Hello CONN community,

I am also interested in conducting a seed-to-voxel gPPI analysis with correlation measures because my data has been obtained in a fast event-related experimental design and from previous forum posts the gPPI analysis has been said to be preferred over weighted-GLM analysis in this case. However, I am not interested in directed connectivity from my seed to each voxel but rather to bidirectional connectivity. Has the gPPI analysis been "forced" to calculate bidirectional connectivity when the correlation measure has been chosen as Dr. Alfonso Nieto-Castanon suggested in his post from 2017? Alternatively, would there be a problem if I stick with a weighted-GLM analysis because from what I understand although a gPPI analysis is preferred for fast event-related designs, it does not seem to be mandatory because I came across posts from 2014 in which people were inquiring whether they can use CONN for their data obtained in fast event-related designs and Dr. Alfonso Nieto-Castanon said that it is fine, although at that time gPPI was not yet introduced (or was just introduced).

Thanks a lot!