Fig. 11.

Model selection using empirical data: This figure summarises the results of model selection using the empirical fMRI data. The upper left panel shows the log-evidence profile over the models considered (reflecting different combinations of edges among the six nodes analysed). The implicit model posterior (assuming flat priors over models), is shown on the upper right and suggests that we can be over 80% certain that a particular architecture generated these data. The parameter estimates of the connections under the full (left) and selected model (right) are shown in the lower panels. Again, we see that certain connections have been switched off as the parameter estimates are reduced to their prior value of zero. It is these anti-edges that define the architecture we are seeking. This architecture is shown graphically in the next figure.