Ariana Cha's article, "Lie Detecting Devices: Truth or Consequences?" in today's (18 Aug. 2002) Washington Post includes an interesting description of the views of Paul Ekman, who is a member of the National Academy of Sciences' polygraph review panel:

Quote:

Gordon,

I would agree with you that Paul Ekman's expressed view is not necessarily shared by the majority of the NAS polygraph panel. Nonetheless, it may provide a hint regarding the direction in which the panel is headed. At a minimum, the scientific evidence on the polygraph reviewed by the committee appears not to have persuaded Dr. Ekman that the government's faith in the polygraph is well-placed.

You also write:

Quote:


Actually, what she wrote is, "No studies have ever proven that lie detectors work" (emphasis added). In context, she seems to group polygraphs and voice stress analyzers together under the term "lie detector."

Still, I think she's right, at least with regard to the polygraph "Control" Question "Test" and the Relevant/Irrelevant "Test." Because these techniques lack both standardization and any scientifically meaningful "control" whatsoever, one cannot prove that the technique "works" (in the sense of having any diagnostic validity).

As David Lykken observed in A Tremor in the Blood: Uses and Abuses of the Lie Detector, "...the theory and methods of polygraphic lie detection are not rocket science, indeed, they are not science at all." And Lykken's views are shared by most of the relevant scientific community. As we note in Chapter 1 of The Lie Behind the Lie Detector:

Quote:


And in any CQT examination, the examiner really has no way of knowing whether sophisticated countermeasures were employed or not, which is but one of the myriad uncontrolled confounding variables that prevent the CQT from having any true diagnostic validity.