Development of a gaze contingent method for auditory threshold evaluation in non-verbal ASD children. (June 2019)

• Record Type: Journal Article
• Authors: Sullivan, Brian; Wilson, C. Ellie; Saldaña, David
• Is Part Of: Research in autism spectrum disorders. Volume 62(2019)
• Page Start: 85
• Page End: 98
• Publication Date: 2019-06
• Subjects: Children -- Audition -- Non-verbal -- Eye tracking -- Gaze contingent -- Perceptual threshold; Autism spectrum disorders -- Periodicals; 616.85882005
• DOI: 10.1016/j.rasd.2019.02.006
• Languages: English
• ISSNs: 1750-9467

Abstract

Highlights: Open-source software was developed to test audition in non-verbal ASD children. Testing is accomplished using a gaze-contingent listening game run in Matlab. Hearing thresholds were obtained for most typical children but testing in ASD children was difficult. Discuss challenges making an engaging and accesible auditory test for non-verbal ASD children. Abstract: Background: Minimally verbal children with autistic spectrum disorder (ASD) make up an estimated third of the ASD population (Downs et al., 2005 ), but have been understudied due to difficulties in running experiments with such participants. We sought to develop an instrument to evaluate auditory perception, with the goal of testing both typically developing (TD) and ASD children, including minimally verbal ASD. Audio difference thresholds are typically measured by an audiologist using visual reinforcement audiometry (VRA) techniques, but this requires a trained clinician. Alternatively, mismatch negativity (MMN) via an electroencephalogram can provide an objective threshold measure and the participant can passively attend to stimuli. However, EEG equipment is expensive, and the procedure can be uncomfortable and difficult with anxious or touch sensitive participants. Method: We developed a testing software for estimating auditory thresholds in children using a gaze contingent 'game'. Our open source software uses an eye tracker, Matlab and child-oriented stimuli to automate aspects of VRA. Initial results suggest that audio thresholds can be obtained using our affordable non-invasive system, operated with minimal training, but refinement is necessary. Results: Our method can obtain thresholds for most typical children, but data collection in young ASD children proved more challenging, yielding poor results, and will require further development to make the game more accessible. While promising, these results need to be corroborated with an alternate measure of difference threshold. Conclusion: We document our efforts to design an effective interactive game to assess auditory perception using gaze-contingent eye-tracking methods; and provide case level insights on the testing individual participants and the heterogeneous ability and performance levels within ASD. We discuss the challenges experienced in testing and eye tracking both typical and ASD children to inform clinical and research groups to advance this promising line of research.