Interactions With an Empathetic Agent: Regulating Emotions and Improving Engagement in Autism

摘要

Based on years of research establishing the utility of socially assistive robots (SARs) for autism spectrum disorder (ASD) intervention, such robots have now become popular tools, widely used in special education schools, autism care centers, and clinical settings. Most previous studies have explored the roles of SARs as instructors, learning aides, and social-skills trainers, focusing on the learning, language, and social impairments associated with ASD. This article addresses aspects of empathy and emotion regulation (ER) impairments, which are important underlying factors for many atypicalities manifested in ASD. We discuss the design of our robot's emotional capabilities, its emotion-based action library, and the algorithm it uses to regulate a user's emotions. In addition, we describe a user study that evaluates the ER capabilities of an emotionally expressive empathetic agent as well as its capability to prime higher social engagement in a user.