Hi Frédéric, The post-stats options allow the time courses of each component to be compared with the time course of the experiment (as in the 1st level FMRI, task-based) or between subjects. It isn't a full comparison of the strength of the networks at each spatial location as provided by the dual regression. All the best, Anderson On 3 November 2015 at 10:18, Frédéric Briend <[log in to unmask]> wrote: > Hi FSLers, > > I apologize in advance for my elementary question. I would like to compare > groups of subjects in an ICA way: So can you enlighten me about the > difference between the dual regression and the post-stats in MELODIC (where > it is possible to create a model setup with GLM to compare groups) > > Cheers. > > Frederic >