NCCA Banner for Research Papers


DOLLINS, A. B., CESTARO, V. L., and PETTIT, D. J. Efficacy of repeated psychophysiological detection of deception testing. April 1995, Report No. DoDPI94-R-0013. Department of Defense Polygraph Institute, Ft. McClellan, AL 36205.

Physiological measures were recorded during repeated psychophysiological detection of deception (PDD) tests to determine if physiologic response levels change with test repetition. Two groups of 22 healthy male subjects completed six Peak of Tension PDD tests on each of two test days. A minimum between test day interval of six days was maintained. The treatment group was programmed to respond deceptively to one of seven test questions while the control group was programmed to respond truthfully to all questions. The respiration and Galvanic Skin Response (GSR) line lengths, GSR peak response amplitude and latency, and cardiovascular inter-beat-interval (IBI) were calculated for each response. Analyses indicated that: except for GSR peak response latency, differential physiological reactivity during a PDD test did not change significantly during repeated tests or days; there was a decrease in average respiration line lengths during the beginning test(s) of each day; and, differential changes in average respiration line length, GSR peak latency, and cardiovascular IBI responses corresponded to deception. Power analyses are presented to assist in result interpretation. It is suggested that PDD decision accuracy, concerning subject veracity, should not decrease during repeated testing. It is further suggested that pneumograph line length and cardiovascular IBI are reliable response measures which may be sensitive to physiological changes associated with deception.

Key Words: psychophysiological detection of deception (PDD), peak of tension (POT), repeated measures, respiration, galvanic skin response (GSR), heart rate, power analysis

Director's Foreword

This study addresses the important issue of reliability as it relates to psychophysiological detection of deception (PDD) in a manner never before undertaken. Most published studies regarding PDD reliability report inter-rater and intra-rater scoring comparisons and thus are not really reliability findings. Reliability results, as reported in this study, are based on physiologic data collected from the same individual, while responding to the same test questions, on two separate occasions.

What makes this study unique is that the response measures were quantified and then compared, whereas other studies, involving test retest on the same subject, merely compared diagnostic accuracies of the results of each test. The finding in this study, that the response patterns did not change significantly between test and retest, provides additional, but not definitive, information to support the practice of repeated testing on individuals under criminal and screening circumstances.

The fact that any type of test can be reliable without being valid, but can't be valid without being reliable, suggests that every PDD test that is taught at the Department of Defense Polygraph Institute (DODPI) should have validity and test retest reliability data included in the administration manuals. Only one of the PDD tests taught at DODPI, the Zone Comparison Test (ZCT) has been studied under test retest reliability conditions. Test retest reliability studies will be pursued and will include all testing formats currently taught at the Institute.

Michael H. Capps