Yes sure you can use randomise to compare different spatial histograms. But make sure you divide all voxels of fdt_paths by the value stored in “waytotal”.
> On 5 Aug 2017, at 21:22, Francesco Sammartino <[log in to unmask]> wrote:
> I want to fit a linear model to each voxels of individual seed to all brain connectivity maps (fdt_paths) . I have the same subject analyzed with different seed sizes (it is a simulation study where we are looking at the different white matter connectivity associated with a series of different electrical stimulation parameters).
> Which would be the best model to apply? Linear model with multiple comparison correction through randomise
> Thanks a lot
> Francesco Sammartino