UK proposing use of lie-detectors for terrorists

Actually this is interesting. Polygraphs are not admissible in court and everyone knows they're not 100%. However, are they sufficiently better than chance as to have some utlilty?

I was watching this YouTube video because it was talking about polygraph tests performed on Luke Mitchell and his mother, which they both passed. I'm perfectly sure Luke Mitchell is innocent and his mother didn't cover up for him, but on the basis of the evidence, not a polygraph test. But it has impressed some people, including the (rather gushing) interviewer in the video. (Go to 1 min 10 sec to skip the plug for the sponsor.)


However the technicalities were quite interesting, including the limitations of who should not be tested and what questions can be asked.

I know they're talking about using polygraphs to help decide which sex offenders are more likely to reoffend after they're released, and as that's an inexact science at the best of times I suppose there's an argument that says any more information you can get that's better than tossing a coin is maybe worth it. This seems to be an extension of that.

Very difficult to do a controlled experiment, and I think the utility is being over-stated in the video, but I'm not convinced it's coin-toss territory either.
 
Last edited:
I know someone who applied to the NSA and went a certain distance in the hiring process. They did use a polygraph...they hooked her up to it and then shouted at her, asked personal quedtions, made dreadful insinuations, and were just generally aggressive and unreasonable. She did not respond well, and that was the end of the line. The test was never what the machine indicated, it was how the subject reacted to the questioning.

So I'd say a "lie detector" may be a useful prop as long as it's used correctly, which is not as a "lie detector".
 
Rolfe said:
Actually this is interesting. Polygraphs are not admissible in court and everyone knows they're not 100%. However, are they sufficiently better than chance as to have some utlilty?
Click to expand...
The accuracy rates are disputed, but almost certainly the controlled question test does have higher than chance accuracy at finding a true positive. However, it also has a fairly high risk of a false positive, possibly as high as 25-27%. Polygraph proponents claim that the accuracy can be 90% or higher but this is likely to be wildly exaggerated and based on selective reporting. Some studies find around 60-70%.


Field studies are difficult to conduct and lab studies have poor external validity. I was just saw a recent review but haven't had time to read it in detail.

I suspect that using it to monitor offenders is partly a bogus pipeline procedure to scare people into thinking they'll get caught.


A forensic psychology text says that false negatives are less likely than false positives, ranging 1-13% across studies.
 
Last edited:
Elaedith said:
The accuracy rates are disputed, but almost certainly the controlled question test does have higher than chance accuracy at finding a true positive. However, it also has a fairly high risk of a false positive, possibly as high as 25-27%. Polygraph proponents claim that the accuracy can be 90% or higher but this is likely to be wildly exaggerated and based on selective reporting. Some studies find around 60-70%.


Field studies are difficult to conduct and lab studies have poor external validity. I was just saw a recent review but haven't had time to read it in detail.

I suspect that using it to monitor offenders is partly a bogus pipeline procedure to scare people into thinking they'll get caught.

A forensic psychology text says that false negatives are less likely than false positives, ranging 1-13% across studies.
Click to expand...

Yes to the highlighted. England and wales have been using lie detectors for sex offenders for a while now and I think the argument is that the offenders are more likely to be truthful about their feelings if they know they are taking a lie detector.
 
Belz... said:
I'd say, use my mom's technique: look at their face and decide if they're criminals or not.
Click to expand...

"Little Timmy I want you look me in the eyes and tell me you didn't plan to detonate a vest full of C4, ball bearings, and rat poison at the bus station?"
 
TragicMonkey said:
I know someone who applied to the NSA and went a certain distance in the hiring process. They did use a polygraph...they hooked her up to it and then shouted at her, asked personal quedtions, made dreadful insinuations, and were just generally aggressive and unreasonable. She did not respond well, and that was the end of the line. The test was never what the machine indicated, it was how the subject reacted to the questioning.

So I'd say a "lie detector" may be a useful prop as long as it's used correctly, which is not as a "lie detector".
Click to expand...

