Hi Gang,

Yes it's a group analysis with 8 blocks x 14 subjects. I wanted to test if a quadratic function would fit the data. In the 1st post I only put the 8 blocks of the 1st subject and the last block of the last subject, but there are really 8 blocks x 14 subjects.

I want to see if the quadratic effect I get is due or influenced by something esle, thus adding a covariate in the model. In the end I want to determine if I can say that this covariate did not have an impact on the results.

Which program should I use then , and what not 3dRegAna...? I though 3dRegAna was espacillay for regression at the group level!

<<You analyzed each block as a separate regressor at individual level? Was the behavioral measure acquired at each block level?

Yes on both questions



Pat