The objective of this preliminary study was to examine the utility of power spectral analysis of recorded surface EMG signals in identifying altered MUAP characteristics in hand muscles of stroke survivors. We derived parameters from EMG power spectral analysis, such as frequency range and median frequency and made comparisons between data obtained from the affected and contralateral sides of stroke subjects. The goal was to identify whether power spectral analysis of the surface EMG could be used to infer changes in action potential characteristics that can occur in post-stroke muscle. We utilized both a standard single differential electrode (Delsys, Inc.) and in a separate set of experiments, a sensor array electrode (Delsys, Inc.) that consists of 5 slender cylindrical probes. Recordings were made during isometric force contractions at varying levels. Utilizing a novel decomposition system (Delsys, Inc.) and the sensor array electrode, we were able to derive MUAP parameters, such as p-p amplitude and p-p duration. Preliminary results suggest that there were significant differences in the median frequency values between the stroke and contralateral side of three tested stroke subjects; however, this did not follow the differences in MUAP p-p duration values. There were also significant differences in the median frequency recorded from the sensor array electrode as compared to the single differential electrode on both the stroke and contralateral sides of our tested stroke subjects, as expected. We discuss the possible explanations for our results, specifically pertaining to the clinical scores for each subject.