KIRCHER, J. C., PACKARD, T.; BELL, B. G.; and BERNHARDT, P. C. (1997) Effects of Prior Demonstrations of Polygraph Accuracy on Outcomes of Probable-Lie and Directed-Lie Polygraph Tests October 2001, Report No. DoDPI02-R-0002. Department of Defense Polygraph Institute, Fort Jackson, SC 29207-5000.
Traditional analog polygraph instruments typically record skin resistance (SR), The present study tested if the stimulation pretest improves the accuracy of probable-lie and directed-lie tests. 336 men and women were recruited from the general community and were paid $30 to participate in a mock crime experiment. Equal numbers of males and females were assigned to one of 16 cells in a 2 X 2 X 4 factorial design, with two levels of Guilt (guilty and innocent), two levels of Test Type (probable-lie and directed-lie), and four variants of pretest procedures. Half of the participants were guilty and half were innocent of committing a mock theft of $20 from a purse, and all participants were promised and paid a $50 bonus if they could convince the polygraph examiner that their innocence. Half of the participants were given probable-lie tests and half were given directed-lie tests. 120 participants were not given the stimulation pretest (no-pretest). 120 participants were given the stimulation test and told that the polygraph clearly revealed their deception (effective-feedback). 48 participants were given the stimulation test but no feedback about the outcome (no-feedback). The remaining 48 participants were given the pretest and told that the polygraph failed to reveal their deception (ineffective-feedback). As compared to the no-pretest control condition, the combination of the pretest and effective feedback increased the accuracy of decisions from 77% to 90% for the probable-lie test and from 75% to 83% for the directed-lie test. Additional comparisons revealed that the observed improvement in decision accuracy for probable-lie tests was due to the pretest and not the feedback. However, for the directed-lie test, decision accuracy was higher when the pretest was followed with effective feedback than when it was not. There were no significant differences between probable-lie and directed-lie tests in the accuracy of decisions by independent numerical evaluators or computer diagnoses, but respiration measures were more diagnostic for the probable-lie test. Among non-traditional physiological measures, skin potential was found to be as diagnostic as skin conductance, and measurements of blood pressure from a finger were found to be at least as diagnostic as the cardiograph.