I am a bit of a novice so my apologies if there are any issues with my post.
I am having trouble understanding how to run a regression analysis in SPM 12. I've just run a group-level regression analysis with covariates A, B, and C. I am interested how my covariate C significantly correlates with BOLD signal. There are 4 columns in the design matrix. Column 1 is the mean BOLD activity (it just says MEAN), column 2 is covariate A, column 3 is covariate B, and column 4 if covariate C. I am told the proper contrast is 0 0 0 1. However, wouldn't the correct contrast be 1 0 0 1? Why would I omit the mean BOLD activation?
Alternatively, if I run a one sample T-test with the same three covariates, then contrast 1 0 0 0 would show the mean sample averaged BOLD signal. But if I run contrast 1 0 0 1 doesn't that also show mean sample BOLD signal correlated with that covariate?
My sincere thanks in advance to anyone who can accurately answer my questions.
|