Counts describe this library, not the strength of evidence.
How to read this evidence
Intervention trials and step-count associations should be read separately. Exercise can be part of care, but bias and evidence certainty affect estimates. Cognitive test improvements do not by themselves demonstrate dementia prevention; wearable stress is not a psychiatric assessment.
Among 1,380 college students from five universities in eastern China, more physical activity was linked to fewer negative emotions, partly through psychological resilience (effect -0.0324) and through resilience then positive coping (effect -0.0099). The survey is cross-sectional and self-reported, so these pathways are statistical associations, not proof that exercise relieves distress.
Pooling 27 studies with 2,242 people with Alzheimer's disease, cognitive scores improved most at about 650 METs-min/week of exercise overall (SMD 0.691) and 660 METs-min/week for aerobic exercise (SMD 0.909). The estimates are model-based and the confidence ranges are wide, so the exact optimum is uncertain.
In 102 community-dwelling older adults with cognitive frailty, 24 weeks of Baduanjin raised Montreal Cognitive Assessment (MoCA) scores by 2.51 points and lowered frailty scores by 1.94 points compared with no specific exercise. The trial is small, the control group had no activity, and the abstract itself urges caution in explaining the biological mechanism.
In 44 older adults with cognitive impairment, 6 weeks of dual-task resistance exercise improved Mini-Mental State Examination (MMSE) scores more than resistance exercise alone (time-by-group interaction p = 0.044). The trial is very small and short, and no other outcome differed between the groups.
Among 41,620 Chinese college students, moderate-to-vigorous physical activity (MVPA) was associated with lower depression scores (B = -0.062, P = 0.002), partly through better self-rated health and self-efficacy. Only 30.2% did MVPA, and a single survey cannot show that activity itself lowers depression.
In an online survey of more than one thousand physically active adults aged over 21, physical activity was positively linked to health behavior, mental health and psychological resilience, and quality of life moderated four of six tested paths. The sample was a convenience sample of already-active people and the design is cross-sectional, so causal claims are not supported.
In a 3-month randomized trial of 92 postmenopausal women with obesity, exercise and combined diet-plus-exercise improved memory versus controls (p < 0.05), and only the combined group also improved executive function. The trial was small (23 per group) and short, and the three active groups did not differ significantly from one another.
The international journal of behavioral nutrition and physical activity
In a network meta-analysis of 218 trials with 14,170 people who met criteria for major depression, walking or jogging, yoga, strength training, mixed aerobic exercise and tai chi or qigong each gave moderate reductions in depression compared with active controls (walking or jogging: Hedges' g −0.62, 95% credible interval −0.80 to −0.45). Confidence in the evidence was low for walking or jogging and very low for the other treatments, mainly because only one study met the low-risk-of-bias criteria.
Across 42 trials with 2974 older adults, mind-body exercise outperformed control for anxiety (SMD -0.87, 95% CI -1.43 to -0.31) and depression (SMD -0.52, 95% CI -0.71 to -0.34), with Tai Chi ranked first for both. The rankings come from indirect comparisons with very high heterogeneity (I2 95% for anxiety), so the order of exercises is tentative.
In 7,518 community-dwelling Chinese adults aged 60 and over, leisure-time, household and work-related physical activity were linked to better physical and mental quality of life, and depression partly explained the link. This is a single cross-sectional snapshot, so the order of activity, mood and quality of life cannot be established.
Using genetic variants as proxies, higher accelerometer-measured physical activity was associated with lower odds of depression (OR 0.92, 95% CI 0.88-0.97), and higher sedentary time with higher odds of anxiety (OR 2.59, 1.10-4.60). Mendelian randomisation makes causal claims stronger but not certain, and the anxiety interval is wide.
In 291 adults aged 40-70 tracked for eight days, being more active than usual in the 20 or 60 minutes before a test went with better processing speed, and more active days went with better self-rated memory that evening. These are within-person associations in everyday life, not proof that activity caused the gains, and visual memory showed no link.
The international journal of behavioral nutrition and physical activity
Across 35 randomized trials with 5393 children and adolescents, exercise improved depressive symptoms, with aerobic exercise ranking highest (66.2%) and the best results in those under 15 (SMD -0.41; 95% CI -0.63 to -0.19). The abstract gives rankings and subgroup effects but no overall pooled effect, and several confidence intervals and P values do not line up.
Among 7005 Norwegians aged 70 and older, those with a stable high occupational activity trajectory from ages 33 to 65 had higher risk of dementia (RRR 1.34, 95% CI 1.04-1.73) and mild cognitive impairment (RRR 1.80, 1.54-2.11) than those with stable low activity. Job activity was estimated from registry data, not measured, so the link may reflect other differences between workers.
Across 58 studies of children and youth aged 2-24 (32 pooled), more physical activity was linked to fewer negative psychological responses (Fisher's z = -0.198) and more positive ones (z = 0.170) during COVID-19, though both links were weak. The abstract does not say which study designs were included, so it cannot show that activity improved mental health.
In 175 adults aged 65-84 with mild cognitive impairment, 20 weeks of aerobic-resistance exercise improved ADAS-Cog-13 scores at 6 months versus a balance-toning control (mean difference -1.79 points; 95% CI -3.27 to -0.31), with extra gain from computerized cognitive training and none from vitamin D. The trial was small and the ADAS-Cog-Plus results were not significant.
In 92 children aged 8-11 with overweight or obesity, 20 weeks of aerobic plus resistance exercise lowered a cardiometabolic risk score by about 0.38 SDs (95% CI -0.74 to -0.02) versus a wait-list control, and had no effect on mental health. This is a secondary, per-protocol analysis from one city in Spain.
Among 1,615 US adults over 60 with short sleep (weighted 28,607,569), more exercise was linked to higher scores on the Animal Fluency and Digit Symbol Substitution tests up to 960 and 800 MET-minutes/week, after which the benefit levelled off. This is cross-sectional survey data, so it shows association, not that exercise caused better scores.