Hi. We are doing a study in which a treatment is administered that has a certain effect on a variable. We study this effect in two ways: directly measuring it using instruments, and studying the changes in the fMRI data. We want to find voxels in the brain that covary with the change in this effect.

i.e. if variable pre-treatment is 4 and post treatment is 8, the goal is finding voxels where activation doubles. Since the voxel data is a time series, maybe using beta-coefficients from a linear regression might be the way to go, but I don't know an (easy) way to do linear regressions on every voxel of a dataset with the corresponding voxel in another dataset. How would I do this? Thank you.