An Experimental Investigation of Social-cognitive Abilities in Infants with Autism: Clinical Implications

Competing theoretical accounts of psychopathological development in individuals with autism emphasize the role of different infant social, cognitive and affective factors, including affective responsivity, pretend play, joint attention and imitation. However, due to the fact that autism is rarely diagnosed before the age of three, until now these abilities have only been studied with school-age children, adolescents or young adults with autism. Taking advantage of a new prospective screening instrument for autism in infancy (Baron-Cohen et al., 1996), the present study compared the performance of 20 month old infants with autism and pervasive developmental disorder to that of children with developmental delay without autism on experimental tasks of empathic response, pretend and functional play, joint attention and requesting behaviors, and imitation. The 20 month old infants with autism failed to use social gaze declaratively in the joint attention task, they showed poor empathic response, fewer imitated modelled actions on objects, and none produced spontaneous pretend play. Surprisingly, the infants with pervasive developmental disorder did not perform significantly differently from the infants with developmental delay without autism on any of the measures. The identification of autism-specific impairments in early social cognitive abilities may have important clinical implications, both for the early diagnosis of the disorder and for the setting of goals and monitoring of progress in early intervention programmes. Infants with autism 3 INTRODUCTION Theoretical accounts of psychopathological development in individuals with autism emphasize the role of various social and cognitive factors that emerge during infancy in typical development. However, to date, research has mostly been conducted with school-age children, adolescents or young adults with autism (for a review, see Baron-Cohen, TagerFlusberg & Cohen, 1993). This is because the disorder is rarely diagnosed before the age of 3 (Gillberg et al., 1990). Research with school-age children, adolescents and young adults with autism has demonstrated autism-specific impairments in a variety of infantile social-cognitive abilities, including empathic response (e.g Sigman, Kasari, Kwon & Yirmiya, 1992), pretend play (e.g Mundy, Sigman, Ungerer & Sherman, 1986; Wetherby & Prutting, 1984), joint attention behaviors (e.g. Baron-Cohen, 1989; Mundy et al., 1986) and imitation (e.g. Dawson & Adams, 1984; Rogers, Benetto, McEvoy & Pennington, 1996). These impairments have been cited as support for the theoretical accounts of psychopathological development put forward by various authors and, in turn, they have been linked to the later problems in social understanding and social communication that characterize autism (Baron-Cohen, 1993, 1995; Hobson, 1993; Leslie, 1987, 1994; Meltzoff & Gopnik, 1993; Mundy, 1995; Mundy, Sigman & Kasari, 1993; Rogers & Pennington, 1991). To briefly summarize the research findings with older school-age children, adolescents and young adults with autism: (1) Individuals with autism show poor coordination of affective response in that they are less likely than controls to combine smiles with eye contact and to smile in response to smiles from their mother (Dawson et al., 1990; Kasari et al., 1990), and are impaired in their empathic responses to signals of distress (Sigman et al., 1992). (2) In unstructured or free-play conditions, children with autism produce significantly less pretend play, but intact functional play, compared to chronological or mental agematched comparison groups (Baron-Cohen, 1987; Lewis & Boucher, 1988; Mundy et al., 1986; see Jarrold, Boucher & Smith, 1993, for a review). (3) There is substantial experimental evidence for impairments in both the production and comprehension of joint attention behaviors in children with autism. Many studies have shown, for example, that while children with autism are able to use eye contact and gestures to request objects (protoimperative gestures) or to engage in social action routines, they nevertheless do not use eye contact and gestures to share interest in objects or their properties