Hi,
I found the examples of the design matrices and contrasts on the FEAT website (http://www.fmrib.ox.ac.uk/fsl/feat5/detail.html). Our experiment has 19 subjects scanned over 17 consecutive days. However, the middle 9 days are our "treatment" while days 1-4 and 14-17 are our "control." From the example on the FEAT page, the multi-subject multi-session paradigm is most applicable to use, but after we combine across subjects, we then want to run essentially a t-test to compare the control days to the treatment days. My problem is that the multi-session/multi-subject description shows two different sets of design matrices and contrasts to do all three levels. However, I can only plug in one matrix and contrast into dual regression. How should I go about doing this?
Thanks,
Brandon
|