Lambourne · Brain research 2010 · meta-analysis and meta-regression · n=?

The effect of exercise-induced arousal on cognitive task performance: a meta-regression analysis.

Cited 1121 times in the scientific literature.

Level 1 - systematic review of randomized trials

Meta-analysis and meta-regression of acute experimental studies

PubMed 20381468 · doi:10.1016/j.brainres.2010.03.091 · record verified 2026-08-26

What was done

Authors conducted a meta-regression analysis of studies examining the effects of acute exercise on cognitive task performance. The meta-analysis assessed moderators including assessment timing (during vs. following exercise, and under vs. over 20 minutes of exercise), cognitive task category (rapid decisions, automatized behaviors, memory storage and retrieval), exercise protocol design (steady-state, fatiguing, or inverted-U paradigms), and exercise mode (cycling vs. treadmill running).

What was found

Cognitive performance varied by timing and mode. During exercise, cognitive task performance was impaired overall with a mean effect size of -0.14, with impairment confined to the first 20 minutes of exercise; otherwise, performance on rapid decisions and automatized tasks was enhanced. Following exercise, performance improved overall with a mean effect size of 0.20, facilitating speeded mental processes as well as memory storage and retrieval across steady-state, fatiguing, and inverted-U protocols. Cycling was associated with improved performance during and after exercise, whereas treadmill running impaired performance during exercise and produced only a small improvement post-exercise.

Why it matters

This meta-analysis demonstrates that acute exercise effects on cognition are not uniformly positive or negative, resolving conflicting literature by identifying timing, task type, and exercise modality as critical moderators.

Limits

The abstract does not report the total number of included studies, overall sample size, participant demographics (such as age or baseline fitness), or variance and confidence intervals around the reported mean effect sizes.

Cited by