p-value of 3dttest++ in SUMA

AFNI Message Board
1

Hello

I used the following command for group analysis on a result of surface-based searchlight decoding [1].

3dttest++ -singletoneA 0.5 -setB 01.niml.dset 02.niml.dset 03.niml.dset … 30.niml.dset

Then, I tried to visualize the result using SUMA; however, the p-value is not shown (N/A) as in the figure below.

Why can’t I see the p-values of corresponding t-values? How can I interpret these t-values without p-values?
I know that there is a step after 3dttest++ for multiple comparisons correction using slow_surf_clustsim.py that I can get some p-values, but I was wondering whether it is possible to get uncorrected p-values of result from 3dttest++ or not?

[1] Demo: fMRI surface-based searchlights with LDA classifier

suma.png
suma.png1054×500 150 KB

2

Hello,

From the image, it looks like you are thresholding on the beta weight. Try thresholding on index 1, which should be the t-stat or z-score.

  • rick