Association for Behavior Analysis International

This symposium highlights three recent studies evaluating variables that affect the assessment and treatment of observational learning deficits by people with autism and developmental disabilities. In the first study, researchers compared the use of adult models versus peer models on the learning of long response chains via observational learning. All participants showed faster acquisition in the presence of peer models as opposed to adult models. In the second study, observational learning responses including attending, imitation, delayed imitation, and consequence discrimination were first assessed in all participants. A multiple baseline design across five tasks (i.e., hidden item, computer game, academic, construction toy, and building) was used to evaluate the effects of video modeling on the acquisition of observational learning responses. Post training measures showed that all participants learned to engage in observational learning responses across all five tasks with the implementation of the video model. In the final study, researchers incorporated known and unknown stimuli into observational learning training. When targets known to the participant were used, participants learned to continue to respond correctly despite correct and incorrect responses modeled by adults. When targets unknown to participants were used, participants learned to imitate modeled responses that were followed by reinforcement, and to say, I dont know when modeled responses were followed by punishment. All three studies demonstrate the importance of evaluating factors that contribute to observational learning and the value of incorporating observational learning protocols into instruction.

We compared the effectiveness of a peer and staff model on observational learning by four adults with developmental disabilities. An alternating treatment design was used to evaluate the effects of a staff-as-model and peer-as-model condition. Results indicated that all four participants acquired the skill at a faster rate in the peer-as-model condition. Generalization and maintenance of the skills acquired with both models was also evaluated. Implications of programming for observational learning in education and habilitation settings are discussed.

Observational learning is a critical repertoire that is likely necessary for the acquisition of certain social behavior and learning in educational settings. Catania (2013) defined observational learning (OL) as learning from observing anothers behavior and subsequently emitting behavior based upon the consequences that followed the models behavior. Although OL occurs early in typically developing children, research has found that OL may be limited in children diagnosed with an autism spectrum disorder (ASD). In MacDonald and Ahearn (2015) following an assessment of prerequisite skills for OL (attending, imitation, delayed imitation, and consequence discrimination), these skills were taught with prompting and reinforcement to children with an ASD who did not reliably demonstrate OL to produce OL across a variety of tasks. The purpose of this project is to replicate the assessment and treatment procedures in MacDonald and Ahearn and to examine whether OL can be assessed with video modeling. Following assessment of OL skills, three participants diagnosed with an ASD acquired OL with video modeling and demonstrated across- and within-task generalization. IOA was calculated in 33% of sessions across participants with a mean of 98%.