English

FMRI Clustering and False Positive Rates

Quantitative Methods 2022-06-08 v1 Applications

Abstract

Recently, Eklund et al. (2016) analyzed clustering methods in standard FMRI packages: AFNI (which we maintain), FSL, and SPM [1]. They claimed: 1) false positive rates (FPRs) in traditional approaches are greatly inflated, questioning the validity of "countless published fMRI studies"; 2) nonparametric methods produce valid, but slightly conservative, FPRs; 3) a common flawed assumption is that the spatial autocorrelation function (ACF) of FMRI noise is Gaussian-shaped; and 4) a 15-year-old bug in AFNI's 3dClustSim significantly contributed to producing "particularly high" FPRs compared to other software. We repeated simulations from [1] (Beijing-Zang data [2], see [3]), and comment on each point briefly.

Cite

@article{arxiv.1702.04846,
  title  = {FMRI Clustering and False Positive Rates},
  author = {Robert W. Cox and Gang Chen and Daniel R. Glen and Richard C. Reynolds and Paul A. Taylor},
  journal= {arXiv preprint arXiv:1702.04846},
  year   = {2022}
}

Comments

3 pages, 1 figure. A Letter accepted in PNAS

R2 v1 2026-06-22T18:19:51.675Z