February 13, 2008

Do lie detectors work?

There's a TV game show on now where they hook contestants to a lie detector machine and ask them personal questions.

Do lie detector machines actually work? Or do they work only on people who are innocent most of the time, but are no use on the kind of sociopaths who commit most crimes? Or are they basically just an intimidating prop that can assist somebody who is good at asking questions to break down somebody who is good at lying? (I suppose I could watch the show and see if that helps answer my question, but that would require me to break my rule of not watching more than three minutes of TV at a time.)

Similarly, have there been any advancements in polygraph technology in decades? I haven't heard of any. Kind of makes you wonder ...

In contrast, in the mid-1890s, my Swiss grandfather was a delivery boy for an optical glass company. One of his customers was physicist W.C. Roentgen, who was using the tubes my grandfather trundled in to invent the X-ray machine, for which he won the first Nobel prize. (Later, my grandfather, knowing a good thing when he saw it, became an international X-ray machine salesman.) The point of this digression is that scanning the inside of the body really works, so they keep inventing new scanners: x-rays, then CT scans, MRIs, and PET scans. What about polygraphs?

My published articles are archived at iSteve.com -- Steve Sailer

35 comments:

Alex said...

Do you only watch three minutes of American Idol at a time?

Lucius Vorenus said...

Steve Sailer: that would require me to break my rule of not watching more than three minutes of TV at a time

The new Terminator series, Monday nights on Fox, is worth catching.

So far the writing has been pretty tight, and I've actually gotten scared a couple of times.

Plus the babes are breathtakingly gorgeous [Summer Glau is back - she played River Tam on Firefly/Serenity].

Anonymous said...

Lie detectors work, except on people who have trained themselves to beat them. The key is to inflict pain on yourself when answering truthfully (a tack in the shoe or biting the inside of your mouth are common techniques), so that the tester just thinks you're a naturally jumpy person and the lies don't stand out.

Anonymous said...

Oh my, I watched that show tonight... it's my husband's fault. He's a very smart man, but idiot t.v. amuses him and sometimes I'll sit with him just to be with him until I can't take it anymore and come to isteve for respite.
I've never felt so much like a voyeur as I did tonight, and I've seen very many talk shows and reality shows, and it made me think a rubicon had been crossed with having such a show like this: Would you sleep with another woman if you couldn't get caught? Sheesh! The promos were even better. Coming up, there will be a woman who gets asked, "Have you ever been paid for sex?" If you haven't seen this show, and I never will again, family members who will be impacted by the answers are present within feet of the contestant, five people total.

I don't know a thing about lie detectors. My mom told me she took one back in the '70's when she worked somewhere and was asked if she knew who had stolen some goods. She lied and passed the test. I know I've read about them in crime books and took criminal justice; from what I've read what you say sounds about right.

tommy said...

There isn't a lot of solid scientific evidence on lie detectors. I believe the scanty research on the subject suggests they are far less accurate than polygraphers claim. Much of their effectiveness likely depends on how good a polygrapher is at spotting a liar regardless of the machine's readings.

Even if they aren't very effective, lie detectors are intimidating and this causes a lot of people to spill their guts before being wired up with electrodes.

Our intelligence agencies employ them during initial screenings of recruits (and during investigations), but the Russians have been far more skeptical. I know that, prior to the collapse of the Soviet Union, the KGB didn't use them; recruits were subject to difficult interrogations in place of polygraphs.

tommy said...

Here is Wikipedia's take on the reliability of polygraphs:

