Hi Alfonso,

Just wanted to check that I'm understanding the below section of the conn_batchexample_attentiondataset_Denoising.m script. The help at the top mentions the removal of White matter/CSF. The inclusion of 'Grey Matter' in the confounds list with 1 dimension and 0 order derivatives will also perform something akin to global signal regression (more precisely mean grey matter signal regression); is that right? In which case I should remove GM from the confounds list to just use your recommended aCompCor (no global signal regression) method?

Thanks,
Brad

---->

batch.Denoising.confounds.names=... % Effects to be included as confounds (cell array of effect names, effect names can be first-level covariate names, condition names, or noise ROI names)
{'Grey Matter','White Matter','CSF','sessions','Effect of Fix','Effect of Stat','Effect of Natt','Effect of Att'};
batch.Denoising.confounds.dimensions=... % dimensionality of each effect listed above (cell array of values, leave empty a particular value to set to the default value -maximum dimensions of the corresponding effect-)
{1, 5, 5, [], [],[],[],[]};
batch.Denoising.confounds.deriv=... % derivatives order of each effect listed above (cell array of values, leave empty a particular value to set to the default value)
{0, 0, 0, 1, 0, 0, 0, 0};