• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Silenced research: lie detectors don't work

Merko

Graduate Poster
Merko
Joined
Nov 29, 2006
Messages
1,899
I just read in a Swedish newspaper (sorry) about a study published in the International Journal of Speech, Language and the Law. In the article (which I have not been able to read, but apparently it can be requested from the authors), it is claimed that a survey of 50 years of research on so-called "lie detectors" shows that these perform no better than chance.

After complaints from a company that manufactures and sells lie detectors, the article was pulled from the magazine's website (it has already been printed). First it disappeared without a trace, but now it has been replaced with a statement.

Comments, anyone?
 
I thought it was always known that they don't work. It was my impression that they were used more for intimidation or coaxing confessions than anything else. In any case I don't think it was ever claimed that they detect lies, despite the name.
 
Here's some more information (scroll down for English). This blog claims that the maker of lie detectors (Nemesysco) threatened the journal's editor with a libel suit.

Regardless of one's opinion about the article in question (as I said, I have not read it), this definitely seems like another case where court litigation is used in order to suppress scientific inquiry. The sad thing here is that in this case, it appears to have been successful - the article is no longer available. And as far as I can see, the authors have already given up copyright to the journal, so they may not have the option of publishing it elsewhere.
 
Merko said:
Here's some more information (scroll down for English). This blog claims that the maker of lie detectors (Nemesysco) threatened the journal's editor with a libel suit.

Regardless of one's opinion about the article in question (as I said, I have not read it), this definitely seems like another case where court litigation is used in order to suppress scientific inquiry. The sad thing here is that in this case, it appears to have been successful - the article is no longer available. And as far as I can see, the authors have already given up copyright to the journal, so they may not have the option of publishing it elsewhere.
Click to expand...
They do work. On rare occassions they can be fooled but they usually work. I work at the georgia Dept of Drivers services and they do work.
 
The data shows it works better than chance, mostly as a marker for stress however it does not do so consistently enough to be really all that useful.
 
Even before I was a skeptic, I was suspicious about lie detectors. The entire concept instantly struck me as very questionable.

"So they don't "detect lies", they measure your sweat and heart rate? They measure if you are nervous? Well of course you're nervous, you're being questioned about a double homicide!"
 
EeneyMinnieMoe said:
Even before I was a skeptic, I was suspicious about lie detectors. The entire concept instantly struck me as very questionable.

"So they don't "detect lies", they measure your sweat and heart rate? They measure if you are nervous? Well of course you're nervous, you're being questioned about a double homicide!"
Click to expand...
Well, the "Lie detector" doesn't really test your overall stress to a situation, it is suppose to test your involuntary stress response to making up a lie on the stop. It is suppose to be a fast and involuntary reaction. The issue with this is that reading the stress response is partially subjective which is why they have to calibrate the machine before the test.

However this stress response can be blunted by very calm and very good liars and it can be set off by individuals who are just confused by the question etc.

Just not a good test.
 
Cainkane1 said:
They do work. On rare occassions they can be fooled but they usually work. I work at the georgia Dept of Drivers services and they do work.
Click to expand...
Man...Georgia...didn't know they interrogated people like that when you're just trying to get a driver's license.
 
paximperium said:
Well, the "Lie detector" doesn't really test your overall stress to a situation, it is suppose to test your involuntary stress response to making up a lie on the stop. It is suppose to be a fast and involuntary reaction. The issue with this is that reading the stress response is partially subjective which is why they have to calibrate the machine before the test.

However this stress response can be blunted by very calm and very good liars and it can be set off by individuals who are just confused by the question etc.

Just not a good test.
Click to expand...

Can't an innocent person just be intimidated by the overall situation, though?

Here you are, held in a police station, flanked by cops on all sides, being accused of a crime you have not committed, after perhaps having been interrogated for hours and afraid that the test will show that you are lying.

It's hard to feel at ease in that situation before they start the actual questioning.
 
I was denied employment (about 20 years ago) due to "failing" a polygraph examination. I was really excited about the job, and it apparently affected the readings. The examiner repeatedly asked me if I was taking any medication.
Coincidentally, I passed a polygraph test a few months prior, for a job I was rather indifferent about.
 
