I'm trying to use the multivariate bayes analysis implemented in SPM8 to compare neural coding models in my data. I have two questions I hope you can help me with:
1. I'm a bit confused by the terminology for the model priors. In the Friston et al. NeuroImage paper where this method is described, the various priors were listed as spatial, smooth, singular and support, while in the SPM GUI the options are compact, sparse, smooth and support. I can't find a description of these new labels anywhere, so I'm not sure if the priors have simply been renamed or if these represent new models that weren't in the old version. What, for example, does a 'compact' model reflect?
2. Is it possible to compare models across subjects using random effects? I know that the Bayesian model selection facility in DCM allows for this, but I don't know how to do it in Multivariate Bayes analysis. The BMS button in the GUI allows comparison of different models, but presumably that uses fixed effects, and isn't suitable for group analysis. I haven't been able to find any options to change the inference method.
Thanks for your help,