Hi,
I am trying to extract regression parameters from a second level
analysis. In each first level analysis I ran (to greatly simplify) a
positive (+1; cope1) and negative (-1; cope2) contrast on the variable of
interest (in this case a seed in a resting state analysis). The goal is
to examine "anticorrelated" resting state networks (a la Snyder et al.,
2005) in my data.
I've extracted the regression parameters for each subject using:
fslmaths -i filtered_func_data -m <cluster_mask_zstat> -o outputfile
on the higher level analysis in cope1.feat. When I do the same thing in
cope2.feat, I still get positive values.
Thus, I'm getting positive values for the regression coefficients
regardless of the direction of the contrast. Can I simply flip the sign
for the regression parameters on the negative contrast?
Thanks,
Matt
|