The Child’s Theory of Mind, Bradford Books 18 Pemer, J. (1991) Understanding the Representational Mind, MIT Press 19 Wimmer, H. and Pemer, J. (1983) Cognition 13, 103-128 20 Wimmer, H., Hogrefe, G-J. and Pemer, J. (1988) Child Dev. 59, 386-396 21 Gopnik, A. and Graf, P. (1988) Child Dev. 59, 1366-1371 22 O’Neill, D.K. and Gopnik, A. (1991) Dev. Psychol. 27, 390-397 23 Povinelli, D.J. and de Blois, S. (1992) 1. Comp. Psychol. 106, 228-238 24 Premack, D. (1988) in Machiavellian Intelligence (Byrne, R. and Whiten, A., eds), pp. 160-179, Oxford University Press 25 Povinelli, D.J., Nelson, K.E. and Boysen, S.T. (1990) 1. Comp. Psychol. 104, 203-210 26 Povinelli, D.J., Rulf, A.B. and Bierschwale, D.T. (1994) I. Camp. Psychol. 108, 74-80 27 Cheney, D.L. and Seyfarth, R.M. (1990) Anim. Behav. 40, 742-753 28 Povinelli, D.J., Parks, K.A. and Novak, M.A. (1991) I. Comp. Psycho!. 105, 318-325 29 de Waal, F.B.M. (1982) Chimpanzee Politics, Harper and Row 30 Harris, P.L. et al. (1989) Cognition Emotion 3, 379-400 and T. Preuss - Evolution of theoty of mind In the book review by J.D. Hahn in the July issue of TINS (Vol. 18, p. 327), the title and the ISBN of the book are incorrect. The correct versions are shown below. We apologize to MIT Press and to the readers for these errors. Letters Vol. 2 Please address letters to: Editor, Trends in Neurosciences Elsevier Trends Journals 68 Hills Road Cambridge, UK CB2 I LA. TINS vol. 18, NO. 9, 1995 Publishers Trends in Neurosciences welcomes books for review. Please send books or book details to: Editor Trends in Neurosciences 68 Hills Road Cambridge UK CB2 ILA. to the Editor Letters to the Editor concerning articles published recently in TINS are welcome. Please mark clearly whether they are intended for publication. Maximum length: 400 words. 424 1164-1174 40 Gopnik, A. (1982) 1. Child Lang. 9, 303-318 41 Kagan, J. (1981) The Second Year: TheEmergence ofSe@wareness, Harvard University Press 42 Bates, E. et al. (1979) in The Emergence of Symbols: Cognition and Communication in Infancy (Bates, E., ed.), pp. 69-140, Academic Press 43 Butterworth, G. and Jarrett, N. (1991) Br. 1. Dev. Psychol. 9, 55-72 44 Baron-Cohen, S. (1994) Cum. Psychol. Cognition 13, 513-552 45 Hobson, R.P. (1993) Autism and the Development of Mind, Erlbaum 46 Baldwin, D.A. and Moses, L.J. (1994) in Children’s Early Understanding of Mind (Lewis, C. and Mitchell, P., eds), pp. 133-156, Erlbaum 47 Gallup, G.G., Jr (1970) Science 167, 86-87 48 Gallup, G.G., Jr (1991) in The Self: an Interdisciplinary Approach (Goethals, G.R. and Strauss, J., eds), pp. 121-135, Springer-Verlag 49 Povinelli, D.J. et al. (1993) I. Comp. Psychol. 107, 347-372 50 Gallup, G.G., Jr et al. Anim. Behav. (in press) 51 Gallup, G.G., Jr (1982) Am. I. Primatol. 2, 237-248 52 Gallup, G.G., Jr and Suarez, S.D. (1986) in Psychological Perspectives on the Self (Suls, J. and Greenwald, A.G., eds), pp. 3-26, Erlbaum 53 Lewis, M. et al. (1989) Child Dev. 60, 146-156 54 Johnson, D.B. (1982) Merrill-Palmer Q. 28, 379-388 55 Povlnelli, D.J. (1993) Am. Psychol. 48, 493-509 56 Povinelli, D.J. and Eddy, TJ. Monogr. Sot. Res. Child Dev. (in press) 57 Povinelli, D.J. and Eddy, TJ. Psychol. Sci. (in press) 58 Tomasello, M., Kruger, A. and Ratner, H. (1993) Behav. Brain Sci. 16, 495.511 Book Reviews Corrigendum Long-term Potentiation, ISBN 0 262 02370 9 31 Wellman, H.M. and Bartsch, K. (1988) Cognition 30, 239-277 32 Bartsch, K. and Wellman, H. (1995) Children Talk About the Mind, Oxford University Press 33 Harris, P.L. in Theories of Theories of Mind (Carruthers, P. and Smith, P.K., eds), Cambridge University Press (in press) 34 Povinelli, D.J., Nelson, K.E. and Boysen, S.T. (1992) Anim. Behav. 43, 633-640 35 Povinelli, D.J., Parks, K.A. and Novak, M.A. (1992) Anim. Behav. 44, 269-281 36 Amsterdam, B. (1972) Dev. Psychobiol. 5, 297-305 37 Leslie, A.M. (1987) Psychol. Rev. 94, 412-426 38 Zahn-Waxler, C. and Radke-Yarrow, M. (1982) in The Development of Prosocial Behavior (Eisenberg, N., ed.), pp. 109-137, Academic Press 39 Brownell, C.A. and Carrlger, M.S. (1990) Child Dev. 61, Reviewers If you are interested in reviewing books for Trends in Neurosciences, please contact the Editor, Dr Gavin Swanson, with your suggestions. Tel: +44 I223 3 I596 I ; Fax: +44 1223 464430.