Evaluating the effects of camera perspective in video modeling for children with autism: Point of view versus scene modeling

Video modeling has been used effectively to teach a variety of skills to children with autism. This body of literature is characterized by a variety of procedural variations including the characteristics of the video model (e.g., self vs. other, adult vs. peer). Traditionally, most video models have been filmed using third person perspective (i.e., scene models), where the viewer is watching the actor perform in a scene. Recently, studies have successfully incorporated the use of first person perspective into video models (i.e., point of view models), where the view is directly from the actor's point of view. Currently, no studies have directly compared the effects of camera angle on learning when video models are used as teaching tools. Six boys with autism ages 4-8 years learned yoked pairs of tasks, with one task assigned to each type of modeling condition. The effects were evaluated using an adapted alternating treatments design that allowed for a direct comparison between conditions with task difficulty held constant. Few differences in rate of acquisition and attention to the model were observed. Video modeling was not always successful as a teaching tool for targeted tasks. Supplemental teaching strategies (e.g., in vivo modeling with error correction) were employed when video modeling was ineffective for one or both tasks. This study provides evidence that camera angle does not generally have an effect on video modeling effectiveness. It also provides further evidence that video modeling may not always be an effective teaching tool for all children with autism.