Slashdot Mirror


Thermal Imaging Lie Detector In Development

beaverdownunder writes with this quote from the BBC: "A sophisticated new camera system can detect lies just by watching our faces as we talk, experts say. The computerized system uses a simple video camera, a high-resolution thermal imaging sensor and a suite of algorithms. ... It successfully discriminates between truth and lies in about two-thirds of cases, said lead researcher Professor Hassan Ugail from Bradford University. ... We give our emotions away in our eye movements, dilated pupils, biting or pressing together our lips, wrinkling our noses, breathing heavily, swallowing, blinking and facial asymmetry. And these are just the visible signs seen by the camera. Even swelling blood vessels around our eyes betray us, and the thermal sensor spots them too."

183 comments

  1. Let's get this out of the way by paiute · · Score: 2

    You're in a desert, walking along in the sand, when all of a sudden you look down...

    --
    If Slashdot were chemistry it would look like this:Cadaverine
    1. Re:Let's get this out of the way by Anonymous Coward · · Score: 0

      which desert? why am I there?

    2. Re:Let's get this out of the way by publiclurker · · Score: 1

      It doesn't make any difference what desert, it's completely hypothetical.

    3. Re:Let's get this out of the way by gknoy · · Score: 1

      Why are you letting that turtle die? Do you hate turtles?

    4. Re:Let's get this out of the way by Anonymous Coward · · Score: 0

      Well depending on the desert and the time of day as well as the season and year I could find any sort of object living or not at my feet.

    5. Re:Let's get this out of the way by JustOK · · Score: 1

      sure it's not just some furry?

      --
      rewriting history since 2109
    6. Re:Let's get this out of the way by wierd_w · · Score: 1

      It's a mutant turtle. It was only a teenager in the mid 80s. Its almost 30 now, and it lived in the newyork sewers in it's rat-like parent's basement, getting morbidly obeise on stale pizza crusts.

      Since its acting career dried up shortly after the first motion picture, I am just trying to be merciful.

      Death in the desert is far better for it than the unlife that is comic cons and newyork after dark.

      I don't hate the turtle, I only want to ease its suffering!

    7. Re:Let's get this out of the way by BluBrick · · Score: 1

      If you are suggesting I might not be able to tell the difference between a turtle and something furry, then yes, I'm sure. Now, I may not be quite so sure it's not a rock, but yes, I really am sure it's not just some furry.

      --
      Ahh - My eye!
      The doctor said I'm not supposed to get Slashdot in it!
    8. Re:Let's get this out of the way by Tsingi · · Score: 1
      LOL!

      That is precisely what I thought of when I read this. Not just the test, the tortoise.

    9. Re:Let's get this out of the way by Anonymous Coward · · Score: 0

      at my boner?

    10. Re:Let's get this out of the way by Anonymous Coward · · Score: 0

      Since its acting career dried up shortly after the first motion picture...

      If only.

    11. Re:Let's get this out of the way by Anonymous Coward · · Score: 1

      Was Decker a robot or what? I have never been able to figure that out.

    12. Re:Let's get this out of the way by FatdogHaiku · · Score: 2

      A furry turtle... is that some kind of euphemism?

      --
      You have the right to remain sentient. If you give up the right to remain sentient, you will be elected to public office
    13. Re:Let's get this out of the way by Alex+Belits · · Score: 1

      No, it's supposed to be a turtle furry.

      --
      Contrary to the popular belief, there indeed is no God.
    14. Re:Let's get this out of the way by bcoker · · Score: 1

      You're in a desert, walking along in the sand, when all of a sudden you look down...

      What's a tortoise?

    15. Re:Let's get this out of the way by Hylandr · · Score: 1

      Point made.

      --
      ~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
    16. Re:Let's get this out of the way by Anonymous Coward · · Score: 0

      Wow... sand.

    17. Re:Let's get this out of the way by Stavr0 · · Score: 1

      Let me tell you about my mother.

    18. Re:Let's get this out of the way by bkr1_2k · · Score: 1

      A turtle that picks its shell up off the ground when it walks, rather than dragging its shell like a half-dead cyborg.

      --
      "Growing old is inevitable; growing up is optional."
  2. YIKES!!! by RyanCheeseman · · Score: 0

    no more cheating the lie detector with a tac in the foot or flexing your sphincter.......

    1. Re:YIKES!!! by Ruke · · Score: 1

      I see no reason why these things wouldn't cause visible discomfort in your face, as opposed to whatever (skin conductivity/heart rate?) a traditional lie detector measures. The whole point is to mask truths and lies alike, so there is no way to tell them apart.

    2. Re:YIKES!!! by couchslug · · Score: 3, Insightful

      "no more cheating the lie detector with a tac in the foot or flexing your sphincter......."

      Damn, I'll have to find another excuse for doing those things.

      --
      "This post is an artistic work of fiction and falsehood. Only a fool would take anything posted here as fact."
    3. Re:YIKES!!! by gatkinso · · Score: 1

      You'd think they don't sit you on a pad and make you take off your shoes.

      --
      I am very small, utmostly microscopic.
    4. Re:YIKES!!! by wwphx · · Score: 1

      They didn't when I took a polygraph to work for the police department. But that was also 20 years ago.

      The thing that's scary is people are going to think these things are infallible when they say only 2/3rds of the time. This thing is garbage, they've discovered a casual relationship.

      --
      When you sympathize with stupidity, you start thinking like an idiot.
  3. sounds like ... by recrudescence · · Score: 4, Interesting

    Tyrell: Is this to be an empathy test? Capillary dilation of the so-called blush response? Fluctuation of the pupil. Involuntary dilation of the iris...
    Deckard: We call it Voight-Kampff for short.

    1. Re:sounds like ... by blair1q · · Score: 1

      I just do eyes.

    2. Re:sounds like ... by MidnightBrewer · · Score: 1

      Sounds like prior art to me!

      --
      "Give a man fire, and he'll be warm for a day; set a man on fire, and he'll be warm for the rest of his life
    3. Re:sounds like ... by daem0n1x · · Score: 1

      Get a really hot interviewer and the machine discovers everybody is a liar.

    4. Re:sounds like ... by anubi · · Score: 1

      Yes, it looks like prior art to me too.

      They are just trying to automate what professional interrogators have known for years. Even in the 1950's, Sergeant Dan Matthews of "Highway Patrol" was shown using a Metrigraph psychogalvanometer to measure skin resistance of a subject. ( Not to be outdone, a whole religion was started by a science-fiction writer that centered on the use of this device, renamed to "E-Meter")

      Lie detection is an art. Very subjective. How does one differentiate the stress caused by being under test and knowing the machine is determining your future from the stress of fibbing?

      Its the same thing that causes blood pressure to rise when your physician measures it, or causes perfectly good students to fail tests. (Especially math tests). I have seen it. And experienced it. Too many times.

      I'll betcha the arresting patrolman's gut feeling will outdo any testing any day. Not to say he's always right, but I would bet probability is in his favor.

      --
      "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]

  4. 2/3 is still not good enough. by hedwards · · Score: 4, Interesting

    All this does is change the rules a bit. All of the things they've listed are things which one could train to do or not do on cue. And even without training if it's only good 2/3 of the time that's not good enough to justify deployment.

    1. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 1

      Or the untrained like me might do when I'm not lying. I get nervous during interrogations even when I have nothing to lose. This is when I'm not lying. Tougher the question, more chance I'm going to be nervous while thinking about it. (what was I doing Wednesday night 3 weeks ago, beats the hell out of me but if I'm going to jail over it, I'll be breathing heavily, swallowing, etc).

    2. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 0

      Errrr, 2/3rds is 66%, or 16% better than blind guessing. Amazing. I would imagine that more than 16% of people fart loudly when lying, so that would be a better test.

