Hi Experts,
I had a quick question on performing group-level FEAT analysis on seed-based connectivity estimates.
The subject-level analysis was performed using FEAT (on whole brain using 1 seed region time course and WM, CSF time course as nuisance variables) and the necessary copes were generated.
I would like to perform group analysis to compare the strength of functional connectivity between 2 groups and I was wondering if I could use the absolute value of the cope estimates (discarding the sign, using fslmaths) to quantify just the strength of dependence. If this is acceptable, would I still be using the original varcope estimates or would that change? It should be noted that the absolute of cope values are not normally distributed, however I am assuming that this can still be submitted to the FEAT analysis since it is the error term of the regression that should be normally distributed. Please correct me if I am wrong.
Any inputs will be greatly appreciated,
Raj
|