REVIEWS ng th the ial ph ia is a debilitating fear of othe publi imation to al life,an e that is dal een found in witl doe b or when ng to g of re of er uli and the ary with d he ion in the ala and th here of more hasic. s tha rk has been do n of whe er soctal comnition is ree uld d by foo way in nt o city of dat akes the ne Pro g of the m of olle ions of pro hat pa ective iour is guid virtual-realitymu es tha every diffe mutual socn e must be systems more differentiated than ust TURE REVIEWSNATURE REVIEWS | NEUROSCIENCE VOLUME 4 | MARCH 2003 | 175 REVIEWS us a better approximation to real life, an issue that is vital for investigating social behaviour. In addition to the examination of the genetic contributions to specific cognitive abilities (for example, in rare human diseases, transgenic mice or different dog breeds), there are also genetic techniques on the horizon that might offer new ways of reversibly lesioning specific brain structures143. Even once future technology shows us more clearly how neural events co-vary with stimuli and behaviour, how to interpret such data will remain a deep theoreti￾cal challenge144. One view is that there are specialized processes for social cognition; another is that social cog￾nition arises out of more basic components that are not so specialized themselves (TABLE 1). This raises the question of whether social cognition is reducible to emotional or motivational processing. For example, when we find a face attractive or trustworthy, do we engage the same mechanisms as when our behaviour (and that of other animals) is reinforced by food? Or does the way in which social stimuli are processed dif￾fer fundamentally from reward and punishment for nonsocial stimuli? There is some indication that the orbitofrontal cortex might be more specialized for social and moral judgements, whereas the amygdala might subserve a broader role in emotional processing that includes basic emotions145,146. On the one hand, all behaviour can be dichotomized as some form of approach or withdrawal, and some the￾oretical schemes have attempted to map all emotions onto a two-dimensional space of reward and punish￾ment147. On the other hand, all the different basic and social emotions, and all the different words that we have for describing patterns in social behaviour, differentiate such behaviour at a considerably finer grain. So the question is: what is the appropriate level of grain at which we should describe social behaviour and the cen￾tral states that regulate it? Although it is implausible that we will find a distinct neural substrate for every differ￾ent emotion and personality factor, it also seems that there must be systems more differentiated than just Social phobia is a debilitating fear of others, public places or public social interaction that might arise from a fear of losing a fundamental aspect of social behaviour — the need to belong to a social group135. Increased amygdala activation has been found in subjects with social phobia when viewing neutral136 or angry 137 faces, or when preparing to give a public speech138. People with psychopathy or social phobia show inverse extremes of emotional responsiveness to social stimuli, and there is evidence that they feature, respectively, exaggerated and reduced activation in the amygdala and the orbitofrontal cortex139. The neuropsychiatric aspects of social cognition that I have discussed so far focus on disabilities. Much less work has been done on exceptional social abilities, although research into Williams syndrome127 and TURNER SYNDROME140 tackle an angle of this issue. But it would be interesting to study what happens in the brains of people who are especially kind, empathic, altruistic or socially skilled. There is a growing interest in enhancement of normal social cognitive skills through interventions ranging from educational strate￾gies during development to pharmacological manipu￾lation in adults, but the paucity of data makes their application to real life premature. Conclusions Progress in our understanding of the neural basis of social behaviour has been rapid. We have identified collections of processes and neural structures that par￾ticipate in our perceptions and judgements of social stimuli, the ways in which we reason about and decide among them, and the ways in which individual and col￾lective behaviour is guided. Further progress crucially depends on advances on two fronts: the development of methods and of theory. In the near future, we will see more studies that use virtual-reality stimuli141, and new techniques such as simultaneous scanning of multiple subjects engaged in mutual social interactions142. Both approaches will give TURNER SYNDROME A genetic disease in which females carry only one healthy X chromosome. It is characterized by an inhibition of sexual development and is accompanied by infertility. There is some evidence from patients with Turner syndrome for the existence of an imprinted X-linked locus that affects social cognition. FLOUR Examiner: I want you to look at the picture and tell me everything that is going on. WMS age 10: (Laughs). Oh no. The mommy left the tap on (pointing to the water). And the boy is trying to get a cookie but the chair is tipping over. (In a high voice, as if addressed to the mother) Mom, won’t you save the boy? (Returning to normal tone) Gosh. She better quickly save her boy. Her son and her daughter. Oh, there’s going to be a flood on her floor. The boy’s in the cookies. Maybe it’s after supper. Maybe. Oh, the mommy is drying the towel. Poor boy, he could get hurt and break his arm. Poor boy, oh poor thing. DNS age 10: Mom wash dishes. A bowl fell. Boy slips, boy pushed. Boy helps mom with dishes. Mom big mess in water. Pushing. (Examiner: Can you tell me anything else about the picture?) (Shakes head.) COOKI Figure 7 | Hypersocial function in subjects with Williams syndrome. Williams syndrome results from the deletion of a small set of genes on one copy of chromosome 7. Although the patients are mentally retarded and severely impaired in visuospatial function, they show an exaggerated interest in other people and remarkable expressiveness and social communicative abilities. When shown a picture, such as the ‘cookie theft’ picture that is commonly used in neuropsychology, people with Williams syndrome tell stories (top) that are not only longer and more complex than those told by same-age subjects with Down syndrome (bottom), but their narratives are also infused with expressive details and social attention devices. DNS, Down syndome subject; WMS, Williams syndrome subject. Modified, with permission, from REF. 127 © (2000) MIT Press