Dear Matthias,
There may be a slight misunderstanding here. spm_dcm_U does not determine whether or not a particular input is being used by DCM. All this function does is to take a pre-specified DCM (with a given set of inputs) and then replace some of these inputs as specified in the function call by new inputs. This is useful when you are writing a batch script to do multi-subject DCM analyses because you can then specify a single DCM for one subject and reuse the DCM structure for any other subject. Whether or not a particular input is used depends entirely on the (binary) structure of the matrices in DCM.b and DCM.c. An input that is stored in a DCM but is not used by any of these matrices has no effect whatsoever.
Best wishes,
Klaas
________________________________
Von: Matthias Schurz <[log in to unmask]>
An: [log in to unmask]
Gesendet: Dienstag, den 7. Juli 2009, 09:41:59 Uhr
Betreff: [SPM] DCM - question about spm_dcm_U
dear spm experts,
i have a question about the spm_dcm_U function (SPM5).
in my experiment, i present targets and invalid stimuli. the invalid stimuli
are
of no interest for my research question. i wish not to analyze them.
i have specified two conditions for targets and one condition for the
invalid stimuli in my 1st level spm.
i used the spm_dcm_U command and set the "invalids" condition to 0
(spm_dcm_U(DCM_filename, 'SPM.mat', 1, {1, 1, 0}))
is the "invalids" condition now removed from the dcm-analysis? (does dcm now
partition out the "invalids" from the voi-timeseries, or something like that?)
or is the "invalids" condition now simply left unexplained - resulting in
worse explanatory power of such models?
thank you very much in advance!
all the best,
matthias
