mental_floss magazine
SUBSCRIBE >
GIFT SUBSCRIPTIONS >
DIGITAL SUBSCRIPTIONS >
subscriber services >
With the possible exception of politicians and small wooden puppets named Pinocchio, most people have a hard time lying with a straight face—and an even harder time lying successfully when their every move, breath, inflection of speech, and variation in blood pressure is being monitored.
While the accuracy of the modern lie detector, or polygraph machine, is considered dubious by many researchers—in 2002, the National Academy of Sciences determined the polygraph to be essentially useless—it’s popularly believed that a simple machine can really determine whether or not a person is telling a lie.
Of course, we’ve bought into a lot of crazier ideas in centuries past. In medieval England, a person thought to be lying might be subjected to a test of fire (walking across hot stones; carrying a scorching hot iron rod) or water (being tossed into the pond). If a person was burned in a trial by fire, it was considered sufficient evidence for a hanging. A person tried by water had an even worse deal: if you floated, you were guilty, and sent to the hangman’s noose. If you sank, you were considered innocent, but since you were dead from the drowning, it didn’t make too much of a difference.
Untruth or Consequences
By the 19th century, governments were no longer throwing people in ponds (not as a measure of truth-telling, anyway), but the methods used to assess a person’s character were still pretty dubious. Phrenology—the study of bumps on the skull—and the new discipline of psychology gave rise to the idea that physical characteristics and behavior could demonstrate a person’s moral character, and thus their truthful or deceitful nature.
In 1895, Cesare Lombroso theorized that sudden changes in a person’s blood pressure could be an indication of lying, and he attempted to chart these changes with a device called “Lombroso’s Glove.” More sophisticated machines—simultaneously recording blood pressure, pulse, respiration, and galvanic skin response—were later developed by Dr. John A. Larson and Leonard Keeler (widely considered the father of the modern polygraph machine) in the late 1920s and early 30s. But it was a psychologist named William Marston who first popularized the “lie detector,” and gave it the cultural prominence it has today.
Lasso of Truth
Marston was hired by the US government during WWI to come up with a way to make sure that prisoners of war told the truth during interrogation. Echoing Lombroso’s experiments, he decided to test his subjects’ blood pressure during their interviews. In 1917, he published his findings to great acclaim in the press, who hailed him as the inventor of the “lie detector.” Marston didn’t shy away from the title. In fact, he publicly stated that his discovery hailed “the end of man’s long, futile striving for a means of distinguishing truth-telling from deception.”
Marston remained a firm advocate for the implementation of the polygraph into the court system, and was brought in to administer a lie detection test for the 1923 case of Frye vs. United States. The court found that the test could not be considered reliable enough to be 
used as evidence, though, so Marston’s tests were thrown out. Essentially, the court ruling established a precedent, and polygraphs have, for the most part, been kept out of the courtroom ever since.
A tireless advocate for truth, Marston was undaunted by the court’s ruling. Instead, his obsession with honesty would later fuel his work in creating the most enduring female superhero in comic book history, and the greatest lie detector of them all—Wonder Woman, whose magical golden lasso compelled villains caught inside it to tell the absolute truth.
This article was written by Ransom Riggs and excerpted from the mental_floss book In the Beginning: The Origins of Everything. You can pick up a copy in our store.
Speaking of Lie Detecting…
We recently released our first iPhone trivia game. You’re given two statements – one true, one blatantly false. Your job is to spot the Big Fat Lies! If you’ve got an iPhone (or iPod Touch) and $2.99, head on over to the iTunes Store and get downloading. (That link should open up your iTunes and send you right to Big Fat Lies! page of the store. If not, you can search for “Big Fat Lies” and it will come right up.)
Comments (14)
Comment
- Follow mental_floss
I think I would get so nervous if I was hooked up to one of these things that I would probably fail even when stating my own name!
This article also reminded me of MEET THE PARENTS when they called the Dad a human lie detector… That was a GREAT movie…
posted by GTT on 12-12-2008 at 11:03 am
I am sure this article is true, but I would like to know why they are “useless”.
Is it because they work on some people, but not all? Does it work on “normal” people but not on pathological liars who have no conscience? Or does it fail randomly on any given subject. Does it work well, but people can be trained to fool it? Is there absolutely nothing coherent at all coming out of the data?
Even if it doesn work, if some people believe it might, then there is value in it.
posted by Morris on 12-12-2008 at 11:40 am
In the eighties, I was given a lie detector test as a prerequisite to being hired at a certain company. I lied several times during the course of the testing, but passed anyway, and was hired by the company…
posted by big fat liar on 12-12-2008 at 12:10 pm
I remember in one of Diane Mott Davidson’s culinary mysteries one of the characters was given a lie detector test after downing a dozen or so expressos during the day and as a result flunked.
posted by Beth on 12-12-2008 at 12:11 pm
Obviously there’s tons of people who would naturally be nervous to take the test even if they had nothing to hide. I think the test facilitor begins the test with normal questions like, “Is your name _______?” They measure your blood pressure and pulse during these obvious truths and falsies and compare them to what happens when they ask the “real” questions. I think the only way to fool the test would be to have the ability to stay calm during it’s entirety.
