Play Trajectories in Infant Siblings of Children with Autism

Background: Christensen, et al. (2009) examined the play behavior of infant siblings of children with autism who later met criteria for ASD (ASD siblings), infant siblings with other non-ASD delays (Other Delays siblings), infant siblings with no delays (No Delay siblings), and typically developing controls (TD controls) at 18 months of age.  Findings indicated that the ASD siblings showed fewer novel functional play actions and more non-functional repeated actions than the TD controls.  The Other Delays and No Delays siblings also demonstrated more non-functional repeated play actions than the TD control group.
            Further examination of the sample at 24 months of age indicated that there continued to be group differences in functional play between the ASD siblings and TD controls, with the ASD siblings showing fewer novel functional play behaviors.  The No Delays and Other Delays siblings did not differ from the TD controls and there were no other group differences in play. 

Objectives: The current paper examined play behaviors at 36 months of age and the development of play from 18 to 36 months of age.  In particular, this paper addressed whether or not the trajectory of development differs across these four groups of children. 

Methods: This paper examined a sub-sample of the participants from the previous study who have data at 18, 24 and 36 months of age.  This sample consists of 41 of the original 77 participants drawn from the study site at UCLA.  The participants were divided among the aforementioned four groups with ASD siblings (n= 7), Other Delays siblings (n= 7), No Delays siblings (n= 19), and TD controls (n= 7).   A four-minute free play assessment performed at 18, 24 and 36 months of age was coded for symbolic, functional and repeated (functional and non-functional repeated) play actions.

Results: Results indicate that all three sibling groups display significantly more non-functional repeated play than the TD controls (p < .02 for all).  Multilevel modeling techniques were used to examine the trajectories of play over time with observations at 18, 24 and 36 months nested within individuals.  Results indicate significant linear increases in object-directed functional play and decreases in self-directed and non-functional repetitive play.  The effect of time did not differ across individuals. There was significant variability in the intercepts of each play variable examined, which is not explained by group membership.    

Conclusions: The results speak to the importance of play as an early indicator of autism and highlight the robust role of non-functional repeated play behaviors in differentiating individuals at risk from individuals not at risk for ASD.  The results also indicate that the trajectory of change in play from 18-36 months of age may not differ across children with and without risk for ASD.  Instead, the starting point may be more important in predicting outcomes.