help
486 Subscribers 
help > Testing the effect of a confound variable
Showing 1-1 of 1 posts
May 8, 2019 07:05 PM | Steven Meisler - Harvard University / MIT
Testing the effect of a confound variable
Dear all,
I am attempting to analyze the impact of RETROICOR physiological noise regression on default-mode network functional connectivity. I have created the regressor time-series for RETROICOR and I know how to include them as first-level covariates and as part of denoising. However, I do not know how to properly set up the project to compare pre vs. post treatment, if such an analysis is possible within a single project file. Is there a way to see first and second level results as a function of what preprocessing methods are included? If not, would a valid workaround be to use each subject's data twice and introduce a second level covariate to denote RETROICOR vs no RETROICOR?
Thank you in advance,
Steven
I am attempting to analyze the impact of RETROICOR physiological noise regression on default-mode network functional connectivity. I have created the regressor time-series for RETROICOR and I know how to include them as first-level covariates and as part of denoising. However, I do not know how to properly set up the project to compare pre vs. post treatment, if such an analysis is possible within a single project file. Is there a way to see first and second level results as a function of what preprocessing methods are included? If not, would a valid workaround be to use each subject's data twice and introduce a second level covariate to denote RETROICOR vs no RETROICOR?
Thank you in advance,
Steven