 |
 |
Hi Tali, If you try taking that one subject alone and doing post-hoc, do the parameters seem similar with the full model? (Back up any DCM files as they'll be over-written.) And for the group and individual post-hoc analyses, does the full model seem to be fitting well? (spm_dcm_fmri_check.m) Best, Peter. -----Original Message----- From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Tali Bitan Sent: 14 July 2014 22:48 To: [log in to unmask] Subject: [SPM] Parameter estimates in DCM post-hoc Dear DCM experts When we run DCM post-hoc on a group of participants, we get the parameter estimates for the winning model in DCM_opt for each individual. However, when we specify a DCM model with the same structure of DCM_opt and estimate it in a conventional manner for the same individual we get entirely different parameter estimates. What is the reason for that? Which one is more reliable? Thanks a lot Tali Bitan