Hi - yes, that's fine - in fact if you look inside the FSLNets code you'll see that's how we do inference on network matrices. randomise doesn't require the matrices to be z-transformed for validity, but in theory it is preferable, yes - again you can see that this is done in FSLNets. Cheers. On 7 Jan 2014, at 00:52, Paul Robinson <[log in to unmask]> wrote: > Dear FSL-ers, > > I have correlation matrices (in .nii format) for two groups of subjects, and was wondering about the legitimacy running randomise on them. I want to know where my groups differ with respect to these correlations, so is this a reasonable way to try to answer this question? Presumably I should also z-transform these correlations? > > Any advice is welcome. > > Many thanks, > Paul > --------------------------------------------------------------------------- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve --------------------------------------------------------------------------- Stop the cultural destruction of Tibet