Hi, Mit & Hector-
There are a couple separate issues here.
To start, just to be clear, am I correct that you have calculated seedbased correlation maps for 9 subjects, and then do a one-sample t-test of those?
- The one-sample t-test will output two things: 1) the mean effect at each location (so the mean of your seedbased correlation maps, per voxel), and then 2) an associated t-statistic (basically the ratio of the mean to the uncertainty at each voxel).
- In most neuroimaging cases, on will expect some high correlations around the area of the seed itself, and then possibly other regions with high correlation. The latter are typically more of interest, because those tell a bit about relations of different parts of the brain.
- Both the effect and the statistic contain useful information and should be displayed. This is something we promote a lot and the AFNI GUI is built to facilitate this. It is described in this thread:
Basic question about value in olay panel
Basically, you can put the t-test effect estimate as the "overlay" volume, and then set the t-test statistic as the "threshold" volume. - ... and even better, you can display your results with transparent thresholding, to show more of your data, which is also described in that thread, as well as in this paper (jumping to Fig 1 there, and emulating the lower panel image there is highly recommended).
--pt