There is little scientific evidence to support the reliability of polygraphs. Despite claims of 90% - 95% reliability, critics charge that rather than a "test", the method amounts to an inherently unstandardizable interrogation technique whose accuracy cannot be established. A 1997 survey of 421 psychologists estimated the test's average accuracy at about 61%, a little better than chance. Critics also argue that even given high estimates of the polygraph's accuracy a significant number of subjects (e.g. 10% given a 90% accuracy) will appear to be lying, and would unfairly suffer the consequences of "failing" the polygraph. In the 1998 Supreme Court case, United States v. Scheffer, the majority stated that “There is simply no consensus that polygraph evidence is reliable” and “Unlike other expert witnesses who testify about factual matters outside the jurors' knowledge, such as the analysis of fingerprints, ballistics, or DNA found at a crime scene, a polygraph expert can supply the jury only with another opinion...”. Also, in 2005 the 11th Circuit Court of Appeals stated that “polygraphy did not enjoy general acceptance from the scientific community”. In 2001 William G. Iacono, Distinguished McKnight University Professor of Psychology and Neuroscience and Director, Clinical Science and Psychopathology Research Training Program at the University of Minnesota, published a paper titled “Forensic “Lie Detection": Procedures Without Scientific Basis” in the peer reviewed Journal of Forensic Psychology Practice. He concluded that “Although the CQT[clarify] may be useful as an investigative aid and tool to induce confessions, it does not pass muster as a scientifically credible test. CQT theory is based on naive, implausible assumptions indicating (a) that it is biased against innocent individuals and (b) that it can be beaten simply by artificially augmenting responses to control questions. Although it is not possible to adequately assess the error rate of the CQT, both of these conclusions are supported by published research findings in the best social science journals (Honts et al., 1994; Horvath, 1977; Kleinmuntz & Szucko, 1984; Patrick & Iacono, 1991). Although defense attorneys often attempt to have the results of friendly CQTs admitted as evidence in court, there is no evidence supporting their validity and ample reason to doubt it. Members of scientific organizations who have the requisite background to evaluate the CQT are overwhelmingly skeptical of the claims made by polygraph proponents.” Polygraph tests have also been criticized for failing to trap known spies such as double-agent Aldrich Ames, who passed two polygraph tests while spying for the Soviet Union. Other spies who passed the polygraph include Karl Koecher, Ana Belen Montes, and Leandro Aragoncillo. Noted pseudoscience debunker Bob Park recently commented, "The polygraph, in fact, has ruined careers, but never uncovered a single spy."

Anonymous said...

I've talked with people who've underwent the government TS/Poly clearances, and the nearest I can tell is that they are actually used to determine how well people can handle interrogations.

Anonymous said...

Summer Glau? She always seemed a bit "Special Ed" to me. Never got the appeal. Heady is something though. Leonidas's wife in "300."

"Chuck" if it ever gets new episodes is worth your while though. Adam Baldwin (NOT of the Baldwin Brothers) and Yvonne Strahowski (the most beautiful woman on TV) along with star Zach Levi will amuse you. Interesting for how well it details the character interaction rather than focusing on silly plots.

I am told by people who work in the military that polygraphs rely on the skill of the interpreter. Which to me indicates it's really the person interviewing and assessing answers rather than the machine.

Anonymous said...

http://antipolygraph.org/

Steve I saw your entry about lie detectors. I found the link above a while back while aimlessly browsing the web. More than s run-of-the-mill "bite your cheek to pass the test" sort of web site it goes into much detail about the set-up interview before the gizmo is used. The web site emphasizes that most of the benefit of the test comes about because the pre-interview is meant to convince the test subject of the value of the machine, despite its many deficiencies. The pre-interview is all about removing the normal qualifications and ambiguities from past events so you will believe the only truthful answers must be yes or no.


BTW an polygraph test about something important will likely be equipped to detect many of the simple techniques advertised as aiding you in deceiving the machine. You can beat the machine but the effort must go into spoiling the results during the pre-interview. Don't admit to knowing any techniques supposed to beat the machine or ever investigating how to beat the machine, every doubt will be seen in an accusing light.

Anonymous said...

In Britain we are about to introduce polygraph testing for the first time - to test whether convicted paedophiles have been behaving themselves.
It is really Government spin that they are seen to be doing something about crime.
Our fatherless society has had a spate of adults being kicked to death by gangs of kids.
The government can't or won't do anything about this but a story with paedophiles, clampdown and polygraph, gives the impression of action. It is activity masquerading as action.

bjdoible said...

One new form of lie detector is showing the suspect pictures of the crime scene and seeing if there is any "recognition" which can apparently be detected by MRI.

http://www.usatoday.com/tech/science/2006-06-26-mri-lie_x.htm

manindarkhat said...

Do lie detector machines actually work? Or do they work only on people who are innocent most of the time, but are no use on the kind of sociopaths who commit most crimes?

