thank you so much, your answer really clarified a lot of things: you
opened an entire new world to me!
I still have one more doubt, when you say:
> Rather than assessing the modulatory influence of all
> your regressors on
> *all* your connections, a better approach might be to make 2
> models, one in
> which the main effect regressors modulates the connections between
> yourregions of interest, and another in which the interaction
> regressor does
> so. You can then compare these models using Bayesian model
> comparison, to
> address your precise question.
If you perform a Bayesian comparison between, say, model 1 (modulation
by the main effect) and model 2 (modulation by the interaction), what
kind of answer would this comparison provide?