I got polygraphed twice for various security clearance related reasons while in the Navy and the guy basically admitted that was the case, the machine was mostly just a prop used for the interview.

That being said going forward under that assumption there's still a difference between an NSA interview doing and every random beat cop doing it to every terrorist suspect.

The worst that could have happened to me was not getting a specific security clearance. Doing it in scenarios that could lead to conviction/arrest would be different.
 
Last edited:
TragicMonkey said:
I know someone who applied to the NSA and went a certain distance in the hiring process. They did use a polygraph....
Click to expand...
My daughter applied for a position at a police department several years ago. Part of the process was a polygraph. She failed; they said she was being deceptive about any past transgressions. When we talked about it on the phone I suggested that perhaps they told everyone that they failed and it was a test of character. I was wrong, she was the only one in her group to have failed.

They polygraphed her again and another failure. But the person who recruited her told the police chief she wanted her in, so she went to the academy anyway.

Ranb
 
Captain_Swoop said:
Are you sure
Click to expand...

It's true:
https://www.gov.uk/government/news/compulsary-lie-detector-tests-for-serious-sex-offenders

https://www.bbc.co.uk/news/uk-48775614

The main reason is to pressure those being tested into revealing more information because they think they will be caught out anyway. This of course raises certain ethical issues, including the fact that it gives authorities motivation to mispresent the accuracy of the tests. A perception that they cannot be relied on would reduce their usefulness for this purpose.
 
Irrespective of how flawed they are, the most important thing about polygraph "lie detectors" is that they produce more tangible and "objective" evidence, in order to reduce the importance of relying on witness testimony by the police who interrogates suspected criminals.

If used in the context of already convicted criminals, one of the motivations could be out trying to enforce some kind of quality standard when it comes to parole, by producing "objective" documentation of the questioning and how the people responsible proceeded. It also gives them more "objective" grounds for restricting the parolee's freedom during parole or rescinding it completely, even if their motivation for doing so is completely unrelated to the results of the "lie detector".

It should be noted that the usage of these "lie detectors" are for the most part restricted to the US. Even police and state security do not use them when giving people security clearances and employment, no matter the sensitivity and secrecy involved. This is indicative of the lack of importance placed on the "objective" evidence these kind of machines produce. Human lie detectors are more than good enough.
 
Last edited:
Polygraph tests are deeply flawed in both directions. The false positive and false negative rates, even with unpreped subjects, are abysmally high and I would hate to see the determination of if a terrorist or potential terrorist was safe to release based on a polygraph examination. Further my understanding is that almost anyone can be quickly trained to pass a polygraph test by thinking agitating thoughts during the positive controls and calming thoughts during the actual test itself.
 
Giordano said:
Polygraph tests are deeply flawed in both directions. The false positive and false negative rates, even with unpreped subjects, are abysmally high and I would hate to see the determination of if a terrorist or potential terrorist was safe to release based on a polygraph examination. Further my understanding is that almost anyone can be quickly trained to pass a polygraph test by thinking agitating thoughts during the positive controls and calming thoughts during the actual test itself.
Click to expand...

I heard they can be passed by just clenching the anus during the control questions.

The person undergoing the test clenches their own anus, I mean, they shouldn't clench the questioner's anus. That would be unexpected behavior that may be interpreted amiss.
 
They'd almost certainly get better results by randomly increasing the scrutiny of parolees rather than wasting everyone's time and money with the stupidity of a polygraph test.

Every time I read a story about increasing polygraph use I imagine a newly employed polygraph expert (sic) telling someone at a bar, "That's why I pay my dues to the union!"
 
Babbylonian said:
They'd almost certainly get better results by randomly increasing the scrutiny of parolees rather than wasting everyone's time and money with the stupidity of a polygraph test.

Every time I read a story about increasing polygraph use I imagine a newly employed polygraph expert (sic) telling someone at a bar, "That's why I pay my dues to the union!"
Click to expand...

A reason for using the polygraph with sex offenders is to prompt more 'disclosures' (e.g. of thoughts or fantasies about future offences). This is probably as important or more important than any actual detection of deception, although of course it is not openly presented that way.

