Hi all,
I am trying to find the optimal model using spm_dcm_post_hoc. And I have some questions to ask:
1)Optimizing for a group: I don't know whether it is right or not to put file names of all subjects into a cell 'P' and input this cell to spm_dcm_post_hoc for group optimization.
2)Which parameters are removed? The sentences like "8 out of 16 free parameters removed " are outputted in the command window in the process of optimization. But how to find out which parameters are removed? Or what're the parameters of the model with the greatest posterior probability among the 256 models outputted in the Graphics?
3)Optimizing for a group in which each subject has multi sessions: How to arrange the input data to spm_dcm_post_hoc if each subject has multi sessions?
4)Whether values in DCM.Ep in DCM_BPA.mat are the estimated connection strengths for the optimal model or not? Or whether the "selected model" referred in spm_dcm_post_hoc is the optimal model or not?
Thank you in advance!
He
|