    3. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 0

      Really? The way it works is, they "know" when you're lying, so (under pressure from a skilled administrator) you get scared and tell the truth. Same as a polygraph, but it's "new tech", you don't know your existing lie-detector fooling skills work, so it undermines your confidence and you lose.

      Seems worth deploying in some cases (i.e. those where you expect reasonably trained subjects, but not professional operators who'll be up-to-date on it) to me.

    4. Re:2/3 is still not good enough. by hedwards · · Score: 1

      Professionals are the ones that you're looking for, the trained but not professionals are easily handled with traditional interrogation methods. Nobody in their right mind agrees to a polygraph test as they're notoriously inaccurate and ultimately even a clean test doesn't mean you're off the hook.

    5. Re:2/3 is still not good enough. by Zancarius · · Score: 2

      Not only that, but...

      We give our emotions away in our eye movements, dilated pupils, biting or pressing together our lips, wrinkling our noses, breathing heavily, swallowing, blinking and facial asymmetry.

      Most people who are nervous for whatever reason will do at least one or more of those things even when they're not necessarily lying. So, congratulations, you just pegged someone for being nervous!

      --
      He who has no .plan has small finger. ~ Confucius on UNIX
    6. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 1

      Pain avoidance does work on such responses and those can be automated. It works better with volunteers who want this skill. It's pretty brutal but no worse than training some special forces receive.

    7. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 1

      Same reason I'd never take a lie detector test -- except if I was guilty and hoped to clear myself.

      I'm one of those who always gets asked lots of questions by US customs and gets my luggage searched when everyone else in my party is waved through after the cursory questions (why were you in X, how long were you there,...). I just don't like people asking me questions about stuff that's MY business and really none of theirs. My resulting body language apparently is interpreted as guilt. I'm not belligerent, I don't refuse to answer their questions or evade them so that doesn't explain it.

    8. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 0

      "Professionals are the ones that you're looking for,..."

      Yep, real professionals who'd inject you with Botox, that's why the LAPD doesn't hold its breath for this new tech.

    9. Re:2/3 is still not good enough. by Antique+Geekmeister · · Score: 1

      I suspect that the training to manipulate a normal skin resistance based lie detector, such as a Scientology "e-meter", would apply equally well to this technology. It would be fascinating to run the test against someone trained to such hypnotic responses to questions and see if this tool is as easily manipulated by the same mental and physical techniques.

    10. Re:2/3 is still not good enough. by sjames · · Score: 1

      I have a much cheaper and simpler "lie detector" that manages to be 75% as good as that but costs only $0.01. There is a slightly sturdier model available for $0.25 though it may have a bit of a truth bias.

      This sounds like yet another fine product of Snake Oil LTD. meant to fleece empty headed DHS officials and the taxpayer.

    11. Re:2/3 is still not good enough. by hedwards · · Score: 1

      More or less. I'm not sure what the Israelis do for training their airport security, but I doubt very much that they're hit rate is as low as 2/3 of people they are really looking at.

      Unfortunately, as expensive as this technology is likely to be, it's unlikely to be as expensive as providing proper training to TSA agents and the salaries to retain them.

    12. Re:2/3 is still not good enough. by Hylandr · · Score: 1

      Specifically, how will this software work when their face is covered with saran wrap and chilled from the water baths?

      - Dan.

      --
      ~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
    13. Re:2/3 is still not good enough. by sjames · · Score: 1

      One fine day in TSA training: No No, No! That's a gumwrapper, see how it smells minty? You want to flip the PENNY, that's the brown metal thing that's shaped like a doughnut without a hole!. Hold on, No Carl, you flip the penny, not the subject!...

    14. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 1

      I'd rather be an autistic than an obnoxious cunt like you.

    15. Re:2/3 is still not good enough. by gandhi_2 · · Score: 1

      yes but ~67% of the time, it works every time.

    16. Re:2/3 is still not good enough. by N1AK · · Score: 1

      Thanks for pointing that out to the 2-3 people who weren't aware of this and didn't assume the reason it is only 2/3rds accurate, as noted in the summary, is because of things like this.
      At 2/3rds accurate, assuming that's from proper testing it's already more reliable than standard polygraphs; as they're already used it doesn't bode well.

    17. Re:2/3 is still not good enough. by UnresolvedExternal · · Score: 1

      From TFA: In a real, high-stress situation, we might get an even higher success rate," noted Professor Ugail, who believes he'll eventually be able to detect around 90% of those who are lying, which is similar to the performance of the polygraph.

      The polygraph is 90% accurate?

    18. Re:2/3 is still not good enough. by flyneye · · Score: 1

      I just flip that little mental switch that reminds me there are infinite dimensions and anything I say is going to be the truth in one of them. If I believe it as I tell it, then it is true and so far no machinery has beaten me. I suspect this new garbage will perform no better. The funny part of sessions with a polygraph is the obvious control questions get downright creepy coupled with my answers.
      9 out of 10 psychologists will tell you , Fly is an honest guy. Suckers...

      --
      *Repent!Quit Your Job!Slack Off!The World Ends Tomorrow and You May Die!
    19. Re:2/3 is still not good enough. by AmiMoJo · · Score: 1

      Even better train yourself to do them all the time, or at least while they are asking the control questions. Lie detection is relies on detecting changes from the normal truth-telling state, so if you mess that up the whole test is invalid.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    20. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 0

      All that and more hedwards. If you want to see more on that, do a search on "lie detector test" and "polygraph". Watch the videos on how each trace of the graph can be controlled at will. The same will be true of video analysis of the face and of the "sophisticated brainwave scans".

      For my job, I had to undergo a polygraph. During the initial briefing the examiner said that the best value of such tests is that people believe they work. During the exam the examiner became very angry because I am unable to sit still for any length of time. He said that a cough, thoughts about anything but the question at hand, and bodily movement greatly corrupt the data. In the end, he said that my results were the worst he had ever seen in all his years as an examiner. Still, I kept my job so I must have "passed".

      Bottom line: lie detectors are a bunch of witchcraft. If you live with a coven of witches (go through "extensive training"), you will believe in witchcraft too. Heaven help us if the courts get snookered into accepting such "evidence".

    21. Re:2/3 is still not good enough. by Anonymous Coward · · Score: 0

      You can dilate your pupils on cue, or prevent it from happening? Interesting.

      As for the accuracy, keep in mind that people are about 54% effective in telling truths from lies unaided (Bond & DePaulo, 2000). If the 66% result suggested by the article is reliable, then it is clearly better than what people can do on their own (again, on average, but people are, on average, average). Whether that is "good enough to justify deployment" is more a matter of the cost of the device, I would think, as it is a clearly better accuracy rate.

    22. Re:2/3 is still not good enough. by repapetilto · · Score: 1

      It probably means that there is only a 10% chance of a false negative (in untrained subjects). Meanwhile false positives are through the roof.

  5. Practiced lying can defeat lie detectors... by blahplusplus · · Score: 3, Interesting

    ... all you have to do is memorize and rehearse lies in advance and imagine them and recall them as if they were memories. People get caught in lies because it's cognitively demanding to make it up on the spot unprepared.

    If you don't believe this consider religious faith. Many people I'm sure believe those falsehoods genuinely because they are well ingrained in their imaginations.

    1. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 2, Interesting

      ... all you have to do is memorize and rehearse lies in advance and imagine them and recall them as if they were memories. People get caught in lies because it's cognitively demanding to make it up on the spot unprepared.

      If you don't believe this consider religious faith. Many people I'm sure believe those falsehoods genuinely because they are well ingrained in their imaginations.

      In which case, it won't be lying anymore, because you genuinely believe it to be true.
      In my humble opinion, lying is only useful when you yourself do not believe the falsehood which you're trying to communicate, resulting in a situation of asymmetrical information. It's called deception, and is widely used in counter-intelligence.

