Neurotypical children begin to engage in social pretend play with peers and/or siblings during the third year of life (Piaget, 1962).
Rakoczy argues that early pretend play is an advanced form of both individual and collective intentionality (Rakoczy, 2008).
Competence at sharing a common play frame, apparent at around the age of six in neurotypical children, is associated with a sharp decrease in pretend play (cf.
76) suggests that 'one of the characteristics of make-believe play is that it is itself essentially a metacommunicative activity', noting that in social pretend play, 'the participants communicate to each other how to interpret and how to respond to the transformed elements of their shared make-believe world'.
In pretend play, this would involve explicit discussion of issues such as role assignment and plot development, prototypically signalled with framing expressions such as 'Let's pretend'.
At the other end of the continuum is the more traditional view of metacommunication in pretend play, which is formal proposals.
Giffin points out that the sociodramatic role-playing she is particularly concerned to investigate involves an inherent paradox: on the one hand, negotiation of the pretend play is achieved most clearly through out-of-frame metacommunication, but on the other there is a strong impetus to maintain the illusion of the pretend world being developed.
3) noted that, because of their difficulties with pretence, 'The task of coordinating play with peers in a social pretend framework is particularly complex'.
The relationship between these known difficulties for children with autism in engaging in social pretend play, and the ability to make use of a range of metacommunicative resources as outlined in Giffin's model, will be explored in the remainder of this article.
The corpus used for the current study was collected as part of a previous language study (Douglas, 2012) which had research foci other than pretend play.
For this study, we identified all pretend play sequences in the corpus.
We first undertook a basic analysis of the speech act functions of the contributions to the construction of the pretend play sequences from both the child and adult play partner.
The limited research available on imaginative social play suggests that children with autism experience great difficulty in participating in social pretend play episodes.
The metacommunicative function of enactment is not to propose any transformation as part of the pretend play but involves action (either verbal or nonverbal) which continues the previously established storyline or script.
The function of ulterior conversation is to expand on the previously established play script by introducing transformations of objects or events but without overt reference to the pretend play frame.