Bayesian Group Analysis for within-subject design


I have a 5 condition repeated-measures design. I am interested in the effect of condition on activation levels in 21 regions of interests (as determined by previous studies). The contrasts of interest between relevant conditions were computed at the first level.

Is it possible to use to determine the likelihood that the ROIs were activated in response to each condition (and for contrasts of interests)? If so, what should the “some-x” variable be?

Thanks again,