    2. Re:Practiced lying can defeat lie detectors... by artor3 · · Score: 3, Insightful

      Great job weaving some bigotry and flame-baiting into an otherwise reasonable post.

    3. Re:Practiced lying can defeat lie detectors... by blahplusplus · · Score: 1

      "In which case, it won't be lying anymore"

      You still know its a lie so technically it is lying. The point is to get your biology to your mind to adapt in such a way as if you are recalling a valid memory hence the imagination portion of it. Since for many of us we're not taught to lie professionally or practice it. We do it mostly on the spot and while we can dupe others with our 'on the spot lying' the average person doesn't have infinite resources to spend considering whether something is true or a lie. Where as the article is really trying to catch 'everyday lying' catching people unprepared and offhand. Most of us don't have to deal with having such resources thrown at our everyday fibs since they usually don't involve serious risks, as those risks go up then practicing and finding away around it become more important.

    4. Re:Practiced lying can defeat lie detectors... by Black+Parrot · · Score: 4, Funny

      ... all you have to do is memorize and rehearse lies in advance and imagine them and recall them as if they were memories. People get caught in lies because it's cognitively demanding to make it up on the spot unprepared.

      From what I've read, you're supposed to randomly lie or tell the truth on the easy questions they ask at the start to gauge your response.

      If you don't believe this consider religious faith. Many people I'm sure believe those falsehoods genuinely because they are well ingrained in their imaginations.

      But all my religious beliefs are true; it's only other people's religions that are unfounded.

      --
      Sheesh, evil *and* a jerk. -- Jade
    5. Re:Practiced lying can defeat lie detectors... by Thunderstruck · · Score: 1

      Don't feed the troll.

      --
      Trying to use sarcasm in text-based forums does not work.
    6. Re:Practiced lying can defeat lie detectors... by A+beautiful+mind · · Score: 4, Insightful

      Saying religion is not some bronze age fairytale and coexisting with the scientific knowledge of the 21st century is something I find deeply offensive.

      It is time we stopped treating fairytales with undeserved respect or reservation.

      --
      It takes a man to suffer ignorance and smile
      Be yourself no matter what they say
    7. Re:Practiced lying can defeat lie detectors... by NonUniqueNickname · · Score: 1

      The Bene Gesserit are unimpressed by your strategy, or the article for that matter.

    8. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      Where's the bigotry? I'm not seeing it. Or is disagreeing with people now a form of bigotry?

    9. Re:Practiced lying can defeat lie detectors... by gatkinso · · Score: 1

      Doesn't really work if you don't know what the questions are ahead of time.

      --
      I am very small, utmostly microscopic.
    10. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      If only i had my mod points!

    11. Re:Practiced lying can defeat lie detectors... by blair1q · · Score: 1

      >From what I've read, you're supposed to randomly lie or tell the truth on the easy questions they ask at the start to gauge your response.

      Which is not calibrating the machine for what you think it is..

      It's not actually a lie detector so much as a fear detector.

      You don't fear being caught telling the lies you're telling at their request. They're not looking for a range, there. They're baselining the machine, and you.

      Later, when you really do start lying about the things they're going to nail you for, the needles will jump.

    12. Re:Practiced lying can defeat lie detectors... by RazorSharp · · Score: 1

      ... all you have to do is memorize and rehearse lies in advance and imagine them and recall them as if they were memories. People get caught in lies because it's cognitively demanding to make it up on the spot unprepared.

      From what I've read, you're supposed to randomly lie or tell the truth on the easy questions they ask at the start to gauge your response.

      From what I've heard, you're supposed to clench your asshole.

      http://www.youtube.com/watch?v=bScv6kfxRyE

      --
      "From the depths of my skeptical and rationalist soul, I ask the Lord to protect me from California touchie-feeliedom."
    13. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 2, Insightful

      It's not bigotry, it's the truth.

      Despite what your mommy told you, you have no "right" to not be made fun of for believing in an invisible sky-daddy.

    14. Re:Practiced lying can defeat lie detectors... by DRJlaw · · Score: 4, Informative

      ... all you have to do is memorize and rehearse lies in advance and imagine them and recall them as if they were memories. People get caught in lies because it's cognitively demanding to make it up on the spot unprepared.

      From what I've read, you're supposed to randomly lie or tell the truth on the easy questions they ask at the start to gauge your response.

      From what I've read, you're supposed to shut the hell up and invoke your right to be silent if you're being questioned about things you have done.

      Lying to public officials, especially federal officers, is in and of itself a crime. Lying gives officials facts which they can cross-check. Lying is something that ordinary people are generally bad at, and interrogators know how to get a suspect to move outside their pre-rehearsed alibis.

      Staying silent is not a crime. Staying silent does not allay an official's suspiscion, but cannot be used to convict you of a crime. Staying silent is something that oridinary people are generally bad at, but it's a hell of a lot easier to practice.

      Identify yourself, produce whatever ID you normally carry, and decline to speak about anything else unless you have carefully thought out what you are about to say, know that it is does not tend to indicate that you've committed some sort of crime, and know that it is the truth.

      The rules are quite different if you are being questioned about someone else or what they have done. But that's another story.

    15. Re:Practiced lying can defeat lie detectors... by HornWumpus · · Score: 4, Interesting

      All really good salesmen temporarily believe whatever bullshit they are selling at the time. It's kind of like method acting.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    16. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 2, Insightful

      Not bigotry. He chose an apt example that clashes with your worldview. He isn't intolerant of you having that worldview - he just disagrees with you and isn't trying to hide it. You, on the other hand, are being a bigot by attacking him because he doesn't believe in what you believe.

    17. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      Lying to public officials, especially federal officers, is in and of itself a crime.

      False. If a cop asks you where you were last night, and you say down at the corner bar when you were actually sleeping with your mistress, the act of lying to the cop isn't a crime. In rare cases, it might be obstruction of justice, but that's it.

      The federal officer thing is different:

      http://en.wikipedia.org/wiki/Making_false_statements

      but even then mere lying isn't sufficient, it must be a material lie.

      Now, you are always better off if you say nothing, but that's a different story.

    18. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      "From what I've read, you're supposed to randomly lie or tell the truth on the easy questions..."

      There's the biggest flaw.

      You don't talk to the police, ever!
      Then it doesn't matter what sort of recording device they have,

    19. Re:Practiced lying can defeat lie detectors... by DRJlaw · · Score: 2

      Lying to public officials, especially federal officers, is in and of itself a crime.

      False. If a cop asks you where you were last night, and you say down at the corner bar when you were actually sleeping with your mistress, the act of lying to the cop isn't a crime. In rare cases, it might be obstruction of justice, but that's it.

      Likely false. For example, ask Casey Anthony. Then read the statute under which she was convicted. Don't imagine that it's the only one. Are they questioning you concerning a felony? *DING*... you've committed a crime. Are you willing to bet that you're only being questioned concerning a misdemeanor? Because the police are not going to tell you that.

      but even then mere lying isn't sufficient, it must be a material lie.

      Materiality is a low bar. And if you read the linked statutes, you'll note that materiality is not necessarily required. If you're taking a lie detector examination, you're answering questions under oath. Public officials are not fools. You can study the entire criminal code of the jurisdiction that you happen to be in, as well as looking for specific provisions pertaining to taxation, permitting, and other topics where intentional inaccuracy is highly frowned upon, or you can assume that lying to a public official is generally a bad idea. ACs are big on giving weasel advice where it's not their butts on the line, but as the rest of us can see, taking your advice is a demonstrably bad idea.

    20. Re:Practiced lying can defeat lie detectors... by sjames · · Score: 1

      If you rehearse a story in your mind often enough, you can know it's a fiction, yet answer questions about it with confidence from memory, just like the truth.

