Hello,
You need to use the -e option with the design.grp file for a paired test ( see https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/Randomise/UserGuide#Two-Sample_Paired_T-test_.28Paired_Two-Group_Difference.29 for more details ) - can you try rerunning with that option and see if anything changes?
Kind Regards
Matthew
> Dear FSL experts,
>
> I am using two-sample paired t-test using Randomise using TFCE (n=15 conditions=pre, peri). When I threshold at 0.95 and 1, there are no significant voxels/clusters at all. This is against what I expected, as normal higher-level FEAT (with clustering threshold) returns several significant voxels/clusters - It seems strange to conclude that none of these clusters are significant because those were not deemed significant in Randomise.
>
> Even in one sample t-test for each condition, for the regions deemed significant in higher-level FEAT contrast images, there is a huge difference in terms of significance, further supporting that the regions should be deemed significant in contrast images.
>
> Can anyone provide me some insight?
> Thanks,
>
> Here are the commands I used. (For design.mat, the entries followed the conventional design matrix scheme for paired t-test)
>
> randomise -i LC_peri_vs_pre.nii.gz -o LC_peri_vs_pre -d design.mat -t design.con -n 5000 -T
>
> #design.grp
> /NumWaves 1
> /NumPoints 30
>
> /Matrix
> 1
> 2
> 3
> 4
> 5
> 6
> 7
> 8
> 9
> 10
> 11
> 12
> 13
> 14
> 15
> 1
> 2
> 3
> 4
> 5
> 6
> 7
> 8
> 9
> 10
> 11
> 12
> 13
> 14
> 15
>
> #design.con
> /ContrastName1 condition A > B
> /ContrastName2 condition B > A
> /NumWaves 16
> /NumContrasts 2
> /PPheights 2.000000e+00 2.000000e+00
> /RequiredEffect 2.368 2.368
>
> /Matrix
> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
|