Twelve weeks of workouts. No clear memory boost. Mood improved.

Randomised controlled trial (RCT). 52 people. 12 weeks of training; outcomes at baseline and week 12 (optional week-24 visit not analysed, too few attended).

A trial. Chance decided who did what.

Who: healthy, physically inactive middle-aged adults in the UK (London, King's College London), aged 45-65, BMI 20-30, <90 min/week moderate-vigorous activity, no psychiatric disorder in the past 2 years, non-smokers; mean age 55.1, 71% women, 50% White British

• Sharper memorynonemain test
• Low-mood symptomslowersmall drop
• Weight, blood pressurenoneno clear change

More to know, from our read of the paper:

• Small trial: 52 randomised (27 exercise, 25 control); 25 and 22 attended the endpoint visit. Powered only for a LARGE memory effect (d = 0.8), so smaller memory effects could have been missed: say 'no clear change', never 'exercise doesn't help memory'
• Primary outcome (pattern separation / lure discrimination index) was null in ITT (beta 0.33, 95% CI -0.24 to 0.89, p = 0.26) and per-protocol (p = 0.47)
• Mood result is a secondary outcome, and small: PHQ-9 interaction beta -1.18 (95% CI -1.95 to -0.41, FDR-adjusted p = 0.011); weaker in per-protocol (adjusted p = 0.047). Participants were NOT depressed at baseline (minimal-symptom range)
• Open label: participants and researchers knew who exercised; PHQ-9 is self-reported. The control group got no attention-matched activity
• Only 3 months; authors note memory/hippocampus changes in other studies emerged after 6-12 months. 'Memory needs longer' is the authors' suggestion, not a finding

A UK trial randomly split 52 healthy but inactive adults aged 45 to 65: half did 30-minute live online workout classes three times a week for 12 weeks, half kept their usual routine. On the main memory test (telling look-alike pictures apart) there was no clear change. Depression-symptom scores dropped a little more in the workout group, even though nobody started out depressed. Weight, blood pressure and sleep showed no clear change.

We read the abstract and the full text of this paper and checked every number on this page against it (18 facts, checked on 2026-10-29). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.