Happy new year to Bob and AFNI Team!
I was trying to play with the new “ETAC” option in 3dttest++ and noticed an issue that I don’t understand.
I simply added “-ETAC” in a command line of regular 3dttest++ without any other options of ETAC_blur or ETAC_opt, requesting the ETAC procedure to be computed by default. In the outputs, I can see “significant” (p<0.001/voxel with a cluster size determined by regular 3dClustSim) activations in the “main t-test” file but all the TT..default.ETACmask files are empty (all voxel values are zero) indicating no significant activation at all. Since my intuition about ETAC is sort of to combine significant activations at different threshold levels, so any significant activation seen in the “main t-test” should warrant a non-empty ETAC result. Am I right? Did I do anything wrong?