Enhancing the Social Interactions of Preschool Children with Autism: The Effectiveness of a Peer-Mediated Intervention

Background:  Deficits in social interactions are a defining feature of autism spectrum disorders (ASD) and a powerful predictor of individual outcomes (e.g., Carter, Davis, Klin, & Volkmar, 2005). Although a wide variety of methods for improving social skills has been examined, it is still unclear which approaches are most effective with very young children. Some of the techniques that have empirical support include ABA, peer training, and video modeling.  However, each strategy has its limitations, especially when used in isolation (e.g., Ferraioli & Harris, 2011; Reichow & Volkmar, 2010).

Objectives:  This pilot study examined the efficacy of a new manualized social skills treatment package for preschool children with ASD that combines several techniques with demonstrated efficacy in order to enhance intervention effects. 

Methods:  The participants were 19 children with ASD who were enrolled in an early intervention preschool program that also served typically developing peers. Children with ASD were assigned to either the treatment group (n = 11, M = 4.9 years, SD = 0.7) or the waitlist control group (n = 8, M = 4.6 years, SD = 1.1). Ten typically developing peers participated in 8 training sessions that consisted of explanations of strategies, modeling, role play, and reinforcement. The intervention itself was delivered across 10 weeks in natural social environments, including recess and classroom playtime.  Each participant in the treatment group received approximately fifteen intervention sessions targeting parallel play, social initiation, imitation, shared positive affect, and turn taking.  In addition to verbal and visual prompts, some sessions also included video modeling of targeted social skills. Participants and peers received immediate social praise and small tangible rewards for their efforts throughout treatment. Data were collected at three time points: pre-intervention, post-intervention, and at a ten week follow-up assessment. Parents and teachers completed the Autism Spectrum Rating Scale (ASRS) which produced Social/Communication and Unusual Behavior composite scores.  Clinicians, who were blind to group assignment, rated participant’s social and play skills using subscales of the Verbal Behavior Milestones Assessment and Placement Program (VB-MAPP). Group differences post-intervention and at the 10-week follow-up were examined using univariate ANCOVAs.

Results:  After adjusting for pre-intervention scores, parents reported that children in the treatment group demonstrated significantly fewer deficits in Social/Communication skills post-intervention compared to those in the control group, F(1,16) = 4.64, p < .05, Cohen’s d =  -1.06. At the 10-week follow-up there were no significant group differences in parent reported Social/Communication skills, F(1,16) = 1.51, p > .05, Cohen’s d = -0.62. Results of clinician ratings revealed that at the 10-week follow-up, children in the treatment group demonstrated significantly more appropriate play skills compared to the control group, F(1,16) = 4.69, p < .05, Cohen’s d =  0.82.

Conclusions:  Integrating evidence-based intervention techniques (e.g., ABA, peer training, video modeling) in a natural social environment may improve the social and play skills of preschoolers with ASD.  Further examination of this intervention, potentially using more sensitive outcome measures and a larger, more diverse sample, is warranted.