Hi Yacila,

In FSLview, the threshold works just as in cluster, with 1-p, so you'd have to threshold at 0.95.

All the best,

Anderson



--
Anderson M. Winkler
FMRIB / Analysis Group
Blog | Twitter ]


On 3 September 2015 at 16:47, Deza Araujo, Yacila Isabela <[log in to unmask]> wrote:
Dear FSL experts,

After 5000 permutations I thought I got some significant results (p < 0.05, TFCE corrected) when I checked them in FSLview. However, when I want to extract the cluster information, I get some inexistent coordinates and there are not  1-p values. The script that I am using is:
cluster -i PDL0_tfce_corrp_tstat1.nii.gz -t 0.95 --scalarname="1-p" > PDL0_visual_neg.txt 

Can it be that the Fslview results (even when the threshold is set at p < 0.001) are not completely true? I am asking this because I got another significant results that are visible at a very low threshold in Fslview but do not appear in the correspondent cluster information extraction.

I would really appreciate any help or advice.

Regards

Yacila Isabela Deza Araujo
PhD Student
TU-Dresden