Hello FSLers,
I have a basic block design: A B C A B C. Each A, B and C block is separated
by an implicit baseline condition.
I would like to extract percent signal change, using featquery, from a region of
interest for each subject, such that I have the percent signal change of "A
versus baseline" and "B versus baseline" for each person.
My question regards which stats images of interest to use as my featquery
input: COPE or PE?
As I understand it, a COPE with a contrast of [1 0 0] asks the question "when
is the first EV's parameter estimate (PE) significantly greater than zero?" But
how does this COPE and PE distinction translate to % sig change?
The reason I ask is because the resulting % signal change I get from the "A vs
baseline" condition is different, depending on whether I use the COPE or PE as
my input.
Many thanks in advance for the help,
Lara
|