15592
Longitudinal Retrospective Video Analysis of Object Play in Infants with Autism

Thursday, May 15, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
K. P. Wilson1, H. Wiener2, M. DeRamus3, J. Bulluck4, G. T. Baranek4, L. R. Watson5 and E. R. Crais6, (1)Kennedy Krieger Institute, Baltimore, MD, (2)Hilltop Home, Raleigh, NC, (3)CIDD, University of North Carolina at Chapel Hill, Chapel Hill, NC, (4)Department of Allied Health Sciences, University of North Carolina at Chapel Hill, Chapel Hill, NC, (5)Allied Health Sciences, University of North Carolina at Chapel Hill, Chapel Hill, NC, (6)Division of Speech and Hearing Sciences, University of North Carolina at Chapel Hill, Chapel Hill, NC
Background: Existing research describes the atypical nature of play behaviors in young children (toddler age and older) with autism; however, little is reported about the earliest play behaviors of infants who are later diagnosed with autism. This study examined the play behaviors of three groups of infants (typically developing, later diagnosed with autism, and later diagnosed with developmental delay) in order to elucidate early patterns of play behavior and inform decisions regarding early diagnosis and intervention.
Objectives: This presentation will focus on the following three objectives: (1) to describe play behaviors in the three groups of infants; (2) to examine group differences in play, including highest level of play achieved; and (3) to examine correlations between highest level of play achieved during infancy and later developmental outcomes (i.e., Vineland Adaptive Behavior Scale, Childhood Autism Rating Scale, IQ).
Methods: This study compared object play of three groups of infants (total n=92) at two time points (9-12 months and 15-18 months) using longitudinal retrospective video analysis of infants as seen in naturalistic home video contexts (e.g., birthday parties, outdoor play). Videos were matched by content factors and an established electronic coding system was used. Analyses included both descriptive and inferential methods. A unique combination of regression models (i.e., tobit and logistic) was used to examine differences between groups, and non-parametric correlations were utilized to determine relationships between early play and later outcomes.
Results: Descriptive analyses indicated that, across groups, the majority of time was spent in exploratory play, and that the highest levels of play (i.e., functional+symbolic) were exhibited by 41% of typically developing infants, but only by 13% with developmental disabilities and 9% with autism. Results of the regression analyses similarly showed that infants later diagnosed with autism used significantly less high level play than did typically-developing infants (p=.034, OR=11.47). Finally, significant correlations were found between the highest level of play achieved by the atypically developing infants (those later diagnosed with a developmental disability or autism) at 15-18 months and their preschool VABS (r=.255, p=.029) and IQ (r=.236, p=.042) scores.
Conclusions: Combined results not only support the importance of play in early diagnosis and intervention of developmental disabilities including autism, but also elucidate key differences in play at these early stages of development. Interestingly, the logistic analysis results suggest that the presence or absence (vs. duration, variety, flexibility) of high levels of play in the first two years of life as a potentially useful tool for diagnosticians, early interventionists and caregivers. This study not only adds to the literature through the observation of naturalistic video footage, which may be more representative of everyday play patterns, but also through the careful coding and analysis of data using methodology best suited to the unique nature of this type of data set.