Brief report: prompted pretend play in autism.

A wealth of studies have confirmed Kanner's (1943) view that children with autism have impoverished pretend play (for reviews see Harris, 1993; Jarrold, Boucher, & Smith, 1993; Wulff, 1985). However, there is disagreement about the exact nature of the impairments, the conditions under which the impairments are manifested, and the processes underlying them (Harris, 1993; Jarrold et al., 1993; Leslie, 1987; Perner, 1993). One source of the controversy lies in the fact that the production of pretense in children with autism has been assessed under two general setting conditions: (a) unstructured, spontaneous conditions, and (b) structured, prompted, or elicited conditions. All studies show an impairment in the production of pretend play in children with autism (in comparison to mental-aged matched controls) in unstructured, spontaneous conditions (Baron-Cohen, 1987; Lewis & Boucher, 1988; Riguet, Taylor, Benaroya, & Klein, 1981; Sigman & Ungerer, 1984; Ungerer & Sigman, 1981). After general or specific prompts are given, children with autism tend to produce some pretend play, although in many studies the group differences remain (Riguet et al., 1981; Sigman & Ungerer, 1984). In two studies they disappeared: Children with autism were unimpaired in pretend play (object substitution) under prompted conditions (Jarrold, Boucher, & Smith, 1996; Lewis & Boucher, 1988). Jarrold et al. (1996) also investigated the ability of children with