PAPER: An Argument For Proper Multiple Comparisons Correction
It has been a long road, but our multiple comparisons paper including the salmon has been published. See below for more details, including the abstract and a link to the download page of the journal. If you have any questions or comments please post them below or send me an email directly.
Neural Correlates of Interspecies Perspective Taking in the Post-Mortem Atlantic Salmon: An Argument For Proper Multiple Comparisons Correction
Craig M. Bennett(1), Abigail A. Baird(2), Michael B. Miller(1) and George L. Wolford(3)
1)Department of Psychology, University of California at Santa Barbara, Santa Barbara, CA 93106
2)Department of Psychology, Blodgett Hall, Vassar College, Poughkeepsie, NY 12604
3)Department of Psychological and Brain Sciences, Moore Hall, Dartmouth College, Hanover, NH 03755
Journal of Serendipitous and Unexpected Results, 2010. 1(1):1-5
Early Access: Oct 20, 2010
With the extreme dimensionality of functional neuroimaging data comes extreme risk for false positives. Across the 130,000 voxels in a typical fMRI volume the probability of at least one false positive is almost certain. Proper correction for multiple comparisons should be completed during the analysis of these datasets, but is often ignored by investigators. To highlight the danger of this practice we completed an fMRI scanning session with a post-mortem Atlantic Salmon as the subject. The salmon was shown the same social perspective-taking task that was later administered to a group of human subjects. Statistics that were uncorrected for multiple comparisons showed active voxel clusters in the salmon’s brain cavity and spinal column. Statistics controlling for the family-wise error rate (FWER) and false discovery rate (FDR) both indicated that no active voxels were present, even at relaxed statistical thresholds. We argue that relying on standard statistical thresholds (p < 0.001) and low minimum cluster sizes (k > 8) is an ineffective control for multiple comparisons. We further argue that the vast majority of fMRI studies should be utilizing proper multiple comparisons correction as standard practice when thresholding their data.
See the JSUR early access page to download the article and supplementary material.