Altamens

Altamens Research · Evidence review

Does Exercise Help the Aging Brain? Reading the Trial Evidence

Randomized trials show physical exercise can improve cognitive test scores — most clearly in people who already have impairment. Here is what that does, and does not, mean.

Altamens Research9 min read

Abstract

Physical inactivity is one of the modifiable dementia risk factors in the life-course model, and exercise is among the most studied non-drug interventions for cognition. This review reads the recent randomized-trial and meta-analytic evidence and asks a careful question: what kind of cognitive benefit does exercise actually produce, and in whom?

Across 2024 meta-analyses, exercise improves cognitive and especially executive-function test scores, with effect sizes from small (in broad dementia populations) to moderate or larger (for multicomponent and dual-task programmes in mild cognitive impairment). Because much of this evidence comes from people who already have impairment, it supports exercise as a reasonable, low-risk part of healthy ageing — not as a guarantee against dementia for any individual.

Key findings

  • Randomized trials find exercise improves cognitive test scores in older adults, with effects ranging from small to moderate depending on the population and outcome.
  • The clearest recent evidence is in people who already have mild cognitive impairment or dementia — not proof that exercise prevents dementia in healthy adults.
  • Aerobic, multicomponent and dual-task programmes tend to show the largest effects on executive function.
  • Dose-response work in Alzheimer's suggests a benefit around 150 minutes of moderate activity per week, with diminishing returns at very high volumes.
  • Confidence intervals are wide and publication bias is a real concern, so individual results will vary.

1The question, framed carefully

Physical inactivity appears as a modifiable factor in the 2024 Lancet Commission's life-course model of dementia risk.2 That places exercise firmly in the conversation about cognitive health — but it does not tell us how large the benefit is, what kind of exercise matters, or whether the effect seen in trials translates into preventing dementia. Those are separate questions, and the honest answers are more specific than the headlines.

2How large is the effect?

In a meta-analysis of randomized trials in older adults with dementia, exercise training produced a small improvement in cognitive function overall.3 In mild cognitive impairment, network meta-analyses report larger effects, particularly for multicomponent and dual-task programmes and for executive function — though these estimates carry wide confidence intervals.4,5

Figure 1Exercise effects on cognition (standardized mean differences)
All-cause dementia · global · BJSM 2022g = 0.19 (0.05–0.33)
MCI · global (multicomponent) · Front Psychiatry 2024g = 1.09 (0.68–1.51)
MCI/dementia · executive (any exercise) · Eur Rev Aging 2024g = 0.69 (0.50–0.89)
MCI/dementia · executive (dual-task) · Eur Rev Aging 2024g = 1.14 (0.24–2.04)
Alzheimer's · optimal dose · BMC Geriatrics 2024g = 0.69 (0.37–1.04)
0.000.501.001.50

By convention ≈ 0.2 is a small effect, ≈ 0.5 medium, ≈ 0.8 large. Wide intervals mean greater uncertainty.

Effects range from small to moderate-plus. These come from different meta-analyses, populations and outcomes and are not directly interchangeable. Wide intervals — especially for dual-task executive function — reflect real uncertainty; one meta-analysis also flagged possible publication bias. Filter by source to compare like with like.

3How much, and what kind?

A dose-response network meta-analysis in Alzheimer's disease estimated an optimal exercise dose of roughly 650 METs-minutes per week — about 150 minutes of moderate-intensity activity — with a nonlinear pattern in which benefits plateaued or diminished at very high volumes.6

  • Aerobic exercise — consistently among the most effective modalities for cognitive outcomes.
  • Multicomponent programmes — combining aerobic, strength and balance work — showed the largest global-cognition effects in MCI.
  • Dual-task training — moving while doing a cognitive task — showed the largest executive-function effect in one meta-analysis, albeit with a very wide confidence interval.

4Limitations

  • Many trials enrol people who already have MCI or dementia, so results may not generalise to healthy prevention.
  • Confidence intervals are wide, and at least one meta-analysis flagged possible publication bias — small positive trials may be over-represented.
  • Cognitive outcomes and exercise protocols differ across studies, so pooled effect sizes are approximate.
  • Trial improvements on cognitive tests do not establish that exercise prevents dementia in the general population.

The reasonable reading is the one national guidance already gives: regular physical activity is a low-risk, broadly beneficial part of healthy ageing that is associated with better cognitive outcomes — decided, when a health condition is involved, together with a clinician.

References

  1. 1.National Institute on Aging (NIA) Cognitive Health and Older Adults. U.S. National Institute on Aging, 2026. https://www.nia.nih.gov/health/brain-health/cognitive-health-and-older-adults
  2. 2.Livingston G, et al. (Lancet Commission) Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. UCL News summary of the Lancet Commission, 2024. https://www.ucl.ac.uk/news/2024/jul/nearly-half-dementia-cases-could-be-prevented-or-delayed-tackling-14-risk-factors
  3. 3.Barha CK, et al. Effects of exercise training on the cognitive function of older adults with different types of dementia: a systematic review and meta-analysis. British Journal of Sports Medicine, 2022. 25 RCTs in the meta-analysis; all-cause dementia g = 0.19 (0.05–0.33). https://doi.org/10.1136/bjsports-2021-104955
  4. 4.Systematic review & network meta-analysis Optimal dose and type of exercise to improve cognitive function in patients with mild cognitive impairment: a network meta-analysis of RCTs. Frontiers in Psychiatry, 2024. 42 studies, 2,832 participants. https://doi.org/10.3389/fpsyt.2024.1436499
  5. 5.Systematic review & meta-analyses Identifying exercise and cognitive intervention parameters to optimize executive function in older adults with MCI and dementia. European Review of Aging and Physical Activity, 2024. 35 RCTs; note: Egger's test suggested possible publication bias. https://link.springer.com/article/10.1186/s11556-024-00357-4
  6. 6.Bayesian model-based network meta-analysis Effective dosage and mode of exercise for enhancing cognitive function in Alzheimer's disease and dementia. BMC Geriatrics, 2024. 27 RCTs, 2,242 AD patients. https://link.springer.com/article/10.1186/s12877-024-05060-8