Hi, I don't know DCM terribly well, so let me jump on the occasion ask a minor question to see if I understand things right. On Thu, Mar 17, 2011 at 03:11:33PM +0100, Christophe Phillips wrote: > 1. Does it make sense to have a model with 4 regions, in which 2 regions > within each of 2 pairs are interconnected, but there are no connections > across pairs (ie, two “isolated” subnetworks; eg, with regions E1<->E2 > F1<->F2, DCM.a = [1 1 0 0; 1 1 0 0; 0 0 1 1; 0 0 1 1])? > Yes it does. > You're "simply" modelling 2 parrallel and independent processes a single > model. In which case, doesn't the likelihood factor in separate independent terms? I have the impression that it amounts to solving the DCM twice on two separate systems. I don't see this as posing a problem at all though for analysis and interpretation. Cheers, Gael