Brittany Travers, PhD – Slide of the Week

Brittany Travers Slide of the Week

Title: Whole-body movement during videogame play distinguishes youth with autism from youth with typical development

Legend: In this study, kinematic (joint movement) and postural sway data (shifts in balance on a balance board) were collected during multiple sessions of videogame play in youth with autism spectrum disorder and youth with typical development (ages 7-17 years). With these data, we examined whether movement patterns could distinguish between the two groups. (A) Study setup: Microsoft Kinect camera and Wii Balance Board sent the joint kinematic and postural sway data to the computer, which controlled the biofeedback-based balance training game and recorded the data. Participants were asked to maintain each balance position for as long as possible. (B) Shows the architecture of our machine-learning classifier ensemble. Specifically, we used an ensemble of random forest (RF) classifiers where each RF dealt with the input vectors (variance and entropy of the kinematic and postural sway data) for a specific pose. Each RF in turn consisted of 10 decision trees (DTs) trained using the CART algorithm. To evaluate our classification approach, we used stratified 5-fold cross-validation (CV) to divide participants into five groups, and then reported the average CV accuracy measures as well as individual fold results. This classification approach showed that whole-body movement was able to distinguish those with autism and those with typical development with high accuracy (78-94%).

Citation: Ardalan, A., Assadi, A., Surgent, O. J., & Travers, B. G. (2019, May). Whole-body movement during videogame play distinguishes youth with autism from youth with typical development. Poster presented at the International Society for Autism Research, Montreal, QC.

Abstract: Individuals with autism spectrum disorder often struggle with motor difficulties across the life span, and these motor difficulties may affect independent living skills and quality of life. Yet, we know little about how whole-body movement may distinguish individuals with autism spectrum disorder from individuals with typical development. In this study, kinematic (joint movement) and postural sway data (shifts in balance on a balance board) were collected during multiple sessions of videogame play in 46 youth with autism spectrum disorder and 18 age-matched youth with typical development (ages 7-17 years). The youth on the autism spectrum exhibited more variability and more entropy in their movements, suggestive of more difficulty with movements. Machine learning analysis of the youths’ motor patterns distinguished between the autism spectrum and typically developing groups with high aggregate accuracy (up to 94%), with no single region of the body seeming to drive group differences. Moreover, the classification results corresponded to individual differences in performance on standardized motor tasks (thereby, validating our machine learning measure) and measures of autism symptom severity (suggesting more movement atypicalities in those with more severe autism symptoms). The machine learning algorithm was also sensitive to age, suggesting that motor challenges in autism may be best characterized as a developmental motor delay rather than an autism-distinct motor profile. Overall, these results reveal that whole-body movement is a distinguishing feature in autism spectrum disorder and that movement atypicalities in autism are not localized to just one part of the body.

About the Lab: The Motor Brain and Development Lab is dedicated to advancing knowledge about motor development, brain development, and independent living skills to promote and enhance quality of life for individuals with and without developmental disorders. Our current projects specialize in examining motor and brain development in individuals on the autism spectrum.

Slide of the Week Archives