The role of imitation in video-based interventions for children with autism

Using a behavioural paradigm, the current thesis investigated the role of imitation in video-based interventions (VBIs) for children with autism spectrum disorder (ASD). A review of the imitation literature was conducted, focusing on imitation skill deficits in children with ASD, followed by a critical review of the VBI literature that focused on pre-intervention assessments of imitation skills and the effect of imitation deficits on VBI outcomes. The need for a standardised multi-dimensional imitation assessment battery for VBIs was established. Based on a comprehensive review of imitation batteries, and considering the nature of VBIs, an imitation assessment tool was developed and called the Video-based intervention imitation battery (VBI-IB). After ethics approval was obtained, the VBI-IB was administered to five participants with ASD (Cane, Shane, Harry, Jake and Matt). The next phase of the research consisted of three experimental studies. Experimental Study 1 was a VBI using full-sequence video point-of-view modelling to teach the construction of a Lego toy. The aim of this study was to compare total imitation scores and match-to-sample scores against video-modelling outcomes. The results indicated that Cane and Harry had good total imitation scores and good delayed-match-to-sample scores, and they performed better on the task of constructing a Lego toy than Shane and Jake, who had lower imitation scores and poorer match-to-sample skills. In Experimental Study 2, the multi-dimensional nature of the VBI-IB was considered against successful outcomes for different kinds of video-modelling tasks. The VBI taught a daily living skill and two imaginary play tasks using a multiple-probe design across three participants (Cane, Harry and Shane). Comparisons were made between the participants’ pre-assessment imitation results on the VBI-IB—with a particular focus on their mimicry and emulation subscale scores—and their outcomes on the daily living task and the imaginary play tasks. Cane demonstrated successful outcomes on all the VBI tasks. Harry, who demonstrated good emulation subscale scores on the VBI-IB, had better outcomes on the daily living task than Shane, who had a better score on the mimicry subscale and better outcomes on the imaginary play tasks. Experimental Study 3 considered procedural modifications to how the VBI was presented in imitation training. A multi-element design was used within a multiple-baseline probe design across four tasks and within one subject (Jake). It was found that modifying a full-sequence video-modelling (FSVM) procedure to a video-prompting (VP) procedure resulted in the successful completion of a task by a participant who had failed to complete it during the FSVM. In addition, the process of video modelling and imitation training improved the participant’s imitation skills by 25 per cent when pre- and post-imitation assessment results were compared.