Thank you for your direction.
To clarify, the threshzstat image generated from each component's probability map are each comsidered statistically independant of all the other IC maps during this mixture model-thresholding process. Right?
To then say afterward, what are the chances of all of the thresholded maps of an individual taken together have clusters or voxels considered signficant when they are not, there needs to be a second over test considered all of the non-noise IC maps to correct for multiple comparisons. Right?
Thank you much,
Varina
|