"Metarepresentation" and "interpersonal relatedness" as explanations of the symbolic play of autistic children

The purpose of this thesis is to examine whether the interpersonal relatedness approach or the metarepresentational (or "theory of mind") approach is a more adequate explanation of the symbolic play of children with autism. This was attempted first, by a comparative theoretical analysis of evidence presented in support of both theories to date, and second, by inclusion of affective and attentive variables in empirical studies of symbolic play in children with and without autism. The roles of affect and attention were crucial, because they are implicit in the interpersonal relatedness theory of symbolic development, but not considered at any length in the metarepresentational approach as it was originally proposed. Current evidence in favour of both theories is presented in chapter I. It was found that findings from studies of 'false belief and symbolic play in autism suggest a 'cognitive deficit' (metarepresentational impairment), but there are also a number of 'affective' impairments which cannot be predicted by such a deficit. It was found that both theories do not have total empirical support. Five empirical studies are presented in chapters 2-7. The influence of a musical stimulus on the symbolic play of children with and without autism is examined in chapters 2 and 4. It was found that more elicited symbolic play (which requires the child to produce the symbolic idea) was demonstrated by children without autism on the inclusion of a musical stimulus. The children with autism were not influenced by the music. An attempt was then made to examine the role of affect and attention during symbolic play, by the development of a rating scale to measure perceived affective mood and attention-to- tasks, in chapters 3, 4, 5 and 6. The scale is used in chapters 4 and 6 to examine relationships of symbolic play, affective mood, and attention to tasks in children with and without autism. Some relationships between symbolic play, affect and attention were found in the children with autism. It was also found that over three occasions the children with autism looked at the experimenter less than the other children. The possibility of a symbolic impairment in children with autism was examined in chapters 4, 6 and 7. Findings were inconsistent, probably because of methodological inconsistency between studies, and flaws in design. Because of this lack of consistent evidence it was not possible to support one of the theories and reject the other. The general discussion (chapter 8) suggests that we cannot choose between the metarepresentational and interpersonal theories on the basis of current evidence, but there remain good reasons for both theories to continue to generate hypotheses to be tested. It is also suggested that recent integrative theories such as that of Baron-Cohen (1994) might bridge the theoretical opposition of Leslie and Hobson.