Hi SPMers
I have a recall condition which I've modeled parametrically so that in
the design matrix I have three columns corresponding to the 0th, 1st
and 2nd order. Am I correct in thinking that the contrast '1 0 0' will
find activity corresponding to the recall condition regardless of the
size of the parametric modulator?
One thing I'm really interested in would be to find areas that are
significantly active during the recall condition but are not modulated
linearly or quadratically. I'm not quite sure how to go about looking
at this though. Should I:
a. Look at the contrast '2 -1 -1'
b. Mask the contrast '1 0 0' exclusively with '0 1 0' and '0 0 1'
c. The contrast '1 0 0' already looks at this
d. something else
Thanks for your help
Tom Couch
|