An investigation into the pretend play shown by children with autism.

This research investigated whether the absence of pretend play typically shown by children
with autism is the result of a global inability to pretend, or reflects a failure to utilise intact pretend
play abilities. A first experiment found that children with autism were impaired in their ability to
produce spontaneous pretend play, relative to a matched group of children with moderate learning
difficulties. They were also impaired in their production of pretence in elicited play conditions, in
which direct encouragement to play was provided by the experimenter. However. a second
experiment revealed that these children were not impaired in their ability to carry out pretend
instructions. Further, a third experiment showed that they were unimpaired in their ability to
comprehend pretend acts which the experimenter demonstrated before them.
These findings suggest that pretend play is something that children with autism can engage
in, at a basic level at least. Consequently, two final experiments aimed to determine why children
with autism do not utilise this capacity spontaneously. The firs~ of these tested an 'executive
deficit' hypothesis, which suggests that a failure to pretend is caused by a failure to disengage from
the functional salience of objects. The results of the experiment disconfumed this prediction. The
second test examined whether children with autism have problems in generating pretend acts, and
found that this was the case.
It is therefore hypothesised that children with autism suffer from some form of generativity
impairment, which impinges on their apparently intact ability for pretence. This suggestion fits in
with the pattern of results obtained from all the studies, as children were only impaired when the
idea for pretence was not provided. Possible cause of such an impairment are discussed. as are the
implications of these findings for our understanding of the psychology of pretend play.