Procedural-Reasoning Architecture for Applied Behavior Analysis-based Instructions

Autism Spectrum Disorder (ASD) is a complex developmental disability affecting as many as 1 in every 88 children. While there is no known cure for ASD, there are known behavioral and developmental interventions, based on demonstrated efficacy, that have become the predominant treatments for improving social, adaptive, and behavioral functions in children.

Applied Behavioral Analysis (ABA)-based early childhood interventions are evidence based, efficacious therapies for autism that are widely recognized as effective approaches to remediation of the symptoms of ASD. They are, however, labor intensive and consequently often inaccessible at the recommended levels.

Recent advancements in socially assistive robotics and applications of virtual intelligent agents have shown that children with ASD accept intelligent agents as effective and often preferred substitutes for human therapists. This research is nascent and highly experimental with no unifying, interdisciplinary, and integral approach to development of intelligent agents based therapies, especially not in the area of behavioral interventions.

Motivated by the absence of the unifying framework, we developed a conceptual procedural-reasoning agent architecture (PRA-ABA) that, we propose, could serve as a foundation for ABA-based assistive technologies involving virtual, mixed or embodied agents, including robots. This architecture and related research presented in this disser- tation encompass two main areas: (a) knowledge representation and computational model of the behavioral aspects of ABA as applicable to autism intervention practices, and (b) abstract architecture for multi-modal, agent-mediated implementation of these practices.