afni.nimh.nih.gov

Jim, you have two options: 1) Use option 'dmUBLOCK(-X)', or 2) model each trial separately and handle the duration variability at the group level.

Gang