The AFNI team is proud to present a new method for false positive rate control - the Haddock Ad-Hoc (HA-H!) clustering method. In this ad-hoc method, clusters are compared to a haddock. While Gaussian random distributions have often been used, this method instead relies on a “Poisson” distribution (in the French sense). Briefly, if the cluster looks like a haddock, that’s no coincidence, and definitely not a “fluke”. As far as processing goes, you are really better off scaling your data but limited to one-tail tests, of course. We expect this method will allow for “fishing expedition” research to net really big results. Ad hoc fish-based methods have a long history in FMRI [1,2], and we are proud to contribute to this tradition.
- Carp J “The secret lives of experiments: Methods reporting in the fMRI literature.” Neuroimage 63(1):289–300, 2012.
- Bennett et al. “Neural Correlates of Interspecies Perspective Taking in the Post-Mortem Atlantic Salmon: An Argument For Proper Multiple Comparisons Correction” Journal of Serendipitous and Unexpected Results, 2010.
Note, please cite this method by the following:
“HA-H!”, April 1, 2017.