Sociopaths can fool lie detectors now because they don't emote like normal people. But sociopaths remember what they've done even if they don't feel emotion about it. Being able to detect recall of a genuine memory, as opposed to emotional reaction to it, will be the key to catching them out.

But good lie detectors, able to work on voices and facial expressions, won't just have huge effects on the present: it will alter history too. There's already a lot of data to work on in speeches by and interviews with men like
Bush and Blair. This story may not be reliable, but it predicts something:

In the mid-eighties, Sacks studied the reaction of people with aphasia as they watched a televised speech by the former-actor-turned-president. Despite being unable to grasp the skillful politician’s words, the patients were convulsed in laughter. “One cannot lie to an aphasiac,” Dr. Sacks noted. “He cannot grasp your words, and so cannot be deceived by them; but what he grasps, he grasps with infallible precision, namely the expression that goes with the words, that total spontaneous, involuntary expressiveness which can never be simulated or faked, as words alone can, all too easily.” So, why did those patients with aphasia cackle at Reagan’s speech? “It was the grimaces, the histrionics, the false gestures and, above all, the false tones and cadences of the voice which rang false for these wordless but immensely sensitive patients,” explained Sacks.

http://www.mickeyz.net/news/mickeyz/reagan_brain_damaged_or_has_something_to_conceal/

Anonymous said...

The episode last night showed a Hispanic guy that makes every other person who has been on that show look like a complete scum bag. The worst dirt they could dig up on him was, "Have you ever resented your mother for wearing black to your wedding?" and "Have you ever thought your wife deserves a better looking husband?"

Good grief. And that really was his worst dirt. He was sitting there with tears streaming down his face after confessing to those ones.

Every other guy who has been on that show ends up confessing to cheating on their girlfriends, stealing things at work, or something else truly reprehensible. This Hispanic guy never cheated on his wife, nothing.

Now, back to our regularly scheduled programming of, "Those Dirty Mexicans."

Matt said...

Mythbusters just did an episode on beating lie detectors. It was still online a couple of weeks ago at the Discovery channel site.

As a defense attorney in Maine, I have seen video's of polygraphs, and agree that the huge thing is the pre-interview, where they get you to admit that you lied about some white lie, it is all very friendly until you admit that you told a lie about something in your past, then the polygrapher freaks out that your a liar and makes it seem like you're public enemy no. 1. Then he asks you about the crime then leaves the room and watches you on hidden video for 15-20 minutes. If you fall asleep during that time, the polygraph comes back fine. If you do a lot of crying, praying, sweating etc, you are a liar. I don't think the machine really does all that much.

Martin said...

Aldrich Ames, the CIA officer who spied for the Soviets, was routinely given polygraph exams, and routinely passed them.

Of course, some policemen do believe in, and use, polygraphs. Some policemen use the advice of psychics. And some policemen in the late 80's / early 90's believed that this country was jammed full of witches covens that were murdering babies in their satanic rituals.

What some policemen believe is often not a good guide to reality.

Mr. F. Le Mur said...

Similarly, have there been any advancements in polygraph technology in decades?

Perhaps:
- Brain Imaging With MRI Could Replace Lie Detector.
- A Good Lie Detector Is Hard To Find:
- New Technology Puts Guilty Verdict To The Test.

TGGP said...

Who says most crimes are committed by sociopaths?

Michael said...

I've had a bit of experience with a few sociopaths, and they really are amazing creatures. They just don't play by the normal rules, and if you attempt to understand them in normal terms you'll just setting yourself up to be taken advantage-of by them. So my hunch is that a machine that's trying to detect whether they're lying or not is doomed to fail -- "lying" doesn't occupy the same place in the mental-emotional universe of sociopaths as it does in our minds. It's one of the quarrels I have with the more sentimental sides of lib-leftyism. So many lib-lefties -- well, really so many sentimentalists of many kinds -- seem convinced that deep inside everyone there's a normal/decent person struggling to come out. That's a hard position to cling to once you've actually interacted with a few real sociopaths.

But if, as one commenter above says, there could be machines that detect simply whether or not someone has seen or witnessed something or not (independent of whether he's lying about it), that seems like it might be useful.

Anonymous said...