I have a hunch that the proposals to use it with terrorists will be the same - to get them to disclose having radical ideas or socially unacceptable thoughts indicating 'deradicalization' failed.

In this context of prompting disclosure accuracy matters little, only perceived accuracy. It wouldn't even matter if the polygraph machine itself was real or fake, as long as they believe it's real and that it will catch them out if they lie.

It also gives the appearance that 'something is being done'. Of course, whether increased monitoring of those who 'fail' actually achieves anything is another matter. Using it to prompt disclosures may just catch those who are more gullible or suggestible.
 
Maybe they can get that British guy who made those magical "bomb detectors" to make them a genuine lie detector.
 
Elaedith said:
A reason for using the polygraph with sex offenders is to prompt more 'disclosures' (e.g. of thoughts or fantasies about future offences). This is probably as important or more important than any actual detection of deception, although of course it is not openly presented that way.

I have a hunch that the proposals to use it with terrorists will be the same - to get them to disclose having radical ideas or socially unacceptable thoughts indicating 'deradicalization' failed.

In this context of prompting disclosure accuracy matters little, only perceived accuracy. It wouldn't even matter if the polygraph machine itself was real or fake, as long as they believe it's real and that it will catch them out if they lie.

It also gives the appearance that 'something is being done'. Of course, whether increased monitoring of those who 'fail' actually achieves anything is another matter. Using it to prompt disclosures may just catch those who are more gullible or suggestible.
Click to expand...
Sorry, but using the polygraph as a prop to deceive people is bad, and certainly not any better than actually believing in the device. Our democratically elected representative governments ought to be able to do better. I will happily settle for interviewers making their own determinations of suspect honesty and proceeding from there, rather than propping up the wildly inaccurate polygraph as having any utility whatsoever.

I no more support polygraphs in this application than I would support Scientology e-meters.

Of course, I also think that if you're going to parole prisoners you have to accept that you're taking the [often significant] risk that they will commit more crimes. The only way to eliminate that risk is to keep them locked up. Claiming otherwise is an attempt to deceive the public, which is even worse than deceiving parolees with polygraphs.
 
I'm not defending it. I'm saying I believe that is a primary reason for it, and therefore disputes about its accuracy may have little impact.
 
Elaedith said:
I'm not defending it. I'm saying I believe that is a primary reason for it, and therefore disputes about its accuracy may have little impact.
Click to expand...
Fair enough. I'm mostly expressing annoyance at the idea that when someone is released from prison and then commits a terrible crime, it means that the system is broken. While that might be part of the problem, the main problem is that we can't see into a convict's heart or, more importantly, see into the future.
 
It's just a gimmick the government can use to assuage the 'something must be done' brigade.
Twenty years from now, when multiple murders and terrorist outrages have been committed by individuals released after lie detector tests, then the 'now discredited' polygraphs can be withdrawn and a fresh gimmick thought up. Today's politicians will have died or retired by then, so they won't have to carry any responsibility.
 
Elaedith said:
A reason for using the polygraph with sex offenders is to prompt more 'disclosures' (e.g. of thoughts or fantasies about future offences). This is probably as important or more important than any actual detection of deception, although of course it is not openly presented that way.

I have a hunch that the proposals to use it with terrorists will be the same - to get them to disclose having radical ideas or socially unacceptable thoughts indicating 'deradicalization' failed.

In this context of prompting disclosure accuracy matters little, only perceived accuracy. It wouldn't even matter if the polygraph machine itself was real or fake, as long as they believe it's real and that it will catch them out if they lie.
Click to expand...

Whoa, ok. I will grant that this is about something being considered in the UK and I am a US citizen. But, does the UK not have something similar-ish to the 5th amendment? Because everything you propose would be in violation for sure, assuming this would be involuntary.
 