In my uneducated and spontaneous opinion that has no credibility, these tests are usually right. Unfortunately, “usually” doesn’t hold up in court.
posted by Bryan on 12-12-2008 at 12:25 pm
I, too, took a lie detector test in the ’80s for employment. (Mine wasn’t a ‘certain company’, it was RiteAid.)Apparently, I passed.
They ask a few obvious questions: “Is your name…” and the like. I guess this establishes your patterns so they can differentiate your truth from lies.
The trouble with these tests, or any simlilar test or survey is that the questions are necessarily phrased as yes-or-no.
Now, when asked if you’ve ever stolen from an employer, for the overwhelming majority of us the answer is yes. You did not buy that pen and you know it is not yours and that your employer doesn’t want you to take it home. But, gee, does that really count? Now you’re really nervous. How do you answer?
That may be a lame example but it’s all I could think of. The point is the questions can be faulty causing you to answer them in ways that are essentially honest, but may indicate dishonesty. A machine cannot make that judgment.
(If anyone knows what I mean and can explain it better, please do.)
Also, I would assume sociopaths can easily pass them.
posted by BassMan on 12-12-2008 at 12:32 pm
Hey BassMan:
What’s the biggest, or most expensive, or most ongoing thing you’ve ever stolen from an employer?
Also – Mental Floss – this’d be a cool
@ of the Day – are there famous cases of workplace hoarding and stealing?
BTW I used to work in a movie theatre and we had a really gross way of balancing the count…. so that we could have free candy.
posted by Mare on 12-12-2008 at 12:53 pm
The way to beat lie detectors is not to try to remain calm during the lie, as that can be difficult, but to be agitated throughout the testing so that the lies and truths are physiologically indistinguishable. For example, right before answering every question you do something that causes yourself some slight pain — bite your tongue, pinch your leg, something like that. The pain will cause your blood pressure and heart rate to rise. The machine will then record the same physical responses with regard to every question, and it looks like there are no lies.
By the way, polygraphs are so out of favor that Congress passed a law basically prohibiting employers from using polygraphs in the employment context — the Employee Polygraph Protection Act (EPPA).
posted by Pete on 12-12-2008 at 1:45 pm
A family member of mine worked at the FBI for awhile, and part of the screening involved was a lie detector test–this was in 1999.
From what she said, they asked you questions that they made you lie about so they could measure what exactly your body does when a lie is told. It seems to me that that would help with the accuracy of the test, but as many have pointed out, they are by no means foolproof. They can be tricked and pathological liars can pass them.
posted by Orange on 12-12-2008 at 1:56 pm
I used to work in a sheriff’s office that required all potential employees to pass a lie detector test as if it were a valid screening device.
The key value of a lie detector machine is in convincing subjects they’ve been found out and had better confess. That’s one reason somebody conducting the test invariably ask an identical question two or three times. There’s a natural tendency to assume that getting asked the same question means your first reply registered as false. From that point, a trained interrogator will have an easier time to provoke confessions from dumb people who fear the machine is legit.
Smoke and mirrors — that’s all that lie detector machines amount to.
posted by Holly B on 12-13-2008 at 8:59 am
i dont think the test proves anything but that the person’s heartbeat quickened when he answered the question and so its completely useless…but unfortunately cops are willing to believe anything that makes a person look guilty
posted by nathenism on 12-13-2008 at 11:18 pm
i’ve also heard that in some military circles, training has been given to deter/obfuscate a lie-detector (if caught, i would imagine).
If this is true…and i’m sure it’s not that hard to ‘redefine’ the question being asked, before answering…then the rest is also hooey.
posted by Gabacho Mike on 12-14-2008 at 6:40 am
I remember a story about some cops during an interrogation who put a colander on a guy’s head with wires leading to a copy machine. Every time they’d ask him a question, someone would hit the copy button and a sheet of paper saying “He’s Lying” would come out. He confessed.
Would that be lie detection via placebo? Or just a stupid crook?
posted by Dave on 12-15-2008 at 12:49 am
I had to take a voice stress test once (a version of the lie detector) I only lied on one question, but it said i lied on almost all of them. I held my ground, though, since I figured they were just trying to get me to confess. I didn’t actually do anything though, so I was pissed when it said i lied when i didn’t. The guy giving it said it was 99.9% accurate, but obviously it was not.
posted by kelly on 6-18-2009 at 12:05 am