Hi all,
I would like to perform a first level analysis where one of the regressors of interest is a function of space (for example, to use a different HRF for different locations in the brain). Ideally, I'd precalculate the regressors outside of FEAT and supply 4D Nifti file with dimensions matching the data, containing the regressors for each point. Poking around film_gls, it seems like the basic machinery for doing this is supported at the second level, where voxel dependent ev's are used for outlier deweighting, etc., but I'm not sure if it's possible to convince film to use this for a first level regressor. Am I missing something, or does this require mucking around in the code?
Blaise
|