Hi Doug,
For some reason, I no longer get an e-mail when you respond to my questions
on the mailinglist, so until today I only saw the first response to this
question (from Bruce).
Anyway, I get your point about smoothing and clusters but I guess showing a
continuous cortical map with vertex values should be ok, right? The map
itself is rather smooth and it's impossible to see the individual vertex in
the figure, so we're basically saying "this region generally has about 0.8%
reduction" (for instance).
Also, how much precision can I assume here? 0.004 mm^2 is not very much.
These changes are apparently very consistent across subjects since most of
the cortex shows significant change. But what if we find very consistent
changes that are on the order of 0.000001 mm², which therefore become
statistically significant? Is that beyond our level of precision here?
Thank you!
yours,
Lars M. Rimol, PhD
Norwegian University of Science and Technology (NTNU)
Trondheim,
Norway