That show, Moment of Truth, has to be faked. To ensure that the "contestants" will be entertaining, they must pre-screen them and find out how they will answer questions. After all, most people lead dull, ordinary lives. They don't have deep dark sectrats ready to be revealed while loved ones watch on. It's a game show, after all. BOGUS.

Michael said...

But I neglected to join in the game of "let's get Steve watching more TV" ...

I don't watch much myself -- I mostly use the TV as a device for watching movies. But! Two shows I love-love-love are "Modern Marvels" and "Dog Whisperer." "Modern Marvels" shows are one-hour docs about real down-to-earth subjects: "Gold mining." "Milk." "The Golden Gate Bridge." They're done with engineering primarily in mind, which is nearly always fun, though it does lead to some absurdities. I remember an episode on tea that seemed to spend half its length talking about the development of tea-bagging machines. But it's a consistently fab show -- informative, well-organized, not-too-flashed-up, and with some of the best educational use of info-graphics I've seen on the tube. And an hour is about as much as I really want to know about many subjects, so the length is great too.

"Dog Whisperer" is pretty cornily produced. But Cesar Milan is amazing with the dogs, and he's an amazing human psychologist too -- his ability to zero in on who people are and what they're to (and not in a negative way) is pretty awe-inspiring. Also, the way people relate to animals and pets is just fascinating -- what does a dog represent to a given individual? A projection? A child? A fun Steve-ish angle on the story is Cesar's own story. Born in Mexico ... And I've run across hints online that his U.S. citizenship status, shall we say, is not as secure as it might be ... Anyway, I generally have no patience with reality TV, but I've spent a lot of enjoyable time watching "Dog Whisperer." The dogs generally have a lot of personality too.

David said...

Summary: It's all a bunch of mystical pseudoscience on the level of theosophy, numerology, and palm-reading, indulged in by the "Idiocracy" types with which our culture and governments are replete, in order to ruin other people/make themselves look-feel important.

And outside of TCM, nothing good is on TV. There has been nothing good on it for 40 years.

beowulf said...

manindarkhat,

The Sacks story reminds me of Paul Ekman's work studying of people who appear to be mind readers, because they decode facial cues the rest of us miss.

Steve's pal Malcolm wrote a great piece about Ekman a few years ago. Its over at gladwell.com, "The Naked Face".

Anonymous said...

Of course people with mental disabilities laughed at Reagan - they were obviously liberals, and for obvious reasons...

Personally, I'd be very suspicious of anything concerning Reagan from the mid-1980's. A lot of that stuff was on the same level as the fifty "psychiatrists" in 1964 who signed an ad claiming that Barry Goldwater was neurotic.

Anonymous said...

I would be curious about the correlation between IQ, sociopaths and the ability to cheat human/machine systems like lie detectors, job interviews, college essays, etc.

It seem low-IQ people are more than fully mentally engaged just functioning in normal society and situations. Without the ability to step outside oneself and rationally analyze actors, rules, sitations and logically connected cause-effect chains, these people would be fully duped by props like lie detector tests.

Higher-IQ people like Bill Clinton seem to be able to abstract away all humanity and instictual morals in pursuit of whatever thier ambition or goals are. I have no doubt that he would pass any lie detector test when denying sexual relations with Lewinsky.

Many other bright people I've known have been able to redefine is to be whatever it is they need it to be. Are high-IQ people intellectual sociopaths? Whites as a group have been accused of this by radical Black racists groups.


http://www.youtube.com/watch?v=3sQJN1cXDO4&NR=1

Anonymous said...

Remember that the Soviets used "psychiatrists" to define the opinions of opponents as mentally ill and worthy of imprisonment for the safety of society.

Will hate-speech laws and other PC movatived ideas find people like Watson and Steve languishing in a gulag somewhere down the line?

dearieme said...

I suspect that most Britons look on lie-detectors as an inexplicable American eccentricity.

Svigor said...

Funny, I was just reading up on them the other day.

Far as I can gather, no, they don't, except as interrogation tools ("c'mon man, give it up, the machine says you're lying and the machine's never wrong").

One source (Wikipedia?) noted that polygraphs have never caught a single spy, and they've missed more than one. One spy or another (Ames?) stated that in nervous anticipation of a polygraph, he consulted his Soviet handler about what to do; his response was "get a good night's sleep, relax, and make nice with the guy running the machine."

