I am dealing with a bit of a mystery (mystery to me anyway). A specific
brain region comes up in a Miss>Hit T-contrast as hypothesised. So far so
With Marsbar, I extracted the eigenvalues separately for Hits and for Misses
using that specific cluster as ROI. The 'y' eigenvalues for both came out
negative. What I donít get is that the mean value for Hits was higher than for
misses (i.e. less negative). How is that possible when the particular ROI came
out of the Misses>Hits contrast? I would understand not finding a significant
difference, but I assume that the means should be in the right direction (i.e.
higher activation for Misses).
What am I missing?