Association for Behavior Analysis International

Video modeling has been used widely in behavior analysis. The literature has cited its use as a consumer teaching procedure (e.g., teaching children with Autisms perspective-taking skills, play skills, etc), as well as a method of staff training (e.g., preference assessments, functional analysis, discrete trial training). Video models afford learners access to a well-produced model, which may increase the over quality of the model, as well as provide opportunities to display and highlight specific features of the modeled behavior. This symposium will discuss several applied studies that attempt to extend the current literature. The reviewed studies will discuss advances in digital technology for the delivery of video models, and conceptual extensions of commonly used procedures which may provide clarity on best practices.

Behavior Analyst, Licensed Psychologist, Graduate Students

A concurrent multiple baseline design, across participants, was implemented to investigate the effects of behavioral skills training (BST) on the identification of interest and disinterest of a conversational listener. In addition to BST, pre-recorded video exemplars were created and used to model the target behaviors in tandem with lag reinforcement schedules. During baseline, all participants were shown three video-exemplars with no feedback. The mean accuracy score for all participants during baseline was 12.2%. During the training condition, BST, multiple-video exemplar training and lag reinforcement was used to teach participants how to identify if a conversational listener is either interested or disinterested in what the speaker is saying. All participants researched a 100% mastery criterion with an average maintenance criterion of 98.7%. From baseline to training, all participants had a non-overlapping data score of 96.7%, acquiring the skills needed to identify the interest and disinterest of a conversational listener. Additionally, all participants generalized skills with family and friends’ post-study, with all families observing participants attempts to re-engage the disinterested conversational partner.

High procedural integrity is essential for the implementation of behavior analytic procedures. Procedural integrity can vary across staff; however, evidence-based training procedures can increase the likelihood interventions are implemented in the manner they were designed. Despite the availability of evidence-based training procedures, organizations continue to rely on didactic based trainings. This may be because many training procedures require a trainer to provide feedback. Video modeling is a training procedure that can be used without the use of performance feedback. The literature is mixed regarding the success of video model trainings without the use of feedback. Therefore, the purpose of the current study was to evaluate the use of a video model training procedure, in which participants viewed content through a head mounted device to eliminate the need for performance feedback. The data for three participants were evaluated using a multiple-probe design across target skills. By the end of intervention, all participants met mastery criteria for each target behavior without the need of performance feedback. One participant required video models with voiceover instructions to master each target skill. Two participants met mastery criteria for a portion of targets skills with video models alone. The results of this study suggest that video models delivered through a head-mounted device, also known as immersive video modeling, can train direct care staff in common behavior analytic procedures without the need for performance feedback.

Previous studies indicate that video-modeling is a more effective intervention method to teach children on the autism spectrum how to imitate actions or activities. However, McDowell, Gutierrez, and Bennett (2015) conducted an alternating treatment design in which they compared live-modeling with prompts to video-modeling without prompts and found that live-modeling was a more effective intervention method. The purpose of the current study was to compare whether video-modeling without prompts and live-modeling without prompts is a more effective intervention for increasing imitation in children diagnosed with autism. The results were inconclusive, however the results of one participant suggest that video-modeling without prompts is a more effective intervention treatment for children on the autism spectrum with some prerequisite imitative skills. Implications related to this finding are discussed.