Differential Effects of Video Modeling As a Function of Targeted Social Behavior for Children with Autism

Background: Video modeling is a commonly used instructional procedure and is now established as an evidence-based practice for teaching individuals with autism (Mason et al., 2012). The procedure involves showing a video display of a model performing a target behavior to an observer and then providing an opportunity for the observer to perform a similar skill. Despite an abundance of recent research examining video modeling for children with autism, several questions remain regarding the efficacy of the procedure when applied to various skills. Additionally, minimal research has examined the possibility of leveraging the attention-capturing qualities of video to simultaneously teach multiple students with autism who would otherwise require one-to-one instruction in order to ensure the student attends to the instructional material.  

Objectives: The primary objective of this study was to examine the effects of video modeling when applied to two types of social initiations: inviting peers to share in play and asking to join others in play. The effects of video modeling on physical and verbal behaviors associated with inviting to play and asking to join in play were compared for 3 children with autism. A secondary objective was to assess the feasibility and preliminary efficacy of video modeling when administered to a small group of preschool students with autism.  

Methods: A single-case reversal design was used to examine the relation between video modeling and the acquisition of specific social initiations directed toward peers by three preschool boys diagnosed with autistic disorder. Two sets of videos of peer models demonstrating the target behaviors were recorded and displayed for all participants simultaneously using an Apple iPad. Participants were then given the opportunity to perform the modeled behaviors by interacting with their peers during a social skills group session.  

Results: During a baseline condition that did not include video modeling, participants demonstrated minimal social initiations. When video modeling was introduced for inviting peers to play, participants imitated some physical initiations though results were variable across sessions. None of the participants engaged in verbal initiations during this condition. All participants showed higher levels of physical and vocal initiations directed toward peers when video modeling was applied to asking to join in play.  

Conclusions: The outcomes of the present investigation support previous research suggesting that video modeling can have differential effects based on skills targeted. The results also suggest video modeling, when used for certain social behaviors, has potential as a methodology for teaching young children with autism in small group settings. These results have implications for the selection of social targets when using video modeling and may offer considerations about procedural elements of the intervention. Future research that examines how video modeling alters behavior is needed to better understand the target behavior selection process and procedural components ideal for an individual student.