Hi Folks,
I know this is a year old but was a solution ever found? I'm having the same problem. I ran a one-sample t-test with SPM and found highly significant results (t scores ranging from 3 to 14 with a sample size of 19). I then wanted to run TFCE on the data so I bundled the con nifti images from SPM for each subject into one 4D image using fslmerge (I used fslinfo to check that the resultant image's dim4 value 19) and ran randomise using the following command:
randomise -i [4D data] -o [outputname] -1 -T
The corresponding output for tstat1 was consistent with the SPM results, showing the blobs when the min threshold was 3 and max was 10ish (using fslview). But the output for tfce_p_tstat1 and tfce_corrp_tstat1 files both had a min and max of 0 (they were empty). Even when I tried to look at the uncorrected p image (tfce_p_tstat) with min value 0.1 (p = 0.90!!), nothing showed up.
What have I done wrong?
Thanks for any help!
Emily
|