You can simply compute subject average PSD maps in source space for each condition. Your atlas (whether it is Destrieux or else) of scouts should be same in the source space for all subjects. Then you can drag all the avg PSD maps of each subject in Process A for Awake and avg PSD maps for Drowsy condition in Process B. This way you will be able to select equal number of files in both tabs and can run paired permutation t-test. Keep all the options e.g. time window, freq bands, source modal options etc. same when you compute PSDs. In the end you will be able to see parametric contrast of two conditions corrected significant or uncorrected scouts. You can do it for different freq bands or single freq depending on your interest.