Apparently there's no evidence that the modest improvement on pure chance shown with polygraphs is anything but a placebo effect.

Anonymous said...

Two words - Aldrich Ames. He was asked how he beat them and he said, 'Simple, they don't work.' The fact that he passed them meant it took much longer to pin him that it otherwise would have.

They might make for fun TV though.

Anonymous said...

Lie detectors are fraudulent. They are at the same level of deception as Lysenkoism. They were invented by noted propagandist and sex criminal William Moulton Marston. No honorable person sullies himself by resorting to such frauds.

Martin said...

"Michael said...

But if, as one commenter above says, there could be machines that detect simply whether or not someone has seen or witnessed something or not (independent of whether he's lying about it), that seems like it might be useful."

Yes, useful for your friendly neighborhood secret policeman.

MRI scans might well have some real value in sniffing out the truth. And therein lies the problem. Prosecutors and policemen might argue for such techniques to be mandatory, like finger-prints are today. Shouldn't their 5th amendment right against self-incrimination prevent them from being compelled to subject themselves to such a procedure? I'm sure some court, somewhere, some day, will rule in favor of the all-powerful state on this matter. But it certainly vitiates the 5th amendment. How is it not self-incrimination when your own brain is used against you?

I'm all for catching criminals. But not at any cost.

Eric said...

I don't think the answer is a simple "don't work". They do work... sometimes. Obviously the can give you false negatives, as in the Ames case.

But criminals do occasionally get tripped up by a polygraph test, so they're probably usefull as a screening tool for CIA-type organizations. As long as the organization doesn't put too much faith in people just because they pass.

Anonymous said...

(I suppose I could watch the show and see if that helps answer my question, but that would require me to break my rule of not watching more than three minutes of TV at a time.)

From "Stuff White People Like":
#28 Not Having a TV
The number one reason why white people like not having a TV is so that they can tell you that they don’t have a TV.

On those lonely nights when white people wish they could be watching American Idol, Lost, or Grey’s Anatomy, they comfort themselves by thinking of how when people talk about the show tomorrow they can say “I didn’t see it, I don’t have a TV. That stuff rots your brain.”

It is effective in making other white people feel bad, and making themselves feel good about their life and life choices.

Though these people often fill their time by talking with other friends who don’t watch TV about how they don’t watch TV, looking at leaves, cooking, reading books about left wing politics, and going to concerts/protests/poetry slams.

Generally this makes them very boring and gives you very little to talk to them about. It’s important that you NEVER suggest they are making a mistake or that there is a value to owning a TV. You should just try to steer the conversation to allow them to talk about how they are better than you.

Anonymous said...

Maybe the main value of telling people with Secret clearances that they will be subject to lie detector tests from time to time would be (1) reducing the temptation to blab about what you do at work; (2) discouraging people who are living "double lives" of one sort or another from applying for the job in the first place.

I think W. F. Buckley suggested that every guy applying to study for the priesthood be given a lie detector test which asked only one question -- something along the lines of, "Since the age of 21, have you slept with anyone under the age of 16?"

IMO, this probably WOULD discourage some people from applying. The MRI brain scan would surely be VERY cost effective for the Catholic Church.

Anonymous said...

I was interested to read the comment from the British reader that New Labour is going to introduce lie detectors over there. That's the first time I have ever heard of them being used anywhere outside the United States.

Also ... "reality television" is always and everywhere an oxymoron. It may be the greatest oxymoron of the past twenty years.

Anonymous said...

The European perception indeed is, that the lie detectors are silly Hollywood B-movie Scifi props ...

Machines cannot detect lies, because "truth" is not a natural property but a semantic value.

But they might be useful occasionally, if they frighten out the guy who's being interrogated. Even a non-naive, non-believing subject may freak out ...

suppose your heart rate is being monitored and you are being told: don't worry, we will only kill you in case your heart rate goes up 20%; why would it go up, you are just sitting there, totally safe ...

What the about the Dying Man testimonies or confessions? Watching movies it would seem that the idea is that a dying man will never lie, because he doesn't want to go straight to hell?