There was no shortage of absurdities during last week's Senate hearing on Christine Blasey Ford's sexual assault accusation against Supreme Court nominee Brett Kavanaugh. But the suggestion that the polygraph test she passed proved she was telling the truth may have been the silliest.
"I understand that you've taken a polygraph test, Dr. Ford, that found that you were being truthful when you described what happened to you," said Sen. Amy Klobuchar (D-Minn.). Sen. Kamala Harris (D-Calif.) twice noted that Kavanaugh, unlike Ford, had not taken a polygraph test, implying that the omission was evidence of his guilt.
Klobuchar and Harris, both former prosecutors, surely know this is bunk. As Kavanaugh pointed out, polygraph results are "not admissible in federal court because they're not reliable."
Even if polygraphs worked as advertised, they would be useless in resolving conflicts between the accounts of two people who both believe they are telling the truth, as seems to be the case with Ford and Kavanaugh. But the problem is deeper than that. The basic premise of polygraphs -- that deception can be detected by looking at variations in someone's blood pressure, heart rate, respiration and galvanic skin response (a measure of perspiration) -- has never been properly validated.
"There is no lie detector, neither man nor machine," a congressional subcommittee concluded in 1965. "People have been deceived by a myth that a metal box in the hands of an investigator can detect truth or falsehood."
More than half a century later, that conclusion is still accurate. "Most psychologists agree that there is little evidence that polygraph tests can accurately detect lies," says the American Psychological Association. "Although the idea of a lie detector may be comforting, the most practical advice is to remain skeptical about any conclusion wrung from a polygraph."
While so-called lie detectors do not detect lies, they may detect signs of fear, stress or anxiety. But these are ambiguous, since a falsely accused person may respond more strongly to "relevant" questions (those related to the incident under investigation) than a guilty one who is calmer under pressure or who knows techniques for beating a polygraph.
"Countermeasures pose a potentially serious threat to the performance of polygraph testing because all the physiological indicators measured by the polygraph can be altered by conscious efforts," the National Research Council warned in a comprehensive 2003 review of polygraph research. "Certain countermeasures apparently can, under some laboratory conditions, enable a deceptive individual to appear nondeceptive and avoid detection by an examiner."
The NRC found that the quality of polygraph studies "falls far short of what is desirable" but concluded that "in populations of examinees such as those represented in the polygraph research literature, untrained in countermeasures, specific-incident polygraph tests can discriminate lying from truth telling at rates well above chance, though well below perfection." Even that assessment may be too generous.
Given the wide variation in polygraph practices and the serious methodological problems with both experimental and observational studies, the NRC noted that "study results cannot be expected to generalize to practical contexts." Psychologists Leonard Saxe and Gershon Ben-Shakhar have argued "it is impossible to predict the conditions under which polygraph test results will be accurate or inaccurate" because there is "no unique physiological reaction to deception" and "no theory which ties deception (or any criminal activity) with physiological reactions."
Polygraphs can be useful, but mainly as pseudoscientific props for cops trying to scare suspects into confessing, employers trying to show they are serious about security and politicians trying to score points. Last December, for instance, Roy Moore, a conservative Republican supported by Donald Trump, tried to contest his loss in Alabama's U.S. Senate race by citing a polygraph test that he said showed he was not guilty of the sexual abuse charges that figured prominently in the election.
I suspect that is one case where Amy Klobuchar and Kamala Harris would be quick to point out the problems with polygraphs.