Brain-wide association studies (BWAS) need thousands of participants for reliable results, as effect sizes are smaller than previously thought. Sampling variability in smaller studies leads to inflated and often irreproducible findings, emphasizing the need for larger, well-powered studies in future BWAS research.

E.A.E., D.A.F., and N.U.F.D. have financial interests in NOUS Imaging Inc. related to FIRMM motion-monitoring software. O.M.-D., E.A.E., A.N.V., D.A.F., and N.U.F.D. may also receive royalties from FIRMM technology.

This study addresses a crucial issue in neuroimaging research: the reproducibility of brain-wide association studies (BWAS). Using massive datasets, the authors demonstrate the need for thousands of participants to achieve reliable results, challenging previous smaller studies. While limitations regarding generalizability and reliance on correlations exist, the sheer scale of the analysis and its implications for future research warrant a strong rating. The declared conflicts of interest are noted and considered in the rating.