      A lot of lie detectors and "tells" used by humans are really "thinking too hard" detectors.

    21. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 1

      Saying religion is not some bronze age fairytale and coexisting with the scientific knowledge of the 21st century is something I find deeply offensive.

      It is time we stopped treating fairytales with undeserved respect or reservation.

      People are always going to think and believe different things. Get used to it.
      If everyone thought and believed the same things, the world would be a very boring place.

      Why bother spending the energy being intolerant towards others? And more to the point, what good can come from it?
      You can't change their point of view any more than they can change yours. What is it that you're scared of about them?

      My last housemate was a Muslim guy - and I learned a lot from him and his culture even though I would never prescribe to his beliefs.

    22. Re:Practiced lying can defeat lie detectors... by IICV · · Score: 2

      Great job weaving some bigotry and flame-baiting into an otherwise reasonable post.

      If you think the OP is merely "weaving some bigotry and flame-baiting" into his post, then you have clearly never heard of the Doctrine of Mental Reservation, which is pretty much "this is how you should lie for Christianity". Someone well-practiced in that doctrine would probably be harder to detect, because as far as they're concerned they're not actually lying.

    23. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      Staying silent does not allay an official's suspiscion, but cannot be used to convict you of a crime.

      Not by itself, but "it may harm your defence if you fail to mention something which you later rely on in court" - at least in England.

    24. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      In the UK you have the right to remain silent, but it may be used against you lol

      In Europe, you have the right to not incriminate yourself under human rights laws.

      We all know the UK is the most tyrannical government in Europe, and has the most surveillance.

      Knowing the UK they will have this built into cameras on the street just as they have ANI (Automatic Numberplate Identification).

    25. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      So if you've committed a crime, then plead the 5th. What about if you're innocent? Can you take the test then? Why not?

      So if you take the test, you're obviously innocent and if you refuse the test, you're obviously guilty. They don't even need a real test, just administer a placebo test.

    26. Re:Practiced lying can defeat lie detectors... by AdamThor · · Score: 1

      you have no "right" to not be made fun of

      tell that to the guy in the UK who got arrested for deriding the suicide...

      --
      -- "Oh. This guy again."
    27. Re:Practiced lying can defeat lie detectors... by stackOVFL · · Score: 1
    28. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      I don't see why, they aren't real, and the method their truthsayers use to detect lies is not documented in the Dune novels.

      If you're going to make a reference to popular fiction, at least make it good or funny.

    29. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      it was clearly "weaving some bigotry and flame-baiting" into the post and to claim otherwise is disingenuous at best. Unfortunately this is a common theme on here on Slashdot. Its too bad too, because an otherwise interesting article and discussion can be turned into a childish war of words. There are many other non-divisive ways to illustrate the concept of "memorize and rehearse lies in advance and imagine them and recall them as if they were memories."

      Also, for future reference, The Doctrine of Mental Reservation was not practiced nor preached by Jesus and hence is not a Christian teaching no matter if the one using it claims to be Christian (or jewish or muslim or hindu etc) This idea can be used by the religious and non religious alike as "In strict mental reservation, the speaker mentally adds some qualification to the words which he utters, and the words together with the mental qualification make a true assertion in accordance with fact."

    30. Re:Practiced lying can defeat lie detectors... by Anonymous Coward · · Score: 0

      I'm perfectly used to the idea that different people have different beliefs. Doesn't change the fact that I'm right and they're wrong (at least, until such time as I hear a convincing argument that says otherwise). Again, I fail to see the intolerance in disagreeing with someone.

      And you're wrong, btw, people's opinions can change. People convert or become atheists all the time. It's not easy, and it generally doesn't happen overnight, but it happens. That's why religious ideas have to be open to free and frank, even robust debate, same as political or scientific ideas.

      Also, why assume that atheists are afraid of religious ideas? Fear of an idea is not the only reason to challenge it, I challenge religious ideas purely because they're wrong.

    31. Re:Practiced lying can defeat lie detectors... by Lost+Race · · Score: 1

      It doesn't matter what you did or didn't do. Police interrogations are almost always fishing expeditions to gather confession-like statements that can be used as leverage in a plea bargain. They are very good at twisting anything you say to look like you've confessed to some crime, perhaps not even what they arrested you for. Always exercising your right to remain silent makes these fishing expeditions less productive and discourages the police from routinely attempting to boost conviction rates by harassing and intimidating innocent people.

  6. Test it on Derren Brown. by Anonymous Coward · · Score: 0

    2/3 of the time? That means 1/3 of the time its wrong.

  7. Correct about 2/3 of the case? by Anonymous Coward · · Score: 0

    Even my wife can do better.

    1. Re:Correct about 2/3 of the case? by Anonymous Coward · · Score: 0

      that's what SHE said.

  8. 67% accuracy? by Anonymous Coward · · Score: 0

    Anyone know how that compares to a polygraph?

    1. Re:67% accuracy? by BlackTriangle · · Score: 0

      17% better than a polygraph, I'd recon.

    2. Re:67% accuracy? by ewibble · · Score: 1

      according to wikipedia http://en.wikipedia.org/wiki/Polygraph "90-95% validity by polygraph advocates" but it is disputed "psychologists estimated the test's average accuracy at about 61%"

      Anyway I think the advantage of this is not that it is more accurate its probably that you can set one up without hooking it up or even asking the person being questioned

  9. Coinflip by tuomasb · · Score: 2

    It successfully discriminates between truth and lies in about two-thirds of cases

    Slightly better than a coinflip. Just like normal lie detectors.

    1. Re:Coinflip by Anonymous Coward · · Score: 0

      It successfully discriminates between truth and lies in about two-thirds of cases

      Slightly better than a coinflip. Just like normal lie detectors.

      A coinflip does better than a polygraph........ which is just an interrogation device that can't detect lies.

  10. Graham Greenes "Travels with my Aunt" by OzPeter · · Score: 1

    There is a character in that book who had his maid wake him up every morning by saying something like "time to get up you war criminal", so that when the authorities would question him about actually being a war criminal, he was so inured by the accusation that it caused no reaction in him at all - so he could happily deny being a war criminal.

    --
    I am Slashdot. Are you Slashdot as well?
    1. Re:Graham Greenes "Travels with my Aunt" by Black+Parrot · · Score: 1

      There is a character in that book who had his maid wake him up every morning by saying something like "time to get up you war criminal", so that when the authorities would question him about actually being a war criminal, he was so inured by the accusation that it caused no reaction in him at all - so he could happily deny being a war criminal.

      So maybe that's why Mom used to always say, "Get up, you slacker!"

      --
      Sheesh, evil *and* a jerk. -- Jade
    2. Re:Graham Greenes "Travels with my Aunt" by blair1q · · Score: 1

      The author did not understand psychopaths at all.

    3. Re:Graham Greenes "Travels with my Aunt" by Anonymous Coward · · Score: 0

      Exactly, there are a lot of people, who actually believe their own lies. Then there are even more who lie daily about a lot of things, minor or major, day and night. For those people it won't work, or at best you'll get mixed results.

      Close counts for nukes and grenades, not 20 to life or death row.

  11. Quick! by olsmeister · · Score: 3, Insightful

    Get this ready in time for the Presidential debates.

    1. Re:Quick! by gewalker · · Score: 4, Funny

      Sad, but true. They tried this with an early prototype in 2008. Within 5 minutes it was engulfed in flames and set back the research by nearly 2 years.

    2. Re:Quick! by Anonymous Coward · · Score: 2, Insightful
      We've already covered Blade Runner, but that's OK. I was actually thinking more along the lines of 1984:

      "It was terribly dangerous to let your thoughts wander when you were in any public place or within range of a telescreen. The smallest thing could give you away. A nervous tic, an unconscious look of anxiety, a habit of muttering to yourself, anything that carried with it the suggestion of abnormality, of having something to hide. In any case, to wear an improper expression on your face, was itself a punishable offense. There was even a word for it in Newspeak: facecrime"

      Get this ready in time for the Presidential debates.

