Hi All,
I'm contemplating the same question, and I'm having a hard time reconciling this response with what's said in the FEAT FAQ:
"...Orthogonalisation should be used to prevent additional EVs changing the parameter estimates (betas) associated with previous EVs. For example, when adding a confound like a motion parameter it is desirable to orthogonalise the confounds with respect to the EVs of interest. This prevents the power in the activation signals being shared between the EVs of interest and the additional confound..."
It seems like the FAQ documentation explicitly recommends orthogonalizing confounds with respect to EVs of interest, while the reply to the original poster's question says that doing so does nothing. Sorry if I'm being dense and missing something; could someone clarify?
Thanks!
-Dan
|