Hello SPMers,
I have two questions regarding adding a motion censoring regressor to the GLM for task-based fMRI data.
1). If, for example, there are five time points to be censored, should these be entered into the GLM as a single regressor or as five separate regressors? I have seen both ways mentioned in the literature and the two yield different results in my data.
2). Do these regressors need to be added to the GLM as covariates of no-interest by manually removing them from the SPM.xX.iC field and placing them in the SPM.xX.iG field before model estimation?
Cheers,
Tim
Timothy D. Ely
Supv. Research Specialist
Dept. of Psychiatry and Behavioral Sciences
Emory University School of Medicine
101 Woodruff Circle
Suite 4000 WMRB – room 4304
Atlanta, GA 30322
404-727-8268 phone
[log in to unmask] email
|