DoDPI Banner for Research Papers


KIRCHER, J. C., PACKARD, T.; BELL, B. G.; and BERNHARDT, P. C. (1997) Effects of Deception on Tonic Autonomic Arousal, May 2001, Report No. DoDPI01-R-0008. Department of Defense Polygraph Institute, Fort Jackson, SC 29207-5000.

The present study tested if measures of tonic arousal are related to the amplitude of responses during probable-lie and directed lie polygraph examinations. It also tested if tonic levels of electrodermal and cardiovascular activity can be used to improve the accuracy of polygraph examinations. Three hundred and thirty-six male and female participants in a previous experiment (DODPI97-P-0016) were interrogated about their participation in a mock crime. Half of the subjects were guilty of committing the mock crime and half were innocent. Half of the innocent subjects and half of the guilty subjects received a probable-lie polygraph examination. The remaining subjects received a directed lie polygraph examination. Subjects were offered a $50 bonus to appear truthful on the polygraph examination. Tonic and phasic measures of skin conductance and skin resistance were obtained from skin conductance recordings. Tonic and phasic systolic and diastolic blood pressure measures were obtained from a Finapres? blood pressure monitor. Tonic and phasic measures of heart period were obtained from the electrocardiogram (EKG). Consistent with prior research, two tonic measures of skin conductance were positively related to phasic reactivity. Weak but significant correlations between tonic and phasic arousal also were obtained for blood pressure and heart period. However, none of the tonic measures improved the accuracy of polygraph outcomes. Tonic arousal accounted for less than 2% of the variance in the guilt/innocence criterion when used in combination with standard measures of differential reactivity to predict group membership. The results suggest that the use of absolute measures of electrodermal and cardiovascular activity would do little to improve the accuracy of computer algorithms for diagnosing truth and deception.