Hi, I've tried this with some artificial data here and it all worked fine. So I don't think the problem is in your matrix or contrast files. Have you checked your input images/masks? All the best, Mark On 28 Dec 2010, at 14:17, SUBSCRIBE FSL Patricia Pires wrote: > Hello Mark, > > thanks for your quick reply. > > I sent the following command for randomise: > > randomise -i all_FA_skeletonised.nii -m mean_FA_skeleton_mask.nii -o resultados_TBSS -d matriz.mat -t matriz.con -n 5000 --T2 -V > > The study has 5 groups and I put in order the data according the order of the groups in origdata. Also I include covariates (age) again in the same order of origdata. > > The matrix and contrast files are attached. > > Thaks very much, > > Patricia. > > > <matriz.con><matriz.mat><matriz.mat>