Skip to main content

Design of an Intelligent Agent to Measure Collaboration and Verbal-Communication Skills of Children with Autism Spectrum Disorder in Collaborative Puzzle Games


AUTHORS

Zhang LLian , Amat AZAshwaq Z , Zhao HHuan , Swanson AAmy , Weitlauf AAmy , Warren ZZachary , Sarkar NNilanjan . IEEE transactions on learning technologies. 2020 10 06; 14(3). 338-352

ABSTRACT

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by core deficits in social interaction and communication. Collaborative puzzle games are interactive activities that can be played to foster the collaboration and verbal-communication skills of children with ASD. In this paper, we have designed an intelligent agent that can play collaborative puzzle games with children and verbally communicate with them as if it is another human player. Furthermore, this intelligent agent is also able to automatically measure children’s task-performance and verbal-communication behaviors throughout game play. Two preliminary studies were conducted with children with ASD to evaluate the feasibility and performance of the intelligent agent. Results of Study I demonstrated the intelligent agent’s ability to play games and communicate with children within the game-playing domain. Results of Study II indicated its potential to measure the communication and collaboration skills of human users.