Research Tuesday- Video Modeling and iMovie (Part 1)

At my recent presentation at Nova Southeastern University, we reviewed research around video modeling and how this speaks to use of the app iMovie (free for devices after Oct, 2013/$4.99). For this reason, I am happy to discuss this material today as part of Research Tuesday.

In researching this topic, I was happy to find a recent meta-analysis (Bellini & Akullian, 2007) that looked at a range of published studies on video modeling (VM). Here are some details of that study:

• The authors emphasize the importance of evidence-based practices in treating autism and the sense of urgency around helping these clients and students to make progress. Single-subject research is defined as EBP when it involved defined practices (methodology and context) that are "implemented with fidelity," resulting in "the practice...functionally related to change in dependent measures," and that the results around the practice be replicated in at least 5 peer-reviewed journals. The study also emphasizes the importance of social validity, meaning that the practice have "social importance and acceptability" from the point of view of the client.
• The study describes video modeling as grounded in Bandura's social learning theory and defines it as "demonstration of desired behaviors through video representation of the behavior." The intervention around VM typically involves watching the video demonstration and then assisting the student in imitating the behavior in context. In video self-modeling (VSM), the student observes himself in video successfully performing a behavior, related to Bandura's concept of the motivation factors around "self-efficacy."
• The meta-analysis examined 29 studies involving VM and VSM applying selection factors (publication in peer-reviewed journal, participants with ASD, focus on VM and VSM for functional behavioral outcomes alone or in combination with other interventions, other factors) and identified 23 studies meeting their criteria. Examining these 23 studies found moderate intervention, maintenance and generalization effects for the methodologies (less generalization found with VSM). Statistical analysis involved application of PND (percent of non-overlapping data), a statistic often used for analysis of single subject designs, for the purpose of establishing these effect sizes. Use of video modeling was also noted as promoting learning factors such as attention and motivation.
• The researchers go on to give specifics about each study, as well as categorized studies according to focus on social or functional (life) skills. For example, in one study, students were supported with prompts to use the skill of requesting items during play, and these prompts were edited out for the video used for repeated viewing by the children. Editing out "hidden supports" is a subcategory of VSM known as video feedforward, according to the article. In this study the 4 participants were found to have a substantial increase in spontaneous requests for toys and items after the intervention. Many other examples are given in the meta-analysis.

A few points in the authors' discussion particularly interested me. As it was written in 2007, the article emphasized the technical expertise required to edit the video footage, particularly in techniques such as feedforward. With the availability of mobile apps for video editing, this is not so much a factor anymore. Additionally, the authors suggested therapists use a storyboard or script, then record the child or children engaging in the desired behavior. I find this a great suggestion as it places video modeling in the context of a larger therapeutic and language process, particularly for higher functioning students who can participate in the storyboarding and planning phase of the activities.

For some practical ways to apply video modeling with pre-made videos in your work, see my recent post for Mindwing Concepts, where I particularly focused on strategies of using situational analysis and story grammar during and after viewing activities.

In next week's post, I'll be looking to back up what I said above about the relative ease of recording videos and editing out prompts (feedforward), in a discussion of iMovie.

Bellini, S, & Akullian, J. (2007). A meta-analysis of video modeling and video self-modeling interventions. Exceptional Children, 73 (3),  pp. 264-287

I will be presenting in the Washington, DC area in September for the Center for Communication and Learning, LLC- hope to see some of you there! Click here for details.

Note: author is a paid consultant for Mindwing Concepts, Inc, for the provision of blog content as relates to the link above.