Dear Danai,
Although DCM for ERP and DCM for fMRI are conceptually similar and
share the same Bayesian optimization framework, their data structures
are not the same because they use quite different generative models.
So the DCM-fMRI structure is probably a bad guide for interpreting the
DCM-ERP structure.
What you are looking for is indeed in DCM.Ep, just what is stored is
the log of the parameter rather than the actual parameter estimate. If
you do something like:
exp(DCM.Ep.A{1})
or
exp(DCM.Ep.B{1})
you will see the same values that are displayed in the reviewing GUI.
In general a good way to find out how something is represented in the
structure is to look at spm_dcm_erp_results, find the place whether
the thing you are interested in is plotted and go from there.
Best,
Vladimir
On Mon, Mar 23, 2009 at 9:43 AM, <Danai> <Dima>
<[log in to unmask]> wrote:
> Dear all,
>
> After you have estimated a model, the actual number of the connections and the modulations (not propabilities) are stored in the DCM. Ep structure?
> In the fMRI data it is under the DCM.A, which is not the case here.
>
> Thanks a lot, for any response.
>
> Best,
> Danai
>
|