Dear Drs. Adeel Razi and Peter Zeidman (and SPM/DCM experts),

I am trying to perform large-scale DCM analysis for resting-state fMRI data. I have 9 regions of interest - defined based on MNI coordinates from literature.

I am interested in comparing the connectivity between these 9 regions between two groups: B and G. So I ran following steps (just to test, I used two subjects per group) using PEB wiki guidelines:

-------------------- Steps I ran:--------------------

% Two Groups: B and G

% DCM_DMN_B1.mat: Full model for subject 1 for group B

% DCM_DMN_B2.mat: Full model for subject 2 for group B

% DCM_DMN_G1.mat: Full model for subject 1 for group G

% DCM_DMN_G2.mat: Full model for subject 2 for group G

GCM = {'DCM_DMN_B1.mat';'DCM_DMN_B2.mat';'DCM_DMN_G1.mat';'DCM_DMN_G2.mat'};

GCM(:,1) = spm_dcm_fit(GCM(:,1));

GCM = spm_dcm_peb_fit(GCM);

save('GCM_example.mat','GCM');

% Specify PEB model settings

N = 4; % number of subjects

M = struct();

M.alpha = 1;

M.beta = 16;

M.hE = 0;

M.hC = 1/16;

M.Q = 'single';

% Specify design matrix for N subjects

M.X = ones(N,1);

% Choose field

field = {'A'};

% Estimate model

PEB = spm_dcm_peb(GCM,M,field);

save('PEB_example.mat','PEB');

BMA = spm_dcm_peb_bmc(PEB(1));

spm_dcm_peb_review(BMA,GCM)

------------------------------------------------------------------------------

This runs successfully, but my output is showing a blank connectivity matrix (no connectivity between any pair of nodes among 9), even at no threshold. For your convenience, here I am attaching one *.mat file (DCM_DMN_B1.mat) (dropbox link: https://www.dropbox.com/s/u8b7009lz86l2aj/DCM_DMN_B1.mat?dl=0) and final output screen shot. Also, its showing that there is one covariate, I am not sure what is that, as I didn't define any covariate in above steps.

Could you please confirm if the above steps are correct and help me in figuring out what I am doing wrong here while comparing two groups: B and G? Also, any documentation to interpret the final results will be really useful.

Thank you so much !

Best,
Sahil