[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.205.87.3. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Article
January 9, 1987

The Predictive Power of the Polygraph: The Lies Lie Detectors Tell

Author Affiliations

Chicago Medical School North Chicago, Ill
New England Deaconess Hospital Harvard Medical School Boston
Georgetown University School of Medicine Washington, DC

Chicago Medical School North Chicago, Ill
New England Deaconess Hospital Harvard Medical School Boston
Georgetown University School of Medicine Washington, DC

JAMA. 1987;257(2):190. doi:10.1001/jama.1987.03390020055022
Abstract

To the Editor.—  We were particularly interested by the recent report on the polygraph,1 and commend the Council on Scientific Affairs for presenting a comprehensive overview of the current state of knowledge. The report drew attention to the problem of false-positive and false-negative errors in the interpretation of polygraph records. We recently attempted to quantify this problem by reviewing the published literature to determine the sensitivity and specificity of the polygraph and then using these data to calculate the predictive value of polygraph screening for the detection of liars.2We found the predictive power of the polygraph to be poor, since it supplies little new information over the "prior probability" that a subject is telling a lie. For example, if an investigator screened a group comprising 50% liars by flipping a coin, diagnosing a "liar" every time it came down heads, he would detect the actual liars in

×