Dear DCM experts,
Now I calculated fMRI data using DCM, but we are having problems.
We are sorry for basic questions, but we hope someone helps us.
(1)We selected five regions that were detected by conventional GLM
analyses: prefrontal cortex (PFC),
bilateral amygdala (R.AMY, L.AMY), and bilateral FFA(R.FFA, L.FFA).
Then we calculated DCM with following parameters;
Intrinsic connection are:
R.FFA - o o o x
L.FFA o - o o x
R.AMY o o - o o
L.AMY o o o - o
PFC x x x x -
the effect of fearful face are:
R.FFA o x x x x
L.FFA o x x x x
R.AMY o x x x x
L.AMY o x x x x
PFC x x x x x
We did not set any mopdulations on inter-area connectivity.
After estimate the model, we got those results.
DCM.A =
-1.0000 -0.0000 0.0000 0.0001 0
-0.0000 -1.0000 -0.0000 -0.0000 0
0.0001 -0.0001 -1.0000 0.0002 0
0.0000 -0.0000 0.0000 -1.0000 0
0 0 0 0 -1.0000
We calculated this model for all the subjects, and we found all DSM.A from
PFC to R/L AMY are zero.
Is there mathematically wrong parameters in our model?
Does DCM model require modulation of inter-area connectivity?
(2)My second question is about spm_dcm_voi and spm_dcm_U
(ftp://ftp.fil.ion.ucl.ac.uk/spm/spm2_updates/).
Our process of analysis were as follows:
First we generated five ROIs in the subject1 directory, and estimated a DCM
model (DCM_1.mat).
Second we copied all the five ROIs and DCM_1.mat to the subject2 directory,
and run the next script.
We have only one session, and length(SPM.Sess.U) is 9.
spm_dcm_voi('1/DCM_1.mat',
{'1/VOI_RFFA.mat','1/VOI_LFFA.mat','1/VOI_RAMY.mat','1/VOI_LAMY.mat','1/VOI_
PFC.mat',})
spm_dcm_U ('1/DCM_1.mat','SPM.mat',1,[1:9])
Then, we estimated the new DCM_1.mat.
The calculateion finished without error log, but we found the DCM.A by this
batch and by manually
assigned is different.
Could me tell me what is wrong with our batch.
Thank you in advance,
Shoichi Ugai
[log in to unmask]
Department of system engineering
Tokyo Metropolitan University
Tokyo, JAPAN
|