Hi Renat > Hi everybody! > > We are trying to find age effect interaction between two groups of > subjects – roughly speaking > > find regions where slope of BOLD signal vs age in one group is > significantly different of that of > > the other group. What fMRI model under “Basic models” should I use? I > don’t see anything > > there that seems to do the job.However, there is a PET model, > Compare-populations: 1 scan/subject (AnCova) seems to do what > > we want, but if we use our fMRI data we get the following error: > The PET model should work, even for fMRI data .. Anyway, you can try a multiple regression analysis where you enter the group and the age leading to 3 or 4 cov (depending how you code the group), then you can compare the regression on the age of one group against the other :-) Best Cyril