Quantification of Predictive Motor Impairments in Children with Autism Spectrum Disorder

Background: Anecdotal reports and selected research results suggest that individuals with autism spectrum disorder (ASD) exhibit difficulties in motor coordination, particularly when interacting with dynamic objects, like catching a ball. A recent theoretical framework from our group suggests that the seemingly distinct manifestations of ASD in multiple domains may share a common core: an impaired ability to make predictions. Difficulties in language, social interactions, repetitive behaviors, and motor coordination may all result from the same underlying impairment, albeit at different time scales. This study examined motor predictive impairment through multiple tasks of catching and bouncing a ball, which involve predicting and intercepting a moving object.

Objectives: Motivated by the theory of predictive impairment in autism, our hypothesis is that individuals with ASD will show impaired motor coordination when interacting with moving objects, specifically when performing actions that involve prediction.

Methods: Four children with ASD (IQ-matched, divided into 2 age groups, 7-9 and 10-12 years) and 21 neurotypical (NT) children performed 5 ball interception tasks in a virtual environment that afforded controlled manipulation of the time window for prediction, while simplifying the coordination challenges for hand movement. Subjects either pressed a button to predict the time or location of a launched ball (low motor challenge), or they vertically moved a paddle to catch or bounce a ball to hit a target (higher motor challenge). Additionally, subjects performed naturalistic ball catching with two hands, where 3D kinematics of joint movements and muscle activity of arm, leg, and postural muscles were recorded to quantify endogenous predictive responses. A set of control tasks assessed more elementary motor abilities, such as postural sway, reaction time, and smoothness of hand movements.

Results: NT and ASD children successfully completed all tasks. However, ASD children performed with significantly higher errors in temporal and accuracy metrics. In virtual ball catching, when the challenge for predicting the ball’s trajectory was increased by partial occlusion, ASD children’s accuracy registered a disproportionately large decrease relative to NTs. Additionally, ASD children displayed lower accuracy when intercepting a ball to bounce it to a target. Importantly, in reaction times and smoothness, tested in control tests designed to closely relate to the experimental tasks, ASD children did not differ from NT children. Similarly, while NT children achieved significantly more safe ball catches during the naturalistic ball catching task, their performance in postural balance alone was not superior to ASD children. Preliminary EMG analysis suggests that when catching balls ASD children show less coordinated muscle activity than NT children in stabilizing trunk muscles.

Conclusions: Results from a sequence of prediction-based tasks reveal that despite considerable individual differences in skill, ASD children on average show impaired motor performance compared to NT children. These data are consistent with our hypothesis that the ability to predict may be an explanatory construct common across ASD individuals despite their phenotypic variation. Given its broader theoretical embedding, the results of this study have implications beyond motor skills towards a more encompassing understanding of autism.