Again, I haven't read the article, but I don't think that the claim is that no lie detector ever has any reliability at all. Rather, the claims appear to relate to one particular type of lie detector, which is based on analysis of the voice only.

And in any case, the question of whether or not lie detectors can be effective appears to me as less important. Even if the claim in this article is wrong, it would be extremely worrisome that the article is withdrawn because of legal threats. After all, when someone publishes an article that claims that mobile phones simultaneously improve and reduce reaction speed in test subjects, the last thing we'd want is for some cell phone company to counter that by bringing the matter to court, right?
 
I'm all for exposing bullcrap.

But the official statement made by the journal, unless they are themselves lying for Nemesysco Limited, indicates that the article contained inaccurate defamatory statements and that they failed to allow Nemesysco to respond.

It sounds as though it was removed for failing to meet the journal's standards in that respect, not because the truth was harmful to Nemesysco. So, there could be more to the story.

However, I think a more responsible decision would have been to give the authors a chance to have the article reinstated after rebutting the complaint. Or, to have published Nemesysco's complaints on the same page and let the readers make up their mind. I'm not a fan of remove-first-then-investigate, especially if no investigation follows the complaint.
 
Last edited:
I've had a glance at the article now, and I can conclude the following:

- it seems like a well researched and serious article, of course this is highly subjective, but my impression is that the article is of high quality;
- Nemesysco and mr Liberman (head of, or possibly the sole operator of Nemesysco) are the subject of very serious criticism that is unusual for a scientific article, in that it examines both the technical quality of their specific product, and also looks into the claimed qualifications of mr Liberman himself.


Given the contents of the article, I'm not entirely surprised that the journal in question felt obliged to remove it, as (unfortunately, I would say) they may be vulnerable to charges of libel in case the criticism is found to be unwarranted. However, this brings me back to the starting point again: regardless of whether the claims in the article are correct or incorrect, it seems quite wrong to silence this discussion with legal threats. How can we ever hope to expose fraudulent products or deceptive marketing if even serious scientists in the field are not allowed to publish their findings in niche publications?

I'll be very interested of course to see the response by mr Liberman that the journal has advertised. I certainly don't deny Nemesysco or mr Liberman the right to defend their technology, as long as they do so on technical grounds.
 
Well, of course polygraphs don't work.

However, one wonders which is scarier -- a lie detector that doesn't work or one that does.


INRM
 
INRM said:
Well, of course polygraphs don't work.

However, one wonders which is scarier -- a lie detector that doesn't work or one that does.


INRM
Click to expand...

One that doesn't work, in my opinion.
 
Polys measure your ability to smooze your investigator.
If you remain calm cool and collected, there's a very good chance a "guilty" person will pass.
My job required passing polys. To get the job, to keep the job, and when leaving the job.
The few in our group that failed were trying to be cute, and "fool" it with those common distractions promoted as foolers, which don't do anything but cause a fail, or, were too intimidated by the process to be calm, or were infuriated by a question.
As the questions aren't sprung out of the blue, the testee can determine before the formal test how to respond during the practice run-through.
 
I Ratant said:
Polys measure your ability to smooze your investigator.
If you remain calm cool and collected, there's a very good chance a "guilty" person will pass.
My job required passing polys. To get the job, to keep the job, and when leaving the job.
The few in our group that failed were trying to be cute, and "fool" it with those common distractions promoted as foolers, which don't do anything but cause a fail, or, were too intimidated by the process to be calm, or were infuriated by a question.
As the questions aren't sprung out of the blue, the testee can determine before the formal test how to respond during the practice run-through.
Click to expand...

The lie detector is just an E-meter on steroids.
 
I Ratant said:
My job required passing polys. To get the job, to keep the job, and when leaving the job.
Click to expand...

What happens if you want to leave your job and fail the test? Do they force you to keep working there for another three months?

If you fail one of the tests whilst working there, will it all be good if you simply claim you were going to quit anyway? :duck:
 
You must log in or register to reply here.

Back
Top Bottom