Sleep, Energy & Cognition
Does Exercise Improve Thinking, or Just Mood?
The short answer
The two effects are not the same size. A 2016 meta-analysis of 25 randomized trials found exercise had a large antidepressant-sized effect on depression (standardized mean difference 1.11, adjusted for publication bias). But a 2024 meta-analysis of 16 long-term trials (11,402 participants) found exercise alone, sustained for a year or more, produced only a small, non-significant effect on global cognition in older adults (SMD 0.05); only when combined with other interventions did the effect reach significance (SMD 0.09). Specific exercise types tell a more encouraging story: 6 months of resistance training preserved Alzheimer's-vulnerable hippocampal brain regions for a full year after training stopped in adults with mild cognitive impairment, and a poststroke cardiorespiratory exercise trial improved executive function scores at 12 months even though it didn't preserve hippocampal volume.
Quick answers
- What is it?
- Whether exercise improves thinking and memory, or mainly just mood.
- Why are people interested in it?
- Exercise is promoted for brain health as much as physical health, and people want to know which claim the evidence actually supports.
- What has actually been studied?
- A 2016 meta-analysis of 25 randomized trials on exercise and depression, a 2024 meta-analysis of 16 long-term trials (11,402 older adults) on exercise and cognition, and two trials measuring brain structure, including hippocampal volume, directly.
- How strong is the evidence?
- The two effects are not the same size — the antidepressant effect is large and well established, while long-term exercise alone produced only a small, non-significant effect on global cognition in the largest pooled analysis.
- What don't we know yet?
- Whether measured cognitive gains outlast the exercise program itself, which exercise type benefits which cognitive domain, and whether the small cognitive effect is partly downstream of the large mood effect.
- Where can I read the original research?
- Sources at the bottom of this page: Exercise as a treatment for depression: A meta-analysis adjusting for publication bias and Effects of exercise interventions on cognitive functions in healthy populations: A systematic review and meta-analysis.
The short answer: The two effects are not the same size. A 2016 meta-analysis of 25 randomized trials found exercise had a large antidepressant-sized effect on depression (standardized mean difference 1.11, adjusted for publication bias). But a 2024 meta-analysis of 16 long-term trials (11,402 participants) found exercise alone, sustained for a year or more, produced only a small, non-significant effect on global cognition in older adults (SMD 0.05); only when combined with other interventions did the effect reach significance (SMD 0.09). Specific exercise types tell a more encouraging story: 6 months of resistance training preserved Alzheimer's-vulnerable hippocampal brain regions for a full year after training stopped in adults with mild cognitive impairment, and a poststroke cardiorespiratory exercise trial improved executive function scores at 12 months even though it didn't preserve hippocampal volume.
What the research shows
The mood effect is large and well established
Start with the least ambiguous finding in this space. A 2016 meta-analysis in the Journal of Psychiatric Research identified 25 randomized controlled trials comparing exercise against control conditions in people with depression, including nine with a formal diagnosis of major depressive disorder. After adjusting for publication bias — which the authors found had likely caused earlier reviews to underestimate the effect — exercise showed a large, statistically significant antidepressant effect (SMD 1.11, 95% CI 0.79-1.43). The effect was largest for aerobic exercise at moderate-to-vigorous intensity, and held up whether sessions were supervised by an exercise professional or not. This is one of the more robust findings in exercise science: the mood benefit is large, consistent across trial designs, and larger than many earlier, unadjusted estimates suggested.
Cognition tells a smaller, messier story
Set against that backdrop, the cognitive evidence looks considerably weaker. A 2024 meta-analysis in Ageing Research Reviews pooled 16 trials lasting 12 months or longer, covering 11,402 community-dwelling older adults without dementia — the largest, longest body of exercise-cognition evidence assembled to date. Physical exercise interventions alone had a modest-to-no effect on global cognitive function (SMD 0.05, 95% CI -0.04 to 0.13, p=0.25) and did not significantly reduce the risk of mild cognitive impairment (relative risk 0.98) or dementia (relative risk 0.61, wide confidence interval). Multidomain interventions — combining exercise with cognitive training, diet, or other components — did reach statistical significance, though the effect size remained small (SMD 0.09, 95% CI 0.04-0.15, p<0.01). The authors rated the certainty of evidence for exercise alone as moderate, and explicitly concluded that "long-term physical exercise alone did not yield any cognitive gains" at the group level, even though it clearly changes mood.
Shorter trials and specific exercise types show more movement
A broader 2023 meta-analysis in the same journal, pooling 54 randomized controlled trials across children, adults, and older adults — most considerably shorter than 12 months — found exercise did benefit overall cognition and every cognitive subdomain examined (global cognition, executive function, memory, attention, information processing), with older adults showing the largest gains. That meta-analysis also found the type of exercise mattered: aerobic exercise showed the greatest benefit for global cognition and executive function, resistance exercise for executive function specifically, and mind-body exercise (such as tai chi or yoga) for memory. This pattern — real signal in shorter or more targeted trials, muted signal in the largest and longest ones — is consistent with genuinely small, hard-to-detect effects rather than a uniformly null relationship.
