Robot-mediated interventions for teaching children with ASD: A new intraverbal skill

ABSTRACT

Socially assistive robots (SAR) have the potential to impact therapies for Autism Spectrum Disorder (ASD) by supporting clinicians in increasing learning opportunities presented to individuals. Recent research on robot-mediated intervention (RMI) delivery has predominantly addressed social deficits in ASD with positive outcomes. Current literature has minimal focus on teaching children with ASD a skill not known apriori by the individual. Furthermore, it is unclear how to integrate robots in clinical settings because current RMIs do not adhere to Applied Behavior Analysis (ABA) protocols. In this work, we investigated whether an RMI could be utilized to teach children with ASD a completely new language and communication skill they could not exhibit at baseline. We utilized a standard ABA assessment tool to first identify appropriate skills to teach children with ASD. We then developed, implemented, and evaluated an RMI intervention that followed standard clinical operating procedures in ABA and targets participants’ unique skill deficits. We examined the effects of the RMI training on teaching wh-question answering. All sessions were conducted with the SAR as the primary therapist using evidenced-based ABA human teaching protocols. All participants acquired the wh-questions answering skill within seven sessions, which is in line with their typical acquisition rates.

KEYWORDS:

• applied behavior analysis
• autism spectrum disorder
• child–robot interactions
• intraverbals
• socially assistive robots

Acknowledgments

This work was supported by the National Science Foundation CRII Award (#1948224). We would like to thank Rachel Lai, Julie Kopp, Rachel Prescott, and all of the staff of the Oakland University ABA Clinic for their assistance in the data collection and data analysis portions of this research.