Last April a young woman placed a desperate call to Dr. David Lykken, a psychology professor at the University of Minnesota.
She'd just flunked a routine FBI lie-detector exam question about whether she used drugs. But she'd never used drugs, she insisted to Lykken, author of "Tremors in the Blood: Uses and Abuses of the Lie Detector."
"She was the daughter of a San Francisco policeman," Lykken recalled in a telephone interview, "and had grown up with a phobia about drugs because of all the horrific stories her dad would tell when she was a child. Her friends all knew that if they lit up a joint when she was around she'd walk out."
Nevertheless, the woman was barred from FBI employment, becoming one of scores of government job applicants victimized by faulty polygraph tests each year. "Polygraphs are a little more accurate than flipping a coin," says Lykken, who has received desperate pleas from people condemned by the all-powerful but errant government tests. There is now serious talk of lawsuits from outraged job applicants who failed the polygraph.
Lykken is not alone in his condemnation of the widely used tests. "The use of the polygraph to look at someone's general honesty is garbage," says John Furedy, a psychologist at the University of Toronto who has written widely on the subject.
In fact, say critics, it can be the most innocent who "fail" the tests. Their nervousness at having their credibility questioned can send the polygraph machine's pens flying over the paper, which credulous examiners count as "a lie." For accomplished liars, like CIA turncoat Aldrich Ames, on the other hand, lie detectors are easy to beat. "It's tea leaves and witchcraft," said a senior U.S. intelligence agency scientist, who has long campaigned for the elimination of polygraph tests.
Defenders of the polygraph, like James Murphy, head of the FBI's polygraph unit, blame "incompetent" examiners for mistakes, according to sources familiar with Murphy's explanations to Senate Judiciary Committee investigators in the wake of the Ames revelations.
Murphy could not be reached for comment, but Lykken, past president of the Society for Psychophysiological Research, blames the technique, not the technicians, for polygraph failures. "The field as a whole is incompetent, and they're using a foolish procedure," he said. "The idea of basing any significant decision on the basis of a polygraph exam is ignorant, it's foolish, it's dangerous."
Lykken believes the polygraph is "perhaps" useful for inducing confessions from criminal suspects, but even that can backfire. When Marine Sgt. Clayton Lonetree was shown the results of a lie detector test indicating he'd given Soviet spies the run of the U.S. Embassy in Moscow in the 1980s, he began to believe it himself -- even though it was untrue, according to later news accounts. Lonetree's "confession" led the CIA on a goose chase that for years deflected suspicion from Ames.
Ames beat the polygraph easily. Anyone can -- with a little coaching. One day an Army intelligence officer called Lykken and poured out his frustration that he'd flunked a polygraph for no apparent reason. He asked for advice. "I told him he could alter his breathing rate and blood pressure when responding to certain kinds of questions," Lykken said. The officer retook the test and called back gushing with gratitude. "The only hard part was learning how to keep a straight face," he said with a laugh.
Likewise, Drew C. Richardson, an FBI supervisor with a Ph.D. in physiology, boasted to a polygraph conference a few years ago that he'd taught his 10-year-old son how to beat the Control Question Test, the standard exam used to ferret out Russian "moles." The FBI transferred Richardson out of polygraphs and forbade him to talk about lie detectors in public anymore -- including to Senate Judiciary Committee investigators.
The Control Question Test measures the difference between a control question ("Have you ever done anything you're ashamed of?") and a specific question ("Have you ever smoked marijuana?"). If the specific question elicits more of an emotional response than the control question, then the subject is said to have "attempted deception." But such a response, say critics, may merely reflect a generalized anxiety, the sources of which can only be guessed at.
"It's a fine instrument for interrogation if you already have strong evidence of someone's guilt," says Furedy, "but a lousy way to find out if someone is lying about smoking pot."
Some critics put slightly more credence in the "Directed Lie" test, developed by Dr. Sheila Reed at the Department of Defense Polygraph Institute at Fort McClellan, Ala., which trains hundreds of polygraphers for the Pentagon, Secret Service, National Security Agency and FBI annually. That test requires a subject to state an obvious lie -- "I have sex with my brother every day" -- on the premise that any lie produces indications of stress, which thus provide examiners a better "base" from which to measure a subject's later, deliberate attempt at deception.
Though an improvement, the Directed Lie Test still does not necessarily separate "the anxious innocent from the anxious guilty," says Lykken. "Perhaps you're anxious because you're guilty, or perhaps you're anxious because you're anxious about the topic." Either way, says Lykken, the charts come out hopelessly muddy -- except to examiners who believe they've "found something."
Ironically, polygraph testing of job applicants has been so widely discredited that only government intelligence agencies are allowed to use it freely today. In 1988, President Reagan signed legislation banning it in private industry. Only employees suspected of theft or sabotage can be "put on the box."
But for all their efforts, critics have failed to budge a powerful "old boy network" of FBI, police and former military officials who have what Reed calls "a child-like faith" in lie detectors -- and especially the standard Control Question Test, which most still rely on despite orders to switch to the Directed Lie Test.
The upshot, say critics, is reliance on a system that lets spies go free while rejecting qualified job applicants.
Or worse. A man named Floyd Fay was arrested for murder in Ohio in 1978. Prosecutors, with no credible eyewitnesses or other solid evidence, offered Fay a deal: Take a lie detector test. If you pass, you go free; if you flunk, the results would be used in court. The gambit -- still legal in 20 states -- seemed a sure thing to Fay, certain of his innocence. But Fay flunked, stood trial and was eventually sentenced to life. Only after the real killers were found -- three years later -- did Fay go free.
The stakes aren't as high for the more than 6,000 FBI applicants polygraphed each year, but Lykken has been so angered by some cases that he recently wrote a personal letter to FBI Director Louis Freeh. "It is not only scandalous what you're doing to the dreams and aspirations of these young people, but the kinds of people that are failing (the polygraphs) are the kinds of people I'd like to see working for your agency," Lykken wrote.
Freeh has not acknowledged the letter.