Dear DCM experts,
There has been some discussion on the list about whether to adjust for
effects of interest when extracting VOI time series. I have tried DCM with
and without adjustment and found I got more significant effects without.
This is unexpected, since adjustment should remove residual movement effects
and improve signal to noise, so I want to check I am doing it correctly.
I adjust using an F contrast covering all experimental conditions, missing
out the movement parameters and the constant. The contrast has a row for
each condition included, with a 1 over each individual condition.
I realize task-correlated movement may make adjustment a bad idea, but I
don't think this is a problem for this data set.
Thank you for your advice,
Paul Wright
|