This review pooled research on exercise in adults diagnosed with ADHD. It covered 14 studies with 417 adults aged 18 or older, run from 2015 to 2024, mainly in the United States, Germany and China. Ten studies tested a single bout of exercise. Four tested longer programs, such as Tai Chi, yoga, Pilates and aerobic or mixed training, lasting 6 to 24 weeks. Eight studies, each comparing exercise with no exercise in the same people, were combined in the main analysis.
After a single bout, adults showed a moderate improvement in inhibitory control, the ability to hold back an automatic response. The studies largely agreed with each other, and checks gave similar results. The gain was larger in people not taking ADHD medication than in people who paused medication before the session.
Core ADHD symptoms improved slightly, but that result was not robust. It faded when one study was removed. There were also signs that studies with weaker results may have gone unpublished. Inattention improved slightly, while hyperactivity and impulsivity did not clearly change.
Longer programs were too few to combine. Two of four found no clear effect and two reported gains. Studies used different measures and little blinding. The authors call the evidence preliminary. In exploratory analyses, they found a single moderate- or high-intensity aerobic session lasting at least 20 minutes was associated with better inhibitory control, and they present this as early support for exercise as a non-drug strategy.
Talk to a clinician about your own situation.
The numbers Stats and what the terms mean
The statistics behind the plain-language claims above.
- Xu, 2026: inhibitory control after a single exercise bout versus no exercise, Hedges' g = 0.55, 95% CI [0.32, 0.79], p < 0.001.
- Xu, 2026: how much the inhibitory control studies disagreed, Q = 5.02, I2 = 20.30%, p = 0.285.
- Xu, 2026: inhibitory control in unmedicated participants, Hedges' g = 0.85, 95% CI [0.51, 1.19].
- Xu, 2026: inhibitory control in participants who stopped medication before the session, Hedges' g = 0.37, 95% CI [0.11, 0.64].
- Xu, 2026: whether medication status changed the inhibitory control effect, p for interaction 0.031.
- Xu, 2026: core ADHD symptoms after a single exercise bout, Hedges' g = 0.23, 95% CI [0.03, 0.43], p = 0.024.
- Xu, 2026: core symptoms with LaCount et al. (2022) removed, Hedges' g = 0.18, p = 0.084.
- Xu, 2026: core symptoms under an assumed correlation of 0.3, Hedges' g = 0.24, p = 0.052.
- Xu, 2026: inattention after a single exercise bout, Hedges' g = 0.28, 95% CI [0.01, 0.55], p = 0.042.
- Xu, 2026: hyperactivity and impulsivity after a single exercise bout, Hedges' g = 0.17, p = 0.264.
- Xu, 2026: test for missing studies on core symptoms, Egger's test intercept = 4.12, p = 0.003; fail-safe N (5) below the threshold of 45. What the terms mean:
- Hedges' g is the size of the difference between exercise and no exercise. Roughly, 0.2 is small, 0.5 is moderate and 0.8 is large. A minus sign shows the direction of the change, not its size.
- 95% CI is the range the true effect likely falls in; a range that stays above zero points to a real gain.
- p is how likely a result this large would be by chance alone; below 0.05 is usually treated as unlikely to be chance.
- Q and I2 show how much the studies disagreed with each other; a low I2 means they mostly agreed.
- p for interaction tests whether an effect differs between two groups.
- Egger's test checks whether small studies with weaker results seem to be missing.
- Fail-safe N is how many unpublished null studies it would take to cancel the result; a low number means the result is fragile.
Source: Xu S, Zhao C, Hu L. The effects of acute and chronic exercise on executive functions and core symptoms in adults with ADHD: A systematic review and meta-analysis. Psychology of Sport and Exercise (2026). doi:10.1016/j.psychsport.2026.103088