Deliberate discrepancies as a design strategy for motivating social communication in virtual environments for young children with autism

Abstract

This PhD project explores the phenomenon of young children with autism spectrum conditions (ASC; 8 children aged 5-8 years) detecting discrepancies (i.e. novel or rule-violating occurrences) in existing video data from a virtual environment (VE), and the children's' subsequent social and non-social reactions. Children frequently initiated positive, social communication to human and virtual partners, with individuals appearing to prefer initiating about some discrepancy types over others. These early results suggest that deliberately including a range of discrepancies in future exploratory VEs may motivate initiation for children in this group. However, little is known about the possible types of discrepancy that might exist in a VE, how this population understands them, and how they might adaptively be incorporated into system designs. Missing too is a clear vocabulary or framework for discussing discrepancy research. The next phases of PhD work aim to approach these questions, with continued focus on taxonomizing discrepancy, and exploring design issues in single-user VEs for young children with autism.