> Why don't you want to get ts with variance related somehow to your task(s) and regress out effects of no interest?
functional connectivity analyisis is usually done this way in order to assure that the connectivity between regions isn't just driven by task-related activation patterns. What we are interested in is spontaneous synchronized brain activity.
>> Yes I think for your purposes it would be the latter (zeros(1:11) 1).
>> As Iwo pointed out, the "adjust data" option removes the null space of your F-contrast, i.e. the zeros, from your timeseries. See lines 143-153 in spm_regions.
I've actually tried that but it gives me very odd results. The two regions I'm looking at are almost perfectly correlated (0.99) if I specify the F-Contrast like that. Another odd thing is that for both regions the eigenvariate captures roughly 99% of the variance. If I specify the contrast as (ones(1:11) 0) the eigenvariates capture a more reasonable 60% of the variance and the correlation between the time-courses of both regions is also in a much more plausible range. It really looks to me as if (ones(1:11) 0) were the right way to go but I would feel much more confident if I knew why.
|