I've some problems with my fMRI analysis. It's a experiment about pain
processing in rats. The pain is evoked by varying heat stimuli using
Peltier elements on the left and right paws of the animals.
The strange thing is that especially in one animal, I observed a
decrease in BOLD that results in negative t values after GLM analysis.
I tried several software implementations including an own one. SPM
suppresses negeative t values, so I had to use negative contrasts (e.g.
First I thought that a possible cause is motion or maybe a effect of
"stealing" of blood from less active regions into areas with increased
neural activity. But the clusters of negative t-values/decreased BOLD
are very large in comparision to the clusters with increased BOLD.
As I said, that effect is very large in one animal, but also existant
in the scans of some more.
So maybe someone has experienced similar effects or can give me some
hints for further investigation?