 |
 |
Dear James,
It is good that you extracted VOIs from individual local maxima. Further than that, from the description you’ve given, it’s not possible
to say what’s wrong with your model. Perhaps you could give more detail on every aspect of your model, including the number of regions, the structure, the task design and details of the effects of interest F-contrast you used, in the hope that this reveals
something which is obviously amiss.
All the best,
Peter
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]
On Behalf Of james pardon
Sent: 16 April 2015 17:57
To: [log in to unmask]
Subject: [SPM] DCM estimation fails
Dear DCM experts,
I'm following DCM manual to perform a similar analysis on my own subjects. In order to extract time series, I use local maximas for each subject as the centre of 8-mm sphere. When I start to estimate my DCM, matlab takes me into a keyboard
debugging mode, where as far as I can tell is the EM step of the nonlinear system identification. Based on previous posts on the mailing list, I re-checked my subjects to make sure that exogenous inputs (modulatory and input matrices) are actually showing
effect in the standard GLM analysis, but the spm_dcm_estimate still fails at the same stage. I'm wondering what else could go wrong here?
Thanks in advance for any inputs.
Best wishes,
James