      Why bother? It would show nothing. Sociopaths can trivially pass these kinds of tests.

      Indeed, the greatest con men not only make their marks believe the lies - they believe the lies themselves, at least while they're being spoken. How do you think you get to become part of the Inner Party in the first place? Unsufficient to be doubleplusgood duckspeaker, must be doubleplusgood doublethinker.

    3. Re:Quick! by Anonymous Coward · · Score: 0

      Do you suspect one of them may be telling the thruth ?

  12. People used to do that by Osgeld · · Score: 1

    but I guess they do not make those anymore, lets spend billons on developing a machine that does the same job as a illiterate con artist can do with zero training.

  13. Of course... by stox · · Score: 2

    It will be illegal to use this on politicians.

    --
    "To those who are overly cautious, everything is impossible. "
    1. Re:Of course... by Black+Parrot · · Score: 4, Funny

      It will be illegal to use this on politicians.

      We already have a visual lie detecting algorithm for politicians: "Their mouth is moving."

      --
      Sheesh, evil *and* a jerk. -- Jade
    2. Re:Of course... by Anonymous Coward · · Score: 1

      Heh, actually I think they lie even when their mouths are shut.

    3. Re:Of course... by sjames · · Score: 1

      They don't need to move their mouths to lie. The smile is a lie covering a sneer of derision. The suit is a lie, the handshake is a lie. Any form of interaction at all is a lie.

      Behind the thin but perfect veneer of an upstanding everyman lies a monster.

    4. Re:Of course... by Holammer · · Score: 1

      This method is also compatible with lawyers.

  14. So... by Black+Parrot · · Score: 5, Funny

    It successfully discriminates between truth and lies in about two-thirds of cases

    Does that mean you might be found 2/3 guilty of a crime, or will they roll a die and send you to prison on 1-4 ?

    --
    Sheesh, evil *and* a jerk. -- Jade
    1. Re:So... by Anonymous Coward · · Score: 0

      Option 2 of course.

    2. Re:So... by Anonymous Coward · · Score: 0

      only if you roll a natural 1s on two 20 sided dice, then you get the max sentence. :(

  15. Now the Politicians will tremble by u-235-sentinel · · Score: 1

    because we'll know when they are lying.. Oh wait! I've mistakenly suggested they tell the truth now and then. Silly me :D

    --
    Has Comcast disconnected your Internet account? Same here. You can read about it at http://comcastissue.blogspot.com
  16. Lie detectors by Windwraith · · Score: 1

    I always wondered, how do lie detectors separate liars from awkward people? I know a lot of people with low confidence (including myself) that gets very nervous when talking to some people. I would assume even more nervous when being monitored or tested actively.

    How do operators of such devices define the proper thresholds for every individual?

    1. Re:Lie detectors by PPH · · Score: 1

      I've always wondered how lie detectors separate truthful people from sociopaths.

      --
      Have gnu, will travel.
    2. Re:Lie detectors by royallthefourth · · Score: 1

      How do operators of such devices define the proper thresholds for every individual?

      Fabricated on the spot based on the operator's whim because they are running a scam, of course.

    3. Re:Lie Detectors by royallthefourth · · Score: 3, Insightful

      Which, of course, raises plenty of important epistemological concerns: chiefly, what is a lie? Perhaps we could say it is speaking with the intent to deceive. But in what way is the speaker attempting to deceive, which pieces of information does he actually wish to conceal and which bits of misinformation are merely the detritus of a twisted story?

      Even if an actual lie detector were to exist, it would be up to the operator to decide what it means. Nobody is really prepared to deal with that sort of weighty thinking on a consistently sound basis, especially not a policeman or a judge.

    4. Re:Lie Detectors by Eskarel · · Score: 3, Interesting

      I suppose it depends on what questions you ask.

      It would certainly be open for abuse, but so is a pipe wrench. You could use a 100% accurate lie detector to invade someone's privacy for sure, but if you stuck to questions like "Are you planning on killing anyone today?" or "Did you kill that person?" you'd probably be fine. Especially since you wouldn't actually need to have a person in the room if you had a 100% accurate lie detector. You need a person now because interrogation requires instinct, but with a machine that could actually detect truth or lies and the right questions you could put someone in a room, have the machine ask them a preselected list of questions and then let them out if they're fine.

      You couldn't take people out of the equation entirely of course as you'd need them for answering "Why did you kill that person?".

      You'd definitely have to be careful about fishing expeditions, and with a much higher solve rate people would be a lot more careful about what they allowed to become law in the first place, but there's nothing inherently unethical about asking someone if they committed or are planning on committing a specific crime and being able to rely on the answer.

    5. Re:Lie detectors by R3d+M3rcury · · Score: 2

      I always wondered, how do lie detectors separate liars from awkward people?

      They don't. They are designed to detect "reactions" which may be due to feeling guilty about something.

      I took a lie detector test once when I was in High School. I was taking a "Behavioral Science" class and, one day, the police showed up to talk to us about "lie detectors" and how they work. One of the things he explained was that it doesn't really detect lies. It detects a reaction to a question. A skilled officer can use this information to consider areas of questioning that may lead to information.

      After the end of the talk, the officer asked, "Does anyone want to take a lie detector test?" I raised my hand, as did about half the kids in the auditorium. The officer said, "Okay, how many of you have driver's licenses." A bunch of hands went down. "How many of you have driver's licenses with you?" And I was the last man standing.

      So I went up on stage in the auditorium and they hooked up various things to measure my breathing and heart rate and such. He told me to tell the truth. And then he started in with his questions. "Is it true your name is John Smith?" "No," I replied. He makes a few marks on the paper showing my heart rate and breathing and such. "Is it true you're 17 years old?" "Yes" and a few more marks. He asks a few innocuous questions and I answer them, each time him making little marks on the paper coming out of the machine. Then he asks the big question:

      "Have you ever driven over the speed of 55 miles per hour?"

      Flashback! The previous night I had been out with some friends. I had an 11 o'clock curfew but had lost track of time. It was a little before 11 o'clock when I realized the time. I called my folks and told them that I was running late and they told me I had to get my butt home before 11. The race was on.

      I jumped in the car and started off. I hit I-91 and punched the gas, hitting about 80 MPH. When I got off the highway, I was doing about 60 down rural roads where the speed limit was about 40 MPH. I made it home right around 11 o'clock, meaning I wouldn't be grounded or anything. But I knew I'd done a stupid thing and could have gotten in to worse trouble if a cop had caught me. Fortunately, I hadn't gotten caught.

      So, as we return the present, here is a law enforcement officer asking me if I had ever driven above the speed limit. Do I lie? Do I tell the truth!? He told me to tell the truth! It's not like he could give me a ticket, could he?!

      "Yes."

      Okay, I told the truth. What's his next question going to be? Is he going to ask if I've ever driven over the speed of 75 MPH?! What do I do, what do I do?!?

      He asked a few more innocuous questions and then said, "Let's look at the results. Here's where I asked you about your name, here's where I asked you about your age. Here's where I asked you abou speeding. Your heart rate went way up! And, look, you stopped breathing! You didn't really start breathing normally for another 30 or 40 seconds after that question. This is the best reaction I've seen a long time!"

      Now, does the reaction show that I was lying? No, because I didn't lie. But my reaction could have been due to other reasons. Perhaps I lost a loved one to an accident due to a speeder and I'm incensed that the police could be asking me such a question. Now, again, a skilled officer would ask more questions in order to try to figure out what provoked such a reaction as the one I gave versus just asking a few more innocuous questions and calling it quits.

