Dear FS team,

I am doing a cortical thickness study using QDEC. I used FDR and got the following clusters:

Generating cluster stats using min threshold of 3.8...
Found 2 clusters
Contrast: 'lh-Diff-Class1-Class2-Intercept-thickness', 15fwhm, DOF: 69
ClusterNo  Max   VtxMax  Size(mm2)   TalX   TalY   TalZ NVtxs Annotation
---------  ---   ------  ---------   ----   ----   ---- ----- ----------
    1   -3.8743   18231      4.55   -39.8    1.5   14.1   14  precentral
    2   -3.8065   47386      2.66    -8.2   33.1   33.9    4  superiorfrontal
============================================================

  I need to get the cortical thickness values from each subject within this clusters so I am trying to use 

mri_glmfit-sim with --cache option and I wonder what value should I choose in this case for cache (2, 3.3 or 4)

and what --cwpvalthresh 0.05 or .025  to get the same results as FDR's results.

I tried:

mri_glmfit-sim \
--glmdir lh.L_frontal.glmdir \
--cache 3 neg \
--cwpvalthresh .05 \
--overwrite

# ClusterNo  Max   VtxMax   Size(mm^2)  MNIX   MNIY   MNIZ    CWP    CWPLow    CWPHi   NVtxs   Annot
   1       -3.874   18231    320.36    -39.8    1.5   14.1  0.02610  0.02410  0.02820   949  precentral
   2       -3.807   47386    700.61     -8.2   33.1   33.9  0.00020  0.00000  0.00040  1053  superiorfrontal

so as you can see the clusters are the same but their Size and NVtxs is so different. What is wrong?

Thank you for your time and help.

Antonella