U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Optimal Decision Rules for Evaluating Psychophysiological Detection of Deception Data: An Exploration

NCJ Number
225354
Journal
Polygraph Volume: 37 Issue: 2 Dated: 2008 Pages: 112-124
Author(s)
Stuart M. Senter; Andrew B. Dollins
Date Published
2008
Length
13 pages
Annotation
This paper reports on a project undertaken to improve decision accuracy for Zone Comparison Test format polygraph examinations.
Abstract
Two conclusions resulted from this study. First, the use of a “three or five” question series rule is the most effective method for maximizing the percentage of correct decisions, compared to making decisions from only three-question series or only five-question series, regardless of the specific decision rule used in conjunction with this approach. Second, the spot score and the total cutoff rules are biased toward the detection of deception and truthfulness, respectively. The authors advise that some caution must be used when evaluating these conclusions, because these results were produced using laboratory data, rather than data collected in field situations. They advise examining field data in order to determine the generalizeability of these findings, not only to additional data but in order to determine the applicability of these decision rules to real-world examinations. Also, the results of the current study do not address the potential for no-opinion decisions to be resolved during subsequent psychophysiological detection of deception examinations. Data were collected during two previous studies. Decisions were made using eight different decision rules. The decision rules varied in the number of question series used in the decision (three, five, or “three or five”), and the type of decision rule used (total cutoff rule, spot-score rule, or both). 5 tables and 20 references