      So, no, the lie detector doesn't show lies. It shows reactions to questions. These reactions could be due to a number of factors, not the least of which is the idea that you might end up in jail for a crime you didn't commit.

    6. Re:Lie detectors by Anonymous Coward · · Score: 0

      How do operators of such devices define the proper thresholds for every individual?

      There are two approaches:

      1) Try to manipulate the subject into lying about something. For instance the interrogator would say: "We are trying to evaluate your patterns of behavior and your inclination to engage in criminal acts. Have you EVER stolen anything?" Measuring the excitement of the subject when responding (untruthfully) to that kind of thing gives the interrogator a baseline. Did the subject get more excited in the test guilty concience question than in the important question or less?

      2) Confront the subject with a specific information regarding the crime ("Did you leave that red pen on the table?"). In theory, an innocent subject should not be bothered by such questions while a guilty person should.

      In practice the first approach leads to false positives and the second to false negatives just a bit too often.

      Conducting research on 'lie detectors' is problematic because you need to control for the following variables:
       
      - the subjects must have a high personal interest in the outcome of the lie (i.e. if you can win 10 bucks from the scientist if you trick him does not produce the same results (level of excitement) as when you can be imprisoned for life if you fail to lie convincingly)
       
      - you must have a reliable independent way of determining whether the person lied or not (i.e. a guilty verdict from the jury is not always correct and if the 'lie detector' result was a part of the court process, it's not independent either)

      So to study the effectiveness of lie detection, you need to form an experiment where the lie is both high-stake for the subjects and near 100% independently detectable. I don't think such a study exists yet, but I'll gladly stand corrected.

    7. Re:Lie Detectors by janimal · · Score: 1

      How would this lie detector react to someone blaming themselves for a death, but being otherwise innocent in the face of the law and reason? They will say no, but will they believe it?

    8. Re:Lie detectors by gatkinso · · Score: 1

      There are very effective psychological screenings for sociopaths, polygraph not being one of them.

      While I was a contractor candidate and did not go through this, government candidates (to NSA) are subject to a psychological battery of tests (well they used to be anyway).

      I suppose if they find a sociopath they forward your resume along to the CIA. Who knows.

      --
      I am very small, utmostly microscopic.
    9. Re:Lie Detectors by Eskarel · · Score: 1

      That would be where well defined questions come in, and of course human judgment. No one is suggesting that such a machine should take over the entirely of the judicial process. Context is still important after all.

  17. I have Rosacea by Y-Crate · · Score: 1

    Control, true and false answers will all produce the same blushing response at random. Good luck with the rest of the population though!

  18. Another pipe dream by taustin · · Score: 1

    It successfully discriminates between truth and lies in about two-thirds of cases

    So it's even less effective than other "lie detectors" that don't work well enough to use for anything important.

    1. Re:Another pipe dream by Anonymous Coward · · Score: 0

      So it's even less effective than other "lie detectors" that don't work well enough to use for anything important.

      No, actually, two-thirds is considerably more effective than a polygraph.

  19. Pinocchio syndrome by tepples · · Score: 1

    We give our emotions away in our [...] wrinkling our noses

    Let me guess: someone's been hitting The Adventures of Pinocchio too hard.

    1. Re:Pinocchio syndrome by PPH · · Score: 5, Funny

      She sat on my face and asked me to lie to her. Honest.

      --
      Have gnu, will travel.
  20. Truth is rarely binary by izomiac · · Score: 1

    A "lie detector" of any mechanism relies on the assumption that there must be some physiologic or behavioral difference in a person telling a lie compared with a person telling the truth. Aside from how there has never been any data to support that assumption (despite how badly people want it to be true), the term "lie" is a human construction. For example, look at Presidents Clinton and Bush. Both were accused of lying. In the case of the former, the statement may have been technically true in a hyper-literal linguistic sense, but designed to deceive. In the case of the latter, the statement turned out to be false, but was believed to be completely true when it was spoken. (Note, I say these things for the sake of argument. I do not claim to know the objective truth for either of these situations, and there's still quibbling about both.)

    There is a grain of truth in every lie and a measure of inaccuracy in every truth. The delineation between the two is poorly demarcated and even humans can't agree about specific fringe cases. As any programmer knows, machines are extremely literal, so how can one possibly define a lie well enough for one to detect? Heck, how does a person's own body tell the difference and why would it bother? Short of a mind-reading device, you can't determine intent, and even that fails for some variants of lies (e.g. stating something you think, but do not know, as true). And what of the delusional psychopath whose thoughts aren't reliable in the first place? Or con men who specialize in fooling 3.8 billion years of evolution to detect their kind?

  21. Lie detectors are a lie by TiggertheMad · · Score: 1

    They don't and that is their dirty little secret. Nervous? Well, the machine says you are a liar. Have a nice life, now that it has been confirmed by technology.

    Go read up on how they work, its a fucking joke. The fact they are used by groups like the FBI is a national embarrassment.

    --

    HA! I just wasted some of your bandwidth with a frivolous sig!
    1. Re:Lie detectors are a lie by gatkinso · · Score: 1

      >> The fact they are used by groups like the FBI is a national embarrassment.

      Is it?

      Take it from a guy who sat in a little room at NSA for several hours hooked up to one: that process is damn effective.... the fact that the machine is a prop matters not a wit. The real lie detector is sitting behind the desk and isn't a lie detector at all: they simply get you to spill your guts.

      --
      I am very small, utmostly microscopic.
    2. Re:Lie detectors are a lie by Anonymous Coward · · Score: 0

      >> The fact they are used by groups like the FBI is a national embarrassment.

      Is it?

      Take it from a guy who sat in a little room at NSA for several hours hooked up to one: that process is damn effective.... the fact that the machine is a prop matters not a wit. The real lie detector is sitting behind the desk and isn't a lie detector at all: they simply get you to spill your guts.

      And what if you don't, then you're not a liar?

    3. Re:Lie detectors are a lie by Anonymous Coward · · Score: 0

      When the institution, the "operator" and the subject know the machine is bullshit it truly is an embarrassment.

    4. Re:Lie detectors are a lie by gatkinso · · Score: 2

      You will.

      If you don't, the test is "inconclusive" and you are retested until you do. Your song of past sin is then adjudicated and you are either passed or failed. After test number 5 or so they just reject you if you haven't sung.

      Plenty of applicants, especially in this economy.

      --
      I am very small, utmostly microscopic.
    5. Re:Lie detectors are a lie by gatkinso · · Score: 1

      Well. Then I would suggest that you give it a whirl, and we'll see how smug you are afterwards.

      --
      I am very small, utmostly microscopic.
    6. Re:Lie detectors are a lie by cheekyjohnson · · Score: 1

      You will.

      Except if you don't, of course. It really depends on the person.

      --
      Filthy, filthy copyrapists!
    7. Re:Lie detectors are a lie by Windwraith · · Score: 1

      Oh man that sucks. Airport sec or something like that?

    8. Re:Lie detectors are a lie by HornWumpus · · Score: 1

      Sure you confess to something, duh. 'I failed to report the gain I made buying and selling an old car in 2004, it's been on my mind for years. Sob, I feel so much better with this off my chest.'

      But that night in college when you did all that cocaine with those improbably good looking (to be partying with you) Russian girls? You just leave that out.

      The pot garden in your yard and all the associated unreported income? I didn't do it, nobody saw me do it, can't prove anything.

      All that free cable and satellite TV you've enjoyed? No need to mention that.

      That time you and your coworkers took all the fractional pennies from the interest calculations? You leave that out too.

      Your time on 4chan\b? Needless to say.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
    9. Re:Lie detectors are a lie by Anonymous Coward · · Score: 0

      Airport square.

