Greg Lavine reports for the Salt Lake Tribune. Excerpt:
Academic experts who study lie detection say that polygraph tests can play a useful role in cases such as the abduction of 14-year-old Elizabeth Smart, but the scientific community remains divided over whether such examinations can accurately determine guilt or innocence.
At least two family members, including Elizabeth Smart’s father, Edward, are known to have taken polygraph tests. Investigators have left the door open to test other family members.
Polygraph testing can be especially useful in cases involving limited physical evidence, said Frank Horvath, a professor of criminal justice at Michigan State University.
“This is exactly the kind of case where polygraph testing can play the greatest role,” said Horvath, who runs the Research Center for the Detection of Deception, funded in part by the American Polygraph Association.
So-called lie detectors can clear potential suspects as opposed to pinning the crime on an individual, he said. This would probably be among the first tools investigators would deploy from their forensic arsenal.
William Iacono, a University of Minnesota psychologist and noted polygraph critic, said he sees some benefits to the test in abduction cases.
But he cautioned that the tests alone do not indicate guilt or innocence, and studies have shown that innocent subjects can “fail” a polygraph test about 45 percent of the time. Polygraph test results are not admissible in Utah and most other U.S. courts.
Undergoing a polygraph test can get subjects to open up to questions they might otherwise not answer, said Iacono, who runs a psychopathology laboratory. A significant physiological response to a certain question may indicate unease with that topic. Investigators can explore that avenue, which could lead to other areas of focus.
Horvath and Iacono agree that there is no such thing as the perfect liar. Several studies indicate that no one can beat a polygraph test by simply spinning lies.
Iacono said there are other methods that can fool polygraph tests, many of which are available on the Internet. Some methods can be as simple as tongue biting.
Lavine fails to note that Professor Iacono has also argued the “control” question “test” (CQT) polygraphy has no scientific basis. The benefits of CQT polygraphy are the admissions/confessions that are sometimes obtained from naive and gullible subjects. For more on Professor Iacono’s views on polygraphy, see his article, “Forensic ‘Lie Detection’: Procedures Without Scientific Basis.”