A comprehensive review of 31 trials reveals that structured exercise, customised to target specific motor abilities, significantly improves object control, locomotor skills, and balance in children and adolescents with autism, highlighting the importance of personalised intervention strategies.
A new meta-analysis in Frontiers in Psychiatry suggests that exercise can meaningfully improve motor abilities in children and adolescents with autism spectrum disorder, with the strongest gains seen in object control, locomotor ability and balance. Drawing on 31 randomised controlled trials published up to July 2026, the review found that structured physical activity produced large positive effects across all three motor domains, strengthening the evidence that movement-based programmes can help address one of the condition’s common but often overlooked challenges.
The analysis pooled data from 880 young people with autism across studies conducted in Asia, Europe, North America, Africa and Australia. According to the authors, the trials covered a wide range of approaches, including fundamental movement skill training, multicomponent exercise, aquatic work, mind-body exercise and exergaming. Participants were mostly aged between about four and 13, and programmes varied from twice-weekly sessions lasting two weeks to longer courses stretching to 18 weeks. The researchers said the breadth of the evidence allowed them to examine whether different types of motor skill respond differently to exercise dose.
That question proved important. Object control skills, such as throwing and catching, improved most when sessions lasted at least 60 minutes, and the authors found a clear dose-response pattern for that domain. Locomotor skills, such as running and jumping, also improved strongly, but the best results in subgroup analyses appeared in programmes delivered three times a week and lasting 10 weeks or less. Stability skills improved as well, although the review did not identify a clear dosage pattern for balance-related gains. The authors said this suggests that exercise programmes for autism should not be treated as one-size-fits-all, because different motor abilities may respond in different ways to the structure of training.
The findings were supported by sensitivity analyses, which held up even when lower-quality trials were removed. The review did note substantial heterogeneity across studies and signs of small-study effects, meaning the results should be interpreted with some caution. Still, the authors argued that the consistency of the overall benefits points to exercise as a practical and low-risk tool for improving motor competence in autism. They also noted that the programmes may work through several pathways, including better coordination, stronger sensory integration and gains in neuroplasticity, the brain’s ability to adapt with experience.
The paper also sits against a backdrop of rising autism prevalence. The authors cited US surveillance data showing that autism was identified in 32.2 per 1,000 eight-year-olds in 2022, or roughly one in 31 children, underscoring the scale of the challenge for schools, therapists and families. Against that context, the review’s message is direct: exercise can help, but the details matter. Longer sessions appear especially useful for object control, while locomotor gains may depend more on how often and for how long training is delivered. The researchers concluded that future programmes should be tailored to specific motor goals rather than using a uniform format for every child.
Disclaimer: This content is for informational purposes only and is not intended to be a substitute for professional medical judgment, advice, diagnosis, or treatment.
