THE PSE IS A COMPACT, PORTABLE, AUDIO SIGNAL PROCESSOR HOUSED WITHIN A STANDARD 125 MILLIMETER THICK ATTACHE CASE. THIS INSTRUMENT, TYPICALLY USED IN CONJUNCTION WITH A UHER MODEL 4000 TAPE RECORDER, SELECTIVELY FILTERS AND GRAPHICALLY PORTRAYS LOW FREQUENCY COMPONENTS OF SPEECH BELIEVED TO CHANGE DURING PERIODS OF MODERATE TO EXTREME STRESS. A CURVILINEAR TRACING OR 'VOICE CHART' IS FORMED ON HEAT-SENSITIVE PAPER. ACCORDING TO ITS PROPONENTS, THE ADVANTAGES OF PSE ARE NOT ONLY ITS NONINVASIVE CHARACTER BUT IS HIGH DEGREE OF VERSATILITY. IN ONE EXPERIMENT, DESIGNED TO COMPARE PSE WITH MEASURES OF STRESS OBTAINED FROM THE STATE TRAIT ANXIETY INVENTORY (STAI) AND OF HEART RATE UNDER CONDITIONS IN WHICH STRESS WAS EXPERIMENTALLY MANIPULATED THROUGH THREAT OF SHOCK OR THE PRESENTATION OF TABOO WORDS, 60 MALE UNDERGRADUATE VOLUNTEERS WERE RANDOMLY ASSIGNED TO ONE OF THREE EXPERIMENTAL CONDITIONS (SHOCK THREAT, EGO THREAT, AND CONTROL). IN GENERAL, THE PATTERN OF RESULTS OBTAINED FROM THE PSE DATA WAS REMARKABLY SIMILAR TO THE FINDINGS FROM THE STAI A-STATE (PSYCHOLOGICAL STRESS) AND HEART RATE MEASURES. A SECOND EXPERIMENT WAS PERFORMED TO REPLICATE THE FINDINGS OF THE FIRST, PARTICULARLY WITH RESPECT TO ESTABLISHING THE VALIDITY OF THE PSE AS A MEASURE OF HEIGHTENED LEVELS OF A-STATE ANXIETY. THIS STUDY FAILED TO REPLICATE THE VALIDITY OF THE PSE. THIS FAILURE IS ATTRIBUTED TO LOWER LEVELS OF INDUCED STRESS AND A REDUCTION IN BASELINE MEASURES OF STRESS. THE DISCREPANCIES BETWEEN THE PSE RESULTS OF THE TWO EXPERIMENTS POINT TO THE PRESENT LIMITATIONS OF THIS INSTRUMENT. ALTHOUGH THE INSTRUMENT WAS FOUND TO BE EFFECTIVE IN SITUATIONS IN WHICH A LARGE GROUP OF PEOPLE HAD BEEN EXPOSED TO RATHER HIGH LEVELS OF STRESS, IT DOES NOT NECESSARILY FOLLOW THAT THIS INSTRUMENT IS SUFFICIENTLY RELIABLE TO DETECT DECEPTION IN INDIVIDUALS IN MORE NORMAL SETTINGS. GRAPHS AND REFERENCES ARE INCLUDED. (PRG)