Hi

We are having a strange outcome at the 2nd level. We have n = 104, preprocessed the data in a standard way using the Phase maps. 
Denoise was done using aCompCor with 5 components for WM, scrubbing was also done.

After all is preprocessed and denoised, if we do a seed to voxel correlation using striatum, pallidum, ACC, OFC and mPFC seeds from the Atlas, we get significant negative correlations with all the Ventricles and Sinuses as you can see in the picture. The positive correlations are in GM, but also WM.

We are trying to figure out where the problem is, the DMN using the PCC also doesnt look great. We checked the WM, CSF masks and they are correct and well normalized and coregistered with the fMRI volumes.

I'm thinking something is going wrong in the denoise step, but we are using everything as standard.

Any help is appreciated.

Thanks

Eduardo