Surface electromyography of the masseter and electrocardiogram recordings of heart activity during sleep were performed on nine subjects who suffer from an oral motor dysfunction (bruxism) during sleep. Signals were monitored in the subject's home sleeping environment over 4 consecutive nights. A total of 36 nights of data were analyzed to perform the following: (1) describe the nature and magnitude of total masseter muscle electrmyographic activity above a minimum threshold of 3% of each subject's indivudally established maximum voluntary contraction level; and (2) describe electrocardiograph rate changes (using the R-R interval) that occurred in relation to these electromyographic elevations. From these data, criteria for detection of bruxism events were established and combined into a fully automated event detection algorithm. The mean number and duration of the detected bruxism events are reported. The underlying logic for the criteria selected, and what effect other possible criteria would have on the separation of abnormal from normal motor events, is also presented and discussed.