Deadie said:
Whoa, ok. I will grant that this is about something being considered in the UK and I am a US citizen. But, does the UK not have something similar-ish to the 5th amendment? Because everything you propose would be in violation for sure, assuming this would be involuntary.
Click to expand...
You're incorrect about US law. The laws concerning how parolees are able to be treated with respect to search, seizure, and interrogation are quite different from those that govern the general population. If you're on parole in the US, just about anything can be used against you when it comes to a decision to revoke parole. For example, the average citizen can refuse to cooperate with police officers while, depending on the state, a parolee who refuses to cooperate with a police investigation can be sent back to prison based on that alone.
 
Hmm.
I always assumed 5th amendment protections effectively applied to someone keeping literal silence in the face of accusation and even conviction or release. But yes, I suppose compelling someone to speak, with the threat of further punishment should they not, is a fairly effective torture device. I should have known it was properly legal. Geez.
 
Deadie said:
Whoa, ok. I will grant that this is about something being considered in the UK and I am a US citizen. But, does the UK not have something similar-ish to the 5th amendment? Because everything you propose would be in violation for sure, assuming this would be involuntary.
Click to expand...
As far as I know it's used as part of the probation service and is voluntary.

I think Elaedith has the right of it, debating the accuracy of lie detectors is missing the point that people sometimes say and act differently when they know they are taking one.
 
Last edited:
Nobody should delude themselves into thinking the polygraph is a lie detector. When used properly, a polygraph shows areas of physiological response to questions. These can be valuable investigative leads as it shows what topics cause the suspect to have a fight or flight response.

What's more important is what's the plan for investigating areas of concern after the exam? Are there dedicated investigators ready to follow up on leads generated by the polygraph? Is there a system in place to where the investigators acknowledge receipt of the results and how is their follow-up tracked?

Reading the article, it seems like an idea someone came up with who doesn't understand the technology and hasn't thought through the resourcing requirements for after the exam.
 
Deadie said:
Hmm.
I always assumed 5th amendment protections effectively applied to someone keeping literal silence in the face of accusation and even conviction or release. But yes, I suppose compelling someone to speak, with the threat of further punishment should they not, is a fairly effective torture device. I should have known it was properly legal. Geez.
Click to expand...

Somewhat pedantically, if someone has been convicted of a crime, the freedoms and rights normally recognized by the state can be stripped from them depending on the specific judgement and legal jurisdiction.

Taking into account the differences of legal jurisdictions, once someone has been convicted of a crime, it can be up to the convicted person to prove that they should be released from prison. Of course in most cases people are simply released after they serve their sentence, so this only applies to indeterminate sentences and parole. So parole and conditional release is not something one has a right to: it's a privilege.
 
Last edited:
Deadie said:
Whoa, ok. I will grant that this is about something being considered in the UK and I am a US citizen. But, does the UK not have something similar-ish to the 5th amendment? Because everything you propose would be in violation for sure, assuming this would be involuntary.
Click to expand...

I'm describing how polygraphs are already used with sex offenders, and how I suspect they will be used with terrorists if it goes ahead. And yes, they are involuntary when the testing is attached to the release conditions.


There was also this recent announcement about a pilot of mandatory polygraph testing of domestic abuse offenders.
 
Craig4 said:
Nobody should delude themselves into thinking the polygraph is a lie detector. When used properly, a polygraph shows areas of physiological response to questions. These can be valuable investigative leads as it shows what topics cause the suspect to have a fight or flight response.

What's more important is what's the plan for investigating areas of concern after the exam? Are there dedicated investigators ready to follow up on leads generated by the polygraph? Is there a system in place to where the investigators acknowledge receipt of the results and how is their follow-up tracked?

Reading the article, it seems like an idea someone came up with who doesn't understand the technology and hasn't thought through the resourcing requirements for after the exam.
Click to expand...

Important to emphasize that these responses can be caused by a wide variety of forms of stress having nothing to do with involvement or guilt by the person being questioned. For example a question about a gun at a crime scene can induce a stress response if the person had experienced a totally different gun-related disturbing incident. Perhaps even more importantly an individual can easily learn how to suppress or mimic, the stress-induced physiological responses, and thereby manipulate the polygraph test at will.
 
Last edited:
You must log in or register to reply here.

Back
Top Bottom