Affect Recognition in Autism: A single case study on integrating a humanoid robot in a standard therapy

Abstract

Autism Spectrum Disorder (ASD) is a multifaceted developmental disorder that comprises a mixture of social impairments, with deficits in many areas including the theory of mind, imitation, and communication. Moreover, people with autism have difficulty in recognising and understanding emotional expressions. We are currently working on integrating a humanoid robot within the standard clinical treatment offered to children with ASD to support the therapists. In this article, using the A-B-A' single case design, we propose a robot-assisted affect recognition training and to present the results on the child's progress during the five months of clinical experimentation. In the investigation, we tested the generalization of learning and the long-term maintenance of new skills via the NEPSY-II affection recognition sub-test. The results of this single case study pilot suggest the feasibility and effectiveness of using a humanoid robot to assist with emotion recognition training in children with ASD.