Print

Print


Dear Ian,

From what you describe, it is not clear to me why the modulatory parameters acting on the self-connections are not estimated.  This is certainly feasible and should be no problem whatsoever.  

One possibility I could think of: is it possible that you have been looking at the wrong file? This could have happened if you were trying to estimate the models directly from the MatLab command line, using spm_dcm_estimate.  In the case, you have to be careful what argument you are passing to this function.  If you pass the name of the DCM file, then the results will be written to that file.  If, however, you are passing the DCM structure, then the parameter estimates will be written to a new DCM file which will contain the current date in its filename.  

Best wishes,
Klaas






________________________________
Von: Ian Ballard <[log in to unmask]>
An: [log in to unmask]
Gesendet: Dienstag, den 16. Februar 2010, 16:18:59 Uhr
Betreff: [SPM] DCM Self-Connections

Hello,
I am conducting a DCM analysis and I am having a problem running models where the modulatory variable (reward level) acts on the self connections.  I have a 3 region model, and I want to test a set of models where all 3 regions' self-connections can be modulated by reward level.  To do this, I set the values of the DCM.b diagonal entries to 1.  However, when I estimate the models, the DCM.B values on the diagonal are 0.  In fact, the DCM.B matrix is identical for models with and without the modulated self-connections.
EG. DCM.B is identical for DCM.b matrices A and B where
A=   [ 1 0 1     and B = [0 0 1
      0 1 1               0 0  1
      0 0 1]              0 0  0]
I know that it is possible to specify self-connections as being modulated by reward (see Lee, Friston, and Horwitz 2006) so I don't understand what is happening.  If it helps, my DCM.A values on the diagonal are -1 for all models.  Any help would be appreciated.
Thank you,
Ian


__________________________________________________
Do You Yahoo!?
Sie sind Spam leid? Yahoo! Mail verfügt über einen herausragenden Schutz gegen Massenmails. 
http://mail.yahoo.com