This repository contains the code used to analyze differences in effective brain connectivity associated with neurofeedback training assessed with DCM fMRI.

Please contact at this e-mail address if you have any question: [email protected]. Gabriela Vargas.

1. DCM Matfiles:

• 1st family-inference : model_space.mat and BMS.mat.
• 2nd family-inferernce: model_space.mat and BMS.mat
• Bayesian model average: BMA.m

2. Plots:

• Reproduce bars
• Reproduce matrices

1. Data: fMRI with contrasts
2. Generate masks: Here we use the following functions: generaunMask.m
3. Generate VOI's: .nii files. Here we used the following functions generaunROI.m
4. Build and Generate 1st set of DCM models specifying A and C parameters. Here we use the following functions generaunDCM.m
5. Bayesian model comparison: RFX group and Family comparison (both use the batch and the function 'spm_run_dcm_bms(job)' )
6. Selection of winning model parameter configuration by model evidence
7. Build and Generate 2nd set of DCM specifying B parameter. Here we use the same function generaunDCM.m
8. Bayesian model comparison: RFX group and Family comparison (both use the batch and the function 'spm_run_dcm_bms(job)' )
9. Bayesian model averaging (BMA). Here we use the function BMA.m by the batch.