A Meta-Analytic Review of the Efficacy of Physical Exercise Interventions on Cognition in Individuals with Autism Spectrum Disorder and ADHD.

Zhu et al. (2016) pooled every exercise study they could find for kids and adults with autism or ADHD. They looked at how biking, running, or games changed scores on memory, attention, and planning tests. The team ran a meta-analysis, a method that averages effects across many separate trials.

Overall, exercise programs gave a small-to-medium boost to thinking skills. The gains were not the same in every area; some tests rose more than others. Still, the average direction was clearly positive for both diagnoses.

Ludyga et al. (2023) and Ludyga et al. (2025) seem to disagree. They showed that one quick bout of cycling right before a social task made autistic children worse at face recognition and emotional reading. The key difference is timing: the meta-analysis looked at weeks-long programs, while the newer studies tested a single 10- or 20-minute session.

Wang et al. (2024) extends the story. Their 12-week program for kids with ADHD improved cognitive flexibility, and the gain was driven by falling sleep-latency scores. This gives a possible pathway: regular activity shortens the time it takes to fall asleep, and better sleep then sharpens flexible thinking.

Zhang et al. (2024) shift the target. They found that exercise improved sleep quality in children with ASD just as much as melatonin pills. Together, these papers show exercise can help both thinking and sleep, but you must watch the clock and pick the right goal.

You can safely sell families on long-term exercise for global cognitive gains, but skip the pre-session bike warm-up right before social-skills groups. Track both sleep latency and executive tasks; if sleep shortens, flexibility may rise. Offer exercise as a drug-free sleep aid when parents ask about melatonin.