Where a structural brain change was actually measured
Two trials measured brain structure directly rather than relying on test scores alone. The SMART trial, reported in NeuroImage: Clinical in 2020, randomized 100 adults with mild cognitive impairment to progressive resistance training, computerized cognitive training, both combined, or neither, and followed them with MRI at 6 and 18 months. Resistance training alone — not cognitive training alone, and not the combination — produced global long-term cognitive improvement and protected hippocampal subfields especially vulnerable to Alzheimer's-related atrophy, an effect still measurable 12 months after the training program had ended. A separate 2025 trial in JAMA Network Open randomized 104 ischemic stroke survivors to 8 weeks of cardiorespiratory exercise versus balance-and-stretching control, with brain MRI and cognitive testing at 2, 4, and 12 months. Exercise did not preserve hippocampal volume relative to control (difference -0.10%, 95% CI -1.10% to 0.87%, not significant) — but the exercise group still scored significantly better on a 12-month executive function test (Trail Making Test B, adjusted mean difference -3.75 seconds, meaning faster completion). A third, smaller trial in early-stage Alzheimer's disease found that fitness gains from a 6-month multicomponent sports intervention were specifically associated with improved cognitive scores and preserved hippocampal-region brain volume — but only in the intervention group, and only in participants who actually improved their cardiorespiratory fitness.
What we don't know yet
Whether measured cognitive gains outlast the exercise program itself. The SMART trial's hippocampal protection persisted 12 months post-training, but this is one trial in one population (mild cognitive impairment); it's not established as a general rule.
Which exercise type benefits which specific cognitive domain, and why. The broader meta-analyses show aerobic, resistance, and mind-body exercise diverging in their apparent strengths, but no trial has directly compared these head-to-head with cognition as the primary, pre-specified outcome.
Whether the small cognitive effect is partly or wholly downstream of the large mood effect. No trial reviewed here formally tested mood improvement as a mediator of the cognitive changes it also measured.
The minimum effective dose for a cognitive benefit, as distinct from the dose that reliably improves mood. The depression meta-analysis found benefits at both moderate and vigorous intensities; the cognition trials didn't consistently test or report a dose-response relationship.
Why exercise alone underperforms multidomain interventions for cognition specifically, when it performs comparably to or better than other single-domain interventions for mood.
Sources
- Schuch FB, Vancampfort D, Richards J, Rosenbaum S, Ward PB, Stubbs B. Exercise as a treatment for depression: A meta-analysis adjusting for publication bias. Journal of Psychiatric Research, 2016;77:42-51. DOI
- Zhang M, Jia J, Yang Y, Zhang L, Wang X. Effects of exercise interventions on cognitive functions in healthy populations: A systematic review and meta-analysis. Ageing Research Reviews, 2023;92:102116. DOI
- Reparaz-Escudero I, Izquierdo M, Bischoff-Ferrari HA, Martínez-Lage P, Sáez de Asteasu ML. Effect of long-term physical exercise and multidomain interventions on cognitive function and the risk of mild cognitive impairment and dementia in older adults: A systematic review with meta-analysis. Ageing Research Reviews, 2024;100:102463. DOI
- Broadhouse KM, Singh MF, Suo C, et al. Hippocampal plasticity underpins long-term cognitive gains from resistance exercise in MCI. NeuroImage: Clinical, 2020;25:102182. DOI
- Brodtmann A, Churilov L, Adkins K, et al. Poststroke Cardiorespiratory Exercise for Brain Volume and Cognition: A Randomized Clinical Trial. JAMA Network Open, 2025;8(8):e2528907. DOI
- David S, Costa AS, Hohenfeld C, et al. Modulating effects of fitness and physical activity on Alzheimer's disease: Implications from a six-month randomized controlled sports intervention. Journal of Alzheimer's Disease, 2025;103(2):552-569. DOI
Article records maintained via PubMed. Peptide Proof reports what studies found. It does not provide medical advice, dosing guidance, or treatment recommendations.
Topic: Brain & Cognitive Health
Sources
- Exercise as a treatment for depression: A meta-analysis adjusting for publication bias
- Journal of Psychiatric Research, 2016 · n=25 · DOI
- Effects of exercise interventions on cognitive functions in healthy populations: A systematic review and meta-analysis
- Ageing Research Reviews, 2023 · n=54 · DOI
- Effect of long-term physical exercise and multidomain interventions on cognitive function and the risk of mild cognitive impairment and dementia in older adults: A systematic review with meta-analysis
- Ageing Research Reviews, 2024 · n=11402 · DOI
- Hippocampal plasticity underpins long-term cognitive gains from resistance exercise in MCI
- NeuroImage: Clinical, 2020 · n=100 · DOI
- Poststroke Cardiorespiratory Exercise for Brain Volume and Cognition: A Randomized Clinical Trial
- JAMA Network Open, 2025 · n=104 · DOI
- Modulating effects of fitness and physical activity on Alzheimer's disease: Implications from a six-month randomized controlled sports intervention
- Journal of Alzheimer's Disease, 2025 · n=46 · DOI
Get the evidence, not the hype.
New research briefs, plainly written and fully sourced. No dosing advice, ever.