search

poldracklab / ds003-post-fMRIPrep-analysis

An exemplary task analysis workflow to run on OpenNeuro's ds000003 data, after preprocessing with fMRIPrep
Apache License 2.0
28 stars 8 forks source link

Doing high-pass filtering twice? #8

Open tientong98 opened 3 years ago

tientong98 commented 3 years ago

Hi,

I thought that high-pass filtering was already achieved by including the DCT-basis regressors in the GLM model. However, you also have an additional highpass filtering step. Can you please explain why we need to do both?

Best,

Tien

effigies commented 3 years ago

I agree, this looks like it will duplicate regressors, unnecessarily. Assuming that uses a cosine regressor, rather than a polynomial, then there should be no effect, as any drift components will be split between the two, and discarded in either case.

Anyway, we should clean this up.