Hi Everyone, I have a question about running multiple regression at the second level in spm8. I have two individual difference variables (variable A and variable B). I want to run a whole-brain regression on a contrast, looking at neural activation that varies as a function of variable B while controlling for the effect of variable A. (Basically like a hierarchical regression). If I select the multiple regression design in my second level model, put variable A as my first covariate and variable B as my second covariate and then put as my contrast 0 0 1, does this achieve what I want? Or is this more like simultaneous multiple regression? If what I'm suggesting is incorrect, any suggestions on how to perform hierarchical regression in spm8 at the second level? Thanks! Meghan -- Meghan Meyer, Ph.D. Candidate Social Cognitive Neuroscience Lab UCLA, Psychology