    10. Re:Lie detectors are a lie by Anonymous Coward · · Score: 0

      Your subjective experience only understrikes the bullshit. You believe it worked because of the impressive dude (did he have a white overall and order you to shock someone by any chance?) when the actual machine has proven over, and over to be useless garbage.

    11. Re:Lie detectors are a lie by gatkinso · · Score: 1

      True... but in that case it really isn't that big of a deal. They simply reject you.

      However I bet you would. Apply. Try it.

      --
      I am very small, utmostly microscopic.
    12. Re:Lie detectors are a lie by gatkinso · · Score: 1

      Nice try, but they specifically ask if there is any drug use you haven't reported on your SF-86.

      My examiner flat out said the most common tactic is to confess to something small to throw them off the trail of something big.

      The machine may be stupid, but the examiners are not.

      --
      I am very small, utmostly microscopic.
    13. Re:Lie detectors are a lie by gatkinso · · Score: 1

      Go for it man! If you have the background to be selected for screening, then give it a try. I tell you, I am not the only one to come out of that room feeling head raped.

      And, if you will take the time to read my posts, I said the machine is indeed BS... but the process of interrogation works.

      NSA flat out admits this in a congressional report, which you can look up if you care to type a few lines into the search engine of your choice.

      --
      I am very small, utmostly microscopic.
    14. Re:Lie detectors are a lie by Anonymous Coward · · Score: 0

      Nice try, but they specifically ask if there is any drug use you haven't reported on your SF-86.

      My examiner flat out said the most common tactic is to confess to something small to throw them off the trail of something big.

      The machine may be stupid, but the examiners are not.

      I think the whole point of this is that people in that situation will tell them things because........ they want the job.
      Of course a criminal investigation would go much differently.

    15. Re:Lie detectors are a lie by HornWumpus · · Score: 1

      Of course the examiner is the only working lie detector in the room.

      Doesn't change the fact the machine is basically useless and it is simply a battle of wits.

      How many times did those two notorious arch traitors (who's names escapes me) pass polygraphs while selling intelligence to the Ruskys? Dozens each anyhow.

      --
      John McAfee 'It was like that time I hired that Bangkok prostitute; to do my taxes, while I fucked my accountant'
  22. Self Reference by Anonymous Coward · · Score: 0

    Did the good Doctor say all this facing his invention?

  23. Oh, Two thirds of the time huh? by __aailob1448 · · Score: 1

    Wake me up when it gets to 99%.

    Unless I'm mistaken, 66% accuracy is ridiculously shitty.

    1. Re:Oh, Two thirds of the time huh? by gatkinso · · Score: 1

      The fact that it is better than chance is quite remarkable, actually.

      If it actually performs as stated that is.

      --
      I am very small, utmostly microscopic.
    2. Re:Oh, Two thirds of the time huh? by RazorSharp · · Score: 1

      I think that by chance a precise 50.0000000000% success rate would be more strange than a 66% one.

      If I flip a coin thirty times and get twenty heads, would you presume that the coin is unfair? Would that really be so statistically bizarre? No. A precise 15 out of 30 would be much more bizarre, despite that there is a 50% chance of receiving a head on each flip.

      I'm not saying that this device isn't measuring real physical reactions in some way, I'm just saying that they seem to correlate with truth/lies no better than chance guesses would. Basically, it's useless.

      --
      "From the depths of my skeptical and rationalist soul, I ask the Lord to protect me from California touchie-feeliedom."
    3. Re:Oh, Two thirds of the time huh? by Nethemas+the+Great · · Score: 1

      Hey it's better than TSA's record. Let them ask the questions "are you a terrorist?", "do you intend to perform acts against the interests of the United States?", etc. while being monitored by this system. It surely beats the voyeurism, and sexual molestation. If they really have to behave as sexual predators let them, but only on those individuals that have first failed the Q&A session.

      --
      Two of my imaginary friends reproduced once ... with negative results.
    4. Re:Oh, Two thirds of the time huh? by Anonymous Coward · · Score: 0

      Somebody was asleep during statistics class....

    5. Re:Oh, Two thirds of the time huh? by Anonymous Coward · · Score: 0

      It is better than what an unaided human can do. We average around 54% accuracy on telling truths from lies (the results from a meta-analysis, Bond & DePaulo, 2000). The only study of which I'm aware that shows unaided people doing as well as the claimed 66% cited in the article is a result from a study by Ekman & O'Sullivan (1991) in which US Secret Service agents did better (64% accuracy) than other tested groups (college students, psychiatrists, judges, polygraphers, robbery investigators) and better than chance (50%).

  24. God says... by Anonymous Coward · · Score: 0

    C:\TEXT\BIBLE.TXT

    t faith.

    5:13 And withal they learn to be idle, wandering about from house to
    house; and not only idle, but tattlers also and busybodies, speaking
    things which they ought not.

    5:14 I will therefore that the younger women marry, bear children,
    guide the house, give none occasion to the adversary to speak
    reproachfully.

    5:15 For some are already turned aside after Satan.

    5:16 If any man or woman that believeth have widows, let them relieve
    them, and let not the church be charged; that it may

  25. Press and Statistics by rechtco · · Score: 1

    Unclear what 2/3 success rate means. Need to know false positive and false negative rates. Does 2/3 mean of those I said were lying? In which case how many liars did I miss? If there are 1000 liars in the sample, and I say 100 are lying, of which 66 are liars that is 2/3 correct. However, I missed 900 liars and called 34 truth tellers liars. Or does 2/3 mean of those actually lying? In which case how many truth tellers did I include to find the 2/3 of the liars? If there are 1000 liars and 1000 truth tellers in a total sample of 2000 and I say 1666 are liars of which only 666 are liars that is 2/3 666/1000 liars correct and 1000 truth tellers are called liars and 334 liars are missed. Or If I say 999 out of the 2000 are liars and only 666 are liars that is 2/3 666/999 correct with 334 liars missed and 333 truth tellers called liars..

  26. Oh, immigration, of course by Rogerborg · · Score: 1

    Having been through US immigration recently, I can attest that "Fear can sometimes be the fear of not being believed rather than the fear of being caught." is putting it mildly. The TSA goon managed to misinterpret me so many times that I started to doubt my own story. He was either a genius, or a tard, but either way he didn't seem even remotely human, so if we can replace him with a very small shell script, go to it.

    --
    If you were blocking sigs, you wouldn't have to read this.
  27. 66% accurate? by Anonymous Coward · · Score: 0

    OMG, Flipping a coin will be 50% accurate. That tool only costs 25 cents (actually it is only a deposit, becsue you can spend the quarter after you use it to prove someone's guilt). This new system is only 16% better at determining a truth than my quarter. I wonder how much do you have to pay to get that 16% improvement?

  28. I sincerely hope NSA switches to this system by Anonymous Coward · · Score: 0

    Because if they do, I am reapplying. If by slim chance I get to be screened with this thing, I will tell the truth.... again.... and when I pass I will scream at those motherfuckers "I TOLD YOU I WASN'T LYING YOU GODDAMN ASSHOLES!"

    Then I will go back to my job at NASA Goddard, feeling much better that I got that off my chest.

    1. Re:I sincerely hope NSA switches to this system by Dunbal · · Score: 1

      No - the way society seems to be headed you will be thrown in jail for not complying with your mandatory test. After all causing a scene like that means that you must have something to hide. All they need to do is hold you on "suspicion of terrorism" and you're gone indefinitely. Is it 1984 yet?

      --
      Seven puppies were harmed during the making of this post.
  29. Lie Detectors by RazorSharp · · Score: 1

    Even if we were able to create a 100% accurate lie detector, would using it be moral?

    I'm not sure, but I have doubts.

