Once when I was in high school my father brought home a lie detector, a simple one that measured galvanic skin resistance. My sister and I immediately began experimenting. I don't recall bothering much with issues like "Did you take the silver dimes out of my drawer?" or "Did you break the leg on my plastic horse?" We focused on learning to Beat the Man.
We weren't particularly good at lying without the machine picking up an emotional response, but we soon discovered we could make the machine give the same reading to every question, simply by panicking no matter what we were asked. If my sister asked if me if I was a Martian spy, I would envision falling off a cliff and say no. My sister would congratulate me: The machine indicated that I was a big, fat, nervous liar. And, therefore, a Martian spy.
It now occurs to me that panicking every few seconds would be exhausting in a daylong interrogation. I bet I could do it, but not everyone has my talent for hysteria. So in reading the polygraph literature at the library, I looked for tips on how to beat the lie detector. Many books on polygraphs have relevant chapters, easy to locate because some hard-working earlier scholar had gone through and marked all sections on "How to beat the lie detector" or "How to defeat the polygraph tests." For some reason.
For damping emotional responses, some sources suggest taking sedatives. Others suggest practicing with a biofeedback device to control your pulse and respiration. (A 1977 book tells the surprising story of a murderer who passed two tests by means of swallowing a lot of aspirin and washing it down with a Coke to stay calm. In 1960. I don't think that works now.) Others focus, like my sister and me, on giving strong reactions to harmless questions as well as loaded ones, by means of biting one's tongue, pressing on a tack in one's shoe, or "contracting one's anal sphincter."
I see that earlier scholar, thinking along these lines, has starred the passage in Gisli Gudjonsson's article on defeating polygraphs that deals with "Artificially producing responses to control questions."
Some people have been trained to beat the polygraph. In a study at the University of Utah, half an hour of training enabled 50 percent of the sophomores studied to fool the polygrapher. In the 1960s, under an Air Force contract, David Lykken (author of "A Tremor in the Blood") taught volunteers to beat the lie detector by damping their responses -- funding was cut before there were publishable results, but Lykken remains certain the tests are beatable. He also suspects that spies are trained to do so.
In "Tremor in the Blood," Lykken describes how he might confuse his results on a polygraph test -- a combination of simple breathing techniques, muscle tightening, tongue-gnawing and tacks in the socks. (Earlier scholar has marked this section and piously written "no-no's" next to it, presumably to avoid accidentally wearing a tack in his or her sock to the polygraph test.)
Note that most forms of the lie detector test demand that you tell lies in order to give the examiner so-called "controls." You are instructed to say "no" when asked, say, if the sky is blue, and also in answer to questions that are supposed to make virtually everyone feel guilty, like "Have you ever lied to get out of trouble?"
What if you refuse to tell the control lies? "It's fairly rare, but there are times when it occurs. If they're adamant, you don't test them," says polygrapher Skip Webb. "They're basically saying 'I don't want to take this polygraph test.'" If you refuse to lie, it's viewed as refusing to take the test. If it were true that George Washington could not tell a lie, he could not take such a polygraph test. While he could not lose his job for that, he might have to be transferred to another position without access to classified information.
It's said that a little knowledge is a dangerous thing, and this is partly true of polygraphs. While it regularly happens that innocent people with perfect faith in polygraph tests fail them, it also seems to be true that believers are somewhat more likely to give readings that correspond to their beliefs about their own innocence or guilt (which is why they perform the acquaintance test: to convert you to the faith). "If you truly believe in the procedure you're more likely to not be disturbed by the relevant questions," says Lykken. "It's always been a mystery to me how anybody passes. It must be because they believe in it."
Which means it may have been a mistake for you to read this article.