I am analyzing an event related design, and would like to model the time
derivatives for the hrf. It seems the most appropriate way to do this is to
use the F-Contrast.
For example (1 run, 2 event-type study)
1 0 0 0 0 0
0 1 0 0 0 0
0 0 1 0 0 0
0 0 0 -1 0 0
0 0 0 0 -1 0
0 0 0 0 0 -1
When I get my composites, it seems that I get a map with acivations
representing any difference between the groups (t squared). How can one pull
apart the (greater) activation for event-type 1 separately from event-type
2?
Also, would one set up a contrast such as
1 0 0 0 0 0
0 1 0 0 0 0
0 0 1 0 0 0
to get the event related activity for the first event type? And if so, is
there a way to view the activations and deactivations separately?
Lastly, is there a way to combine the time derivatives with the hrf for a
t-test? I read in the archives that:
1 1 1 0 0 0
was not appropriate.
Thanks
-Wil
|