    --
    "From the depths of my skeptical and rationalist soul, I ask the Lord to protect me from California touchie-feeliedom."
  30. Disorders by Anonymous Coward · · Score: 0

    I would imagine that individuals with anxiety, social or other disorders or insecurities would skew results so, where's the test to determine who is the best candidates? I know if I felt I was being accused, assessed or examined for truthfulness, I would more than likely have a subtle guilty responses all over, it's just the way I am and I'm sure others who are hyper sensitive and perceptive, may have a similar reaction.

    That said though, 66% accuracy is a joke. Unless it's close to perfect, it's only maybe relevant enough to determine who may have taken that candy bar.

  31. Even if you're innocent by ronmon · · Score: 1

    You should refuse to give statements to police and prosecutors. Your words will be twisted in order to convict you. They are much more interested in winning a case than finding the truth.

    1. Re:Even if you're innocent by Anonymous Coward · · Score: 0

      It's because they've convinced themselves you did it. Anything that leads to others believing what they already do is good for them.

  32. Lie Detectors easly defeated by old scam by Anonymous Coward · · Score: 1

    If George Costanza has taught us anything, it's not a lie if you honestly believe it, well at least to polygraph...

  33. TV already has a series on this by Anonymous Coward · · Score: 0

    The TV series "Lie to me" is based on this technology, plus a whole lot more
    http://www.imdb.com/title/tt1235099/

  34. Needs more accuracy by tsotha · · Score: 3, Funny

    Two thirds? Most women can do that without any fancy equipment.

  35. Great by Dunbal · · Score: 1

    It successfully discriminates between truth and lies in about two-thirds of cases

    Yeah, it only detects when normal people are lying and it doesn't work at all when sociopaths, psychopaths and habitual liars are lying - because these people feel no shame when they lie or might even believe their lie to be true and thus don't have a vasomotor response. But sure, feel free to re-invent the polygraph. Only you are not addressing the basic flaw in designing the "lie detector machine": the people who are really good at lying are most likely the ones you want to catch, and by definition they are also the people you are not going to catch.

    Academics can be so brilliant and so dumb at the same time. But hey I hear that if you measure people's skulls and classify their shapes into groups, you can also tell who the liars are. Give it a shot!

    --
    Seven puppies were harmed during the making of this post.
  36. Much ado about nothing by Anonymous Coward · · Score: 0

    Thermal polygraph systems have already been shown to be substantially worse than trained interviewers (see paragraph below) and, as previous people have already noted, ~66% accuracy is essentially useless in a realistic context. Some of the text in the BBC article is rather misleading, too. For example, the statement "the thermal sensor even detects changes in blood vessels" implies that this type of detection can be performed easily and reliably across multiple individuals, which isn't the case unless the vessels in question are extremely close to the skin (for instance, on the top of the hand). In my experience (using a FLIR 600 series thermal camera), whether or not it's possible to obtain clear thermal images of blood vessels on the face is very much dependent on the person being imaged and usually isn't possible.

    A good paper that addresses the use of thermal imaging as a lie detection tool in an airport setting more realistically is "Thermal Imaging as a Lie Detection Tool at Airports" by L. Warmelink, A. Vrij, S. Mann, et al (Law and Human Behaviour, 2011) (despite LHB not being a technical journal, the paper includes sufficient technical detail to be credible). That paper concludes by saying:

    "Based on our analysis of the NRC report and the present (baseline) data, it is unlikely that thermal imaging can be used effectively at airports as a general screening device. Thermal imaging will classify too many non-deceptive passengers as liars because, as we outlined above, they may be anxious for non-deceptive reasons. In addition, our data showed that, prior to being interviewed, thermal imaging did not register differences between truth tellers and liars."

    Overall, it's interesting research, but it's nowhere near the point of being sufficiently effective to be useful in an operational context and it's really rather irresponsible of the BBC (and the researchers whom they interviewed) to portray it as such.

  37. I can make it more accurate. by ron-l-j · · Score: 1

    0 = false 1 = false There now its a hundred percent accurate. Just in time for the elections.

  38. Which Desert? by Anonymous Coward · · Score: 0

    Well, how come I'd be there?

  39. What about false positive? by Anonymous Coward · · Score: 0

    As with all reporting on statistics, this story is missing the most important thing. So it successfully knows when someone is lying 2/3 of the time, but what about the number of times it falsely accuses someone? That's the most important issue in this case.

  40. Lost getting from A to B by LostMyBeaver · · Score: 1

    If you're still reading AC, how exactly did you deduce that he's ruining it for others?

  41. I'm with you on this by Anonymous Coward · · Score: 2, Interesting

    I personally grew up in a house where I was guilty until proven innocent and proving myself innocent was considered to be mouthing off.

    In my subconscious I believe that it doesn't matter what I say, I'll be considered guilty no matter what. As a result, I've taken on a superior attitude towards the interviewers in these circumstances. I have real skills, a real education and a real job. I don't have to work as a rent-a-cop in an airport to support my habits. Yes, it's being an asshole, but I refuse to let people who are working a power trip career that was a result of being a bully and a dick throughout their youth to bully and be a dick to me now that I've worked to get to where I am at. Oddly enough, this seems to keep them from being a dick to me since bullies only bully people when they believe they can win without real resistance. There are thousands of "whimps" waiting in line... why bother with a difficult one.

  42. Lies, damned lies and ... by Anonymous Coward · · Score: 0

    [q]It successfully discriminates between truth and lies in about two-thirds of cases[/q]

    That's about 10% better than a coin flip. What do you have left when you run it through the Bayes theorem?

  43. The scary flaw with lie detection by Anonymous Coward · · Score: 0

    You get a guy who is more than a little conscious of the fact that he's being subjected to an unavoidably flawed lie-detection system whose results are going to make or break his interview, and his emotional state will more than likely trip alarms even when he's telling the truth.

  44. 2/3 is not much better than a trained human by Anonymous Coward · · Score: 0

    Let us be honest, decerning truth from lie should be at worst, a 50/50 affair. If you train yourself you can probably go above that. A machine doing a 66% by looking at some phisiological sign , which may or may not be correlated to lie, is the same as the infamous polygraph : a piece of shit.

  45. Advantage "Did Too" by retroworks · · Score: 1

    And thus ended remakes of Monty Python's Dead Parrot Sketch.

    --
    Gently reply
  46. Running a trace now by ThatsNotPudding · · Score: 1

    And even without training if it's only good 2/3 of the time that's not good enough to justify deployment.

    'Classify as terrrorist / terrorist supporter.'

  47. Hello Sociopath by handy_vandal · · Score: 1

    The world's most perfect lie detector will fail to detect the world's most perfect liars.

    Worse, the machine will assure us that those who are not liars, are good and honest men and women.

    And we're supposed to trust this machine?

    --
    -kgj
  48. H. Beam Piper - Little Fuzzy by jimwelch · · Score: 1

    Had a Lie Detector, that measured your brain waves. It was used routinely in court. It is no longer under copyright, so it is free (or should be) on most of the e-readers or on the web.

    --
    Never trust a man wearing a coat and tie!
  49. Deckard will be most pleased by Amtrak · · Score: 1

    So they finally invented a device for detecting those tricky replicants.

  50. If you act nervous on purpose? by Holammer · · Score: 1

    Will the system be able to determine if someone acts guilty on purpose by following a laundry list of nervous behaviours. I'd love to subjected to a controlled test and take it for a ride.

  51. What about drugs? by swb · · Score: 1

    Mild sedatives like Xanax? Beta blockers? Anti-depressants? Anything to make the emotional dynamic range lower?

  52. Ye who approaches the by stackOVFL · · Score: 1

    I.R. lie detecting camera of death must answer me these questions three, or a full body cavity search ye shall receive!

  53. hmm by Anonymous Coward · · Score: 0

    so I guess psychopaths are safe to keep lying.