In our fMRI study, we have an event related design with a painful stimulus
We run this on 10 subjects (experiment run twice on each subject). We
wanted to run a RFX analysis for the group
I modeled the two identical sessions on a single subject basis (using temp
and disp derivatives) and created 3 contrast images for each subject (one
for the main effect, one for the temp derivative and one for the disp
derivative. To do so I used t-contrasts:
Main effect: 1 0 0 1 0 0.
Temp derivative: 0 1 0 0 1 0
Disp derivative: 0 0 1 0 0 1
Is that correct?
The 10 x 3 contrast images obtained from the 1st level analysis were then
put on the 2nd level. I used full factorial with one factor and 3 levels
and entered 3 cells containing each of the 10 sets of contrast obtained
from 1st level.
In the contrast manager I used F contrast: [1 0 0; 0 1 0; 0 0 1]
Is this correct?
Now, I want to use VAS scores obtained for each subject to check if there
is any correlation. Can I use Covariates in the same Full Factorial design
described above?
Many thanks
Annie
|