Dear Andre, No this is not possible, you can only compare models with the same data. If you have a missing ROI due to a brain lesion in some subjects then there are ways round this - extracting a noise ROI from the lesion site in those subjects or using the new DCM.options.hidden option. Best, Peter. -----Original Message----- From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of André Schmidt Sent: 23 July 2014 06:41 To: [log in to unmask] Subject: [SPM] BMS Dear list, Is it valid to compare DCMs (or families) with different numbers of VOIs? I've 6 models, three with 6 and three with 5 nodes - I then applied BMS with a family file considering the different voi's. Can I first use BMS for family inference (two families with different numbers of voi's) and then subsequently for the models only in the winning family? Many thanks for your help Andre