Cannot tell positive clusters from negative clusters in palm outputs

[zhenghanQ]

The non-parametric outputs of palm don't seem to display the correct sign of stats, especially in the case where both positive and negative clusters exist.

In palm_clustere_ztstat_fwep.nii.gz, both positive and negative clusters were shown in positive p (actually -log10(p)) maps. Is there anyway we can generate an output that gives us either separate files for positive vs. negative results or a combined file with correct signs of stats?

Here is my palm config file:-i /om/scratch/Fri/zqi/bildl2_t1_nonpara_mask/mult_regress/model_001_task_001_contrast_old_kids_gt_young_kids/_cope_id_2/merge_copes/cope02_merged.nii.gz
-m /om/project/bild/Analysis/task_openfmri/second_level/l2output_28_f1_sd_1.96/model001/task001/all_taskmask.nii.gz
-d /om/scratch/Fri/zqi/bildl2_t1_nonpara_mask/mult_regress/model_001_task_001_contrast_old_kids_gt_young_kids/l2model/design.mat
-t /om/scratch/Fri/zqi/bildl2_t1_nonpara_mask/mult_regress/model_001_task_001_contrast_old_kids_gt_young_kids/l2model/design.con
-eb /om/scratch/Fri/zqi/bildl2_t1_nonpara_mask/mult_regress/model_001_task_001_contrast_old_kids_gt_young_kids/l2model/design.grp
-T
-C 1.96
-Cstat extent
-fdr
-noniiclass
-twotail
-logp
-zstat