Brock, J., Einav, S., & Riby, D. (2008). The other end of the spectrum? Social cognition in Williams syndrome. In T. Striano & V. Reid (Eds.), Social Cognition: Development, Neuroscience and Autism. Oxford: Blackwell.
In typical development, social-cognitive abilities are, by definition, predictable based on the child’s chronological age. As a consequence, it is diffi cult to know whether associations between different social-cognitive skills and other capabilities refl ect underlying causal relationships or mere maturational coincidence. In atypical development, this association between age and ability is disrupted, and, in many cases, social-cognitive functions appear to develop “out of synch” with one another or with the child’s general developmental level. By looking at developmental disorders and trying to determine the reasons why specifi c skills may be relatively more impaired in one disorder (or individual) compared with another, it may be possible to tease apart hypothetical causal mechanisms and determine the factors that act as constraints upon development. The preceding chapters in this book have illustrated how the study of social-cognitive defi cits in autism can illuminate the processes involved in typical social development. In this chapter, we review the social-cognitive capabilities of individuals with Williams syndrome – a rare genetic disorder that is often seen as representing the opposite of autism.
Williams syndrome is caused by the deletion of about twenty-fi ve genes in the 7q11.23 region of chromosome seven (see Donnai & Karmiloff-Smith, 2000) and is associated with a number of medical and physical characteristics, including cardiac anomalies, excessive blood calcium levels, and an unusual “elfi n” facial profi le (see Morris, 2006). Although there is considerable individual variation, the overwhelming majority of people with Williams syndrome are characterized as having mild to moderate intellectual disability (see, e.g., Howlin, Davies, & Udwin, 1998). Crucially, however, the cognitive profi le is somewhat uneven: visuo-spatial and number skills are particularly weak (e.g. Farran & Jarrold, 2003; Paterson, Girelli, Butterworth, & Karmiloff-Smith, 2006), whereas, by comparison, language and face-processing skills are considered to be much less severely affected.
Of particular relevance to this book and chapter, individuals with Williams syndrome are often described as having a characteristic “hypersociable” personality, behaving “as if everyone is their friend” (Jones et al., 2000). Parents of individuals with Williams syndrome rate their children as being more empathetic, sensitive, and gregarious than do parents of typically developing children or individuals with other developmental disorders such as Down syndrome, autism, or intellectual delay of mixed aetiology (Doyle, Bellugi, Korenberg, & Graham, 2004; Dykens & Rosner, 1999; Gosch & Pankau, 1997; Jones et al., 2000; Klein-Tasman & Mervis, 2003). Researchers have also reported increased use of social engagement devices and emotion inferences (Reilly, Harrison, & Klima, 1995; Reilly, Losh, Bellugi, & Wulfeck, 2004; see also Jones et al., 2000) and an increased tendency to react empathetically toward another person’s distress (Tager-Flusberg & Sullivan, 2000).
Moreover, in initial experimental cognitive studies, individuals with Williams syndrome were found to perform well on formal tests of theory of mind and emotion recognition that individuals with autism typically fail (Karmiloff-Smith, Klima, Bellugi, Grant, & Baron-Cohen, 1995; Tager-Flusberg, Boshart, & Baron-Cohen, 1998).
Such fi ndings led Bellugi and colleagues to conclude that individuals with Williams syndrome “exhibit a striking contrast to the social and language profi les of individuals with other disorders such as autism” (Bellugi, Lichtenberger, Mills, Galaburda, & Korenberg, 1999, p. 200; see also Bellugi, Wang & Jernigan, 1994). In a similar vein, Baron-Cohen and Hammer (1997) argued that, whereas individuals with autism have extreme “male brains,” with better spatial skills than social skills, those with Williams syndrome show the reverse pattern and, as such, may be characterized as having extreme “female brains.”
This view of Williams syndrome and autism as diametric opposites has, however, proven to be somewhat simplistic (Tager-Flusberg, Plesa-Skwerer, & Joseph, 2006). Despite their sociable and empathetic personalities, individuals with Williams syndrome are often reported as having high levels of social anxiety (Dykens, 2003; Udwin, Yule, & Martin, 1987). Children with Williams syndrome typically prefer adult company to mixing with their own age group, and have great diffi culty making and sustaining friendships (Einfeld, Tonge, & Florio, 1997; Rosner, Hodapp, Fidler, Sagun, & Dykens, 2004; Udwin et al., 1987). The two disorders also overlap clinically. Leyfer, Woodruff-Borden, Klein-Tasman, Fricke, and Mervis (2006) reported that 7 percent of children with Williams syndrome met Diagnostic and Statistical Manual (DSM) criteria for autism spectrum disorders – considerably higher than in the general population. Similarly, Leekam, Burt, & Arnott (2006) noted that, although individuals with Williams syndrome were less impaired than those with autism on the socialization and repetitive behavior scales of the Diagnostic Interview for Social and Communication Disorders (Wing, Leekam, Libby, Gould, & Larcombe, 2002), ratings of communication skills were in fact comparable across the two groups.
Since the mid-1990s, experimental research on Williams syndrome has also moved on considerably, and this is the main focus of the present chapter. We aim to provide a comprehensive review of recent fi ndings in relation to social cognition in Williams syndrome and, where possible, make direct comparisons between studies of Williams syndrome and studies that have employed similar methodologies to investigate autism. We begin by briefl y reviewing the language capabilities of individuals with Williams syndrome, with particular reference to pragmatic skills. We then consider the performance of individuals with Williams syndrome on formal tests of theory of mind and its potential precursors in joint attention, before moving on to look at various aspects of face processing. Finally, we review recent evidence concerning the neural mechanisms that potentially underlie the hypersociable personalities of people with Williams syndrome.