Hello,
I want to do a convolution using flobs for my regressors of interest in the first-level feat analyses. For the higher-level analyses, I will pass up the canonical HRF regression parameter estimates, as suggested on your wiki.
However, I am not sure if I should keep the orthogonalization of the basis functions turned on (as it is by default in feat) in the first-level analyses or if I should turn it off.
Could you explain to me what the best approach would be please?
Thanks a lot for your help,
Marie-Eve
|