Dear all
I did a group seed-based functional connectivity from a DMN (the PCC) roi.
i found the classical set of correlating (contrast +1, MTL, AG, mvPFC) and anti-correlating (contrast -1, action-related areas) voxels. so ok!.
Then, I added a variable of interest to my group GLM and I want to test its pos & neg correlation pattern.
I found several voxel that are positively correlated with this variable that are located in those regions that are anti-correlated in the group.
foreseeing a reviewer question, I would like to understand if I can assess if:
- those voxels are really positively correlated in those subjects that have higher values of this variable (inverting the conventional pattern of anti-correlation).
OR
- the higher is the variable, the more positive is the correlation or, simply put, they still were anti-correlated, but such anti-correlation is reduced.
thanks in advance
Alberto
|