 |
 |
Dear FSLers, I have a question concerning the corrp images obtained with the TFCE option in randomise. I performed a vertex analysis as explained on the FSL wiki (the "new" version), and now have corrected 1-P images, derived from TFCE in randomise. I noticed that in these images, which are supposed to only contain voxels that meet correction criterion, some voxels have a value below 0.95. Does it mean that these voxels meet the cluster correction criterion (but are not 'individually' significant at 0.05) or does it mean that they are not significant (and thus the corrp image is in fact NOT thresholded to only include corrected-significant voxels)? I guess from the randomize page I thought that only voxels that pass correction are included in the corrp image. Could you please clarify that for me? Thanks! Evan