NCJ Number
187443
Journal
Polygraph Volume: 29 Issue: 4 Dated: 2000 Pages: 344-356
Date Published
2000
Length
13 pages
Annotation
This paper examines various aspects of polygraph validity and reliability, with the major portion of the paper dealing with the evaluation of 100 confirmed truthful and deceptive single-issue computerized polygraph charts that were blind scored independently by two polygraphists.
Abstract
Using 7-position and 3-position scales with various cutoff points, the findings indicate that the 7-position scale with a +/-6 cut off had the highest accuracy, but also one of the highest inconclusive rates. In contrast to prior research, accuracy was higher for the truthful than the deceptive. Comparisons were made with the validity findings of Patrick and Iacono (1987, 1991) and the original examiners as well as the blind evaluators. It was determined that the accuracy of the original examiner was always higher, but there is reason to believe that it was because they used extrapolygraphic information not available to the blind evaluator. In the second part of this study, various indices of reactivity were evaluated to determine which of these were the most effective in evaluating truthfulness and deception. Particular emphasis was placed on respiration and some of the Department of Defense Polygraph Institute's recommended 22 evaluative criteria. In the third phase of this investigation, the various sensors were compared to determine which provided the most accurate data for a final determination of truth or deception. The electrodermal response contributed the most, followed by the cardiograph and then respiration. In the fourth part of this work, a study was made of the validity of Patrick and Iacono's argument that one cannot generalize from confirmed polygraph findings that were based on admissions to unconfirmed examinations, such as those being considered for admissibility in court. 3 tables and 34 references