International Meeting for Autism Research (May 7 - 9, 2009): An Extensible, Experimental Video Game for Autism Research and Therapy

An Extensible, Experimental Video Game for Autism Research and Therapy

Friday, May 8, 2009
Boulevard (Chicago Hilton)
M. K. Belmonte , Department of Human Development, Cornell University, Ithaca, NY
Background: Video games offer a way to combine the experimental control of precisely known stimulus timing and parameters with the ecological validity of engaging and motivating tasks. Recently we introduced a video game for autism research encapsulating measures of motion coherence threshold, distribution and shifting of visual spatial attention, auditory-visual integration, visual perceptual disembedding, motor inhibition, central coherence, and first- and second-order "theory of mind."
Objectives: Validate the game software via play-testing with children with and without autism spectrum conditions. Apply the game software with autism-spectrum probands, their clinically unaffected sibs, and unrelated normal controls, to test (1) whether the multiple levels of perceptual and cognitive performance measured by the game are correlated, (2) whether the same matrix of correlations identified in the autism-spectrum group exists in sibs and in controls, and (3) whether game-based training within each measure affects performance in correlated measures.
Methods: The game allows a broad range of cognitive and perceptual domains to be evaluated in a repeatable, learnable, anxiety-minimising context. Choices and decisions in the game are player-centred rather than computer-centred, and largely event-driven rather than time-driven. The game is extensible and inclusive; much of its artistic content has been provided by people with autism spectrum conditions as part of Google Project Spectrum, a project that teaches autism-spectrum schoolchildren how to use three-dimensional modelling software.
Results: The game will be demonstrated. Preliminary findings indicate correlations between low-latency (facilitated) responses for perceptual disembedding, narrow distribution and high shift latency of visual spatial attention, elevated (impaired) motion coherence threshold, and high-latency (impaired) "theory of mind" responses, in all groups studied.
Conclusions: The video game paradigm demonstrates correlations amongst autistic cognitive strengths and weaknesses extending not only within the autism spectrum but also in the broader autism phenotype and in normal cognitive variation. Video games hold strong potential for cognitive skills training not only for people with autism spectrum conditions but for all who can benefit from training. Future work will explore not only behavioural but physiological parameters within the game format.