 |
 |
Dear Sara / mailing list Sorry for the delay in getting back to you. So to summarise... you've done a PEB analysis and you find all the between-group effects are small (the "MAP (full)" part of the figure) and so they are removed during Bayesian model reduction (the "MAP (reduced)" part of the figure). That violates your prior belief that there was a between-groups effect. Please could you summarise your complete experimental design and analysis pipeline, including the regressors you included in your PEB design matrix? Otherwise it's a bit hard to know where to start in helping you unpack this result. Best Peter