How reliable are polygragh tests?

[Bikewer]

A device that's making a resurgence in law-enforcement circles is the "Voice-Stress" test.
Detectives will blatantly lie to suspects to the effect that it's "impossible to fool".....

(I should note that such fibbing is quite acceptable interrogation practice)

Some years ago, 60 Minutes set up a fake business, a camera shop with several fake employees. They then approached three polygraph testing firms with the tale that they suspected one of the employees of stealing.
The polygraph folks were to determine who was the thief. They were all told the same thing, "We're not sure, of course, but we think it's "X".
All three found the "X" employee "deceptive".
After the results, one polygraph fellow, an ex police officer, admitted that the machine was mostly window dressing and that it's effectiveness was 90% operator-dependent.

 

[-Axiom-]

I was accused of an arson when I was a teenager and took a polygraph test.
I was told that I failed the test but nothing ever came of it.

At the time I didn't know who did it, a few years after the fact the guilty party told me that he did it.
The guilty party was never even questioned.

The arson was the burning of our attendance officer's car late at night.
It turned out that the guilty party was walking home late at night while drunk.
He walked past this car and lit his baseball cap on fire and threw it in the front seat.

I never did understand why I was accused of this, I wasn't a school skipper and had no issues with this attendance officer.

An aside, after this incident I and a couple other classmates were harassed by the school officials to the point that our parents had to get lawyers to get them to leave us alone.

The guilty party got away with this scot free, the reason he got away with this is that he was smart enough to not tell anybody that he did this, for over 5 yrs.

So in my personal experience I am quite certain that polygraphs do not work at all.

 

[quadraginta]

True, but this is from the perspective of someone who already knows they are useless. Consider the point of view of someone who thinks lie detectors are accurate. Now, the whole setup becomes useful. Through this and other deceptions, it is possible to obtain useful information. In other words, the trick might be useful.

How does one determine if the subject is or is not equipped with such knowledge?

I suppose you could ask them, but how would you know if they were telling the truth or not?

 

[Mr. Scott]

$cientology, anyone? Their 'auditing' device is not more than a galvanometer that uses two tin cans for electrodes. Sweaty palms and the force of one's grip can deflect the meter more than any lie one might tell.

Maybe you missed my earlier post:

I suspect Scientology audits using their "E-meter" as a lie detector to discover wealthy closeted gay celebrities they can blackmail.

 

[Christian]

How does one determine if the subject is or is not equipped with such knowledge?

I suppose you could ask them, but how would you know if they were telling the truth or not?

First, one must know the person fairly well (e.g. an employer who has worked for a while with a specific employee). Then, the most one can hope for is an educated guess. It can be a pretty good guess if you know for a fact you are dealing with someone who seldom reads or studies.

 

[quadraginta]

My thought was that you could put 'em on a polygraph and ask 'em.

 

[rjh01]

Mythbusters did a session on this subject. I was a bit disappointed in what they did, but they could not beat it. They did admit that this did not prove it could not be beaten.

 

[Sledge]

Steven Seagal beat one in The Glimmer Man, but then he is not as other men.

 

[elgarak]

The "polygraph" is an interrogation technique that is based on the assumption that the interrogated person believes there is such a thing as a reliable lie detector, and s/he is hooked up to it.

As such, the question is meaningless.

 

[steve s]

Mythbusters did a session on this subject. I was a bit disappointed in what they did, but they could not beat it. They did admit that this did not prove it could not be beaten.

Grant beat the machine. He got to fly back to SF in first class. Tory and Kari had to take a bus back.

They were testing ways to beat the machine, which assumes it works in the first place. I would have preferred seeing them test whether it's accurate or not with simple questions such as "Is it raining outside?" They could blind the test by having a second polygraph expert who didn't know what questions were being asked review the printout and make a determination if the subject was lying or not.

Steve S.

 

[shadron]

The reason why it persists in the agencies that use it is that they have nothing better. There is simply no way yet to read an man's mind. But security is security, and where would a security agency be if they admitted they could not deliver? Polygraphers, like graphologists, homeopaths and chiropractors, have been assiduous in supporting their life's work, and of course polygraphy has a long history, but most don't seem to get into the contents of that history in any depth.

 

[carolynf1]

Penn and Teller's ******** did a pretty good show about the Polygraph. They interviewed a guy whose career was ruined (he worked for the pentagon, I believe) and lost his top security clearence because the polygraph indicated deception. I heard they don't even stand up in court as evidence.

 

[quadraginta]

We know what at least one Supreme Court Justice thinks about polygraph tests. I wonder if he's ever had any cases before him where someone has been denied a job for refusing to take one?

 

[Aitch]

Steven Seagal beat one in The Glimmer Man, but then he is not as other men.

Ah, but he was using worry beads to keep his cool - I don't think they would have allowed that.

Still one of his better films, though. And I speak as someone sad enough to have most of them on DVD. Even The Patriot.

 

[Sledge]

Ah, but he was using worry beads to keep his cool - I don't think they would have allowed that.

Hmm, true. Although as Seagal has no emotional range, I don't think the machine would have registered anything anyway.

Still one of his better films, though. And I speak as someone sad enough to have most of them on DVD. Even The Patriot.

I think Gimmer Man is the last of his films I can unironically recommend. It's not a bad action/cop flick. Everything after that was direct to video pap.

 

[brenn]

Most of what I've been reading about polygragh testing is that it's pseudoscience. Still, even agencies like the FBI and CIA use them regularly. Is this method useful at all?

The machine is as useful at detecting the truth as the operator is (by which I mean, he could do it equally well with no machine) - however, it is very useful for getting people to tell the truth, under the belief that they will be, or have been, found out by the machine.

 

[Christian]

My thought was that you could put 'em on a polygraph and ask 'em.

Good idea.

 

[Christian]

The machine is as useful at detecting the truth as the operator is (by which I mean, he could do it equally well with no machine) - however, it is very useful for getting people to tell the truth, under the belief that they will be, or have been, found out by the machine.

Yes, I agree.

 

[Biff Starbuck]

I thought this 60 Minutes segment was well done. Even the doctor who developed the counterintelligence questions said the test was not intended for mass screenings.

 

[George II]

Aparently, Aldrich Ames was able to pass two polygraph tests after he started selling secrets to the Soviet Union.

Anyway, it seems pretty logical that since a polygraph basically only detects signs of nervousness, you would still have to determine what the specific source of that nervousness is. Is it the fear of being caught lying, or is it the fear of failing the test despite telling the truth, or something else? But I guess people who want easy answers would feel this is making things too complicated?