Dear Michaël

If the GLM fails to identify a relationship between your contrast and the activity in these brain regions, then you are unlikely to find a modulation of the connectivity in DCM. If you don’t find a difference in modulation, you could model each main effect as separate driving inputs and look at the average connectivity over both conditions (matrix A).

 

Best

Peter

 

 

Good morning Peter,

 

I would like to describe you a situation and ask you what do you think about it.

I am interest to test difference between condition A and B in some regions which belong to a known brain network in literature.

These regions showed activation for A and B separately. However, the classical T-test comparison highlighted no difference of activation inside of these regions.

 

I was thinking to study the connectives computed by DCM in this network for A and B and then compare them.    

 

Do you think that it is useless to perform DCM analysis because of the absence of difference in classical fMRI comparison or it could be interesting to test it ?

 

Michaël Mouthon, PhD