Mental fatigue: averages, differences and publication bias
· 1 min read
First published:
Research through: 24 April 2026
A review and a corrected bias-sensitive analysis show why an average fatigue effect needs uncertainty attached.
Selected historical research available by this Friday. Older sources are foundational context, not news from this week. Researched in October 2026; no exhaustive surveillance claim.
Topics: mind
Finding
The 2023 review reported a small average reduction in endurance performance after mental-fatigue induction.
Limitations: High risk of bias and publication bias limit confidence and individual transfer.
Evidence level: Moderate evidence
Finding
Age, sex, body mass index and fitness did not significantly moderate the pooled effect in that review.
Limitations: Incomplete reporting and limited power do not prove that individual susceptibility is identical.
Evidence level: Moderate evidence
Finding
The corrected bias-sensitive analysis found a smaller, nonsignificant performance effect after modelling selective publication.
Limitations: A model-dependent nonsignificant estimate does not prove mental fatigue has no effect.
Evidence level: Moderate evidence
Interpretation
A pooled effect cannot tell you how much cognitive strain will change your next race. Participant features did not clearly explain variability in one review, but that does not erase individual differences. Publication-bias modelling changes the estimate, without proving fatigue harmless. No cognitive-training programme or psychological diagnosis follows from these studies.
Further reading: current guides
Update history
- · Historical edition researched, translated and assessed in October 2026; Friday publication date retained.
Copy the site guide and this page. Paste into your AI and ask naturally.