It is certainly possible. Are some of the clusters in abs negative? Also, I would strongly recommend using permutation if you are using such a low clusterwise threshold. See https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/MultipleComparisonsV6.0... The monte carlo simulation is not valid at such low thresholds
On 2/14/19 11:50 AM, Barletta, Valeria wrote:
Dear Freesurfers,
I am performing a GLM analysis on two groups (patients versus controls) and one variable (age).
We are using the following option to correct for multiple comparisons:
mri_glmfit-sim --glmdir ${glm_dir}/${hemi}_contrast-${contrast}_${surfname_t1} --cache 1.3 pos/abs
Is it normal to have those 2 different results when using pos or abs option? (see screenshots)
Thank you!
Vale
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer