Affect-sensitive assistive intervention technologies for children with autism: An individual-specific approach

Abstract

This paper presents an overview of our work on the investigation of an affect-sensitive system to be applied in future autism intervention. A physiology-based affect-inference and adaptation framework was proposed, which could endow the assistive intervention technology with the capability of detecting the affective states of a child with autism spectrum disorder (ASD) and responding to them accordingly. Given the importance of affective cues in human-machine interaction and its significant role in autism intervention practice, this work marks an important step towards intelligent intervention systems that embody human-like functionality - affect recognition and adaptation. To account for the spectrum nature of autism and the differences of emotional expression, an individual-specific approach was employed for affective modeling. Two computer-based cognitive tasks were designed for eliciting target affective states considered important in autism intervention. Experimental results on real-time affect recognition and adaptation are presented based on a study of 6 children with ASD in a proof-of-concept experiment (i.e., robot-based basketball game). The preliminary results demonstrated that such an affect-sensitive adaptive system could hold promise for computer/robot-assisted autism intervention.