Hi,

I have one question about using the batch.Denoising.confounds specifications.  I am trying to change the confound list so that it includes realigment/scrubbing/conditions (as is default) but does not include White Matter or CSF, and instead includes a brain-extracted structural image (input as an ROI).  For some reason when I input values in batch.Denoising.confounds and then execute Setup.done=1 in the script, some the specified confound information seems to be overwritten. Specifically, CONN_x.Preproc.confounds.names is correct, but CONN_x.Preproc.confounds.types is not correct based on my inputs (it seems to correspond to what would be correct if I left the confounds alone/on the defaults with White Matter / CSF).  Additionally, when I leave CONN_x.Preproc.confounds.dimensions empty (so that it will fill in with values based on the max dimensions of the effect), it remains empty.  

This results in a problem when I load the .mat file into CONN.  Specifically, the histogram of voxel-to-voxel r values does not show up, and the GUI instructs me to redo Setup.done (by pressing "done" in the Setup tab of the gui).  When this new "done" finishes importing the functional/ROI data, the new CONN_x structure seems to be formatted correctly.  Is there something I am doing wrong in specifying confound information that might be leading to this problem?

Thank you in advance for any help on this question!

Ruth