In the GUI, when I run Preprocessing on a set of subjects, it looks like the previously-processed data gets labeled as the "original data." This is particularly a concern when I add more subjects to the pipeline and re-run Preprocessing. For example, I'm concerned that if I apply a particular smoothing kernel and Preprocess one batch of subjects, then add more subjects and re-run Preprocessing, that the smoothing kernel is applied twice to the first batch of subjects. (If I keep the same kernel, this shouldn't be a problem, but what if I want to re-run with a 4mm or 5mm kernel?) 

Can someone confirm that the original data (say, subject01_rsfmri.nii) is replaced by the processed data (say, suwususus_subject01_rsfmri.nii) after one Preprocessing run? Is there a way to avoid this?