Speed-Dependent Visual-Motor Tracking Differences in Children With Autism Relate to Core Symptoms

Children with autism spectrum disorder (ASD) often experience challenges integrating visual information to guide motor behavior, particularly in dynamic (speeded) contexts. We investigated the effects of varying stimulus tracking speed on visual-motor integration (VMI), addressing our hypothesis that ASD diagnosis and symptom severity would be associated with impaired performance on dynamic, speeded (vs. slow/static) VMI.

Fifty-four children aged 8-12 years (ASD: n = 16; typically developing [TD] controls: n = 38) successfully completed a continuous grip-force tracking task involving three conditions: static, slow, and fast visual trajectories. A linear mixed-effects model was used to examine the effects of ASD (vs. TD) diagnosis on speeded VMI. We further examined the correlation between speeded VMI and both clinician-rated (Autism Diagnostic Observation Schedule, Second Edition [ADOS-2]) and parent-reported Social Responsiveness Scales, Second Edition (SRS-2) autism symptoms.

A significant interaction between diagnosis and condition was observed (p = 0.03), indicating that group differences in tracking accuracy varied by stimulus speed. Children with ASD showed significantly greater tracking error than TD peers in the fast condition (ASD: 13.5 ± 1.0 [95% CI: 11.3-15.7]; TD: 10.5 ± 0.7 [95% CI: 9.1-12.0]; p = 0.02), but not in the static (p = 0.80) or slow (p = 0.55) conditions. Both groups showed increased error as speed increased, but the ASD group showed greater impairment under speeded conditions. Higher ADOS-2 Total scores predicted greater error in the fast versus slow (p = 0.009) and fast versus static (p = 0.06) contrasts and in the fast condition alone (p = 0.02). Elevated SRS-2 Total scores were similarly associated with greater error in the fast versus slow (p = 0.02) and fast versus static (p = 0.02) comparisons, though not in the fast condition alone (p = 0.23).

The findings support difficulty with speeded dynamic VMI as a scalable autism biomarker. Further development of these biomarkers could be helpful to guiding behavioral interventions, for instance, identifying children who would best respond to slowing visual cues during therapy. Future studies should refine assessment tools to extend assessments of dynamic VMI to younger and more affected children and explore developmentally appropriate interventions tailored for children showing difficulties with speeded VMI.