Hi Everyone,
Thank you all for your very helpful feedback. So, just to summarize on the
consensus, and please correct me if I am wrong:
ICA+FIX denoising presumably already removes the WM and CSF related
variance from the signal, in a non-aggressive way (which is one benefit of
the FIXed data) -
Hi all
I think Greg was making the valid point that: if you want to regress the full
space of A and B out of your data, it is not correct to regress out A and then
afterwards regress out B, unless A and B are orthogonal. To do it correctly
you either need to combine [A B] into a single model
That’s a reasonable summary, though I wouldn’t worry about being “too
aggressive” unless your WM and CSF masks are including grey matter signal.
Also, making sure that you regress the appropriate residual time course out of
the data is not a difference between aggressive and non-aggressive
1) My point was that it isn’t the same because of the aggressive vs non
aggressive difference. I wasn’t making an argument for using WM and CSF
instead of ICA+FIX or in addition to it. It doesn’t really add anything
because WM and CSF don’t contain clean regressors of any noise signal.
3) I
