 |
 |
Hi carlos, Glad it makes sense! It's best if you use just one model with multiple covariates. Eugene On May 22, 2009 7:57 PM, "Carlos Faraco" <[log in to unmask]> wrote: Eugene, I see, that makes sense now! I think what I need is the second-level analysis. Just a quick question about this though. Besides the behavioral measures acquired in the scanner, we also have some assessment data, e.g. WMS-III working memory index scores. If I want to see how the WMS-III correlates with the fMRI data is OK to specify these as another EV in the setup or should this EV only be setup separately? In other words should I have one setup where there is the group EV and the response and/or RT data, and a separate setup with group EV and WMS-III scores, or would it do just as well to have one setup which orthoganalizes the WMS-III wrt the RT and response scores? Thanks, Carlos