The Eyes Have It

Feelgood writes: "Yahoo is carrying a Reuters report that thermal imaging may be used in airports to detect liars. Shouldn't be a problem that 1 out of 4 liars will get away and 1 in 10 innocents will be incorrectly nailed." There's a UPI story about the lie detector possibilities and a blurb in Nature. From the UPI article, the inventor has a good appreciation of the ethical considerations. Will anyone else care?

Comments

Liars

[i_am_nitrogen]

Can liars really be detected by thermal imaging? I think they're lying.

Re:Liars

[spongman]

i have to agree with this guy.

too often we see people complaining about rejected posts after the same subject matter has been brought up by someone else later on.

I'm not grinding my axe here but I've reported a few stories that have been rejected, only to appear much later.

Are we free?

[Mean_Nishka]

My buddy Ben Frankline summed this up the best: They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.

Once more, for all the slow JBT's.

[jcr]

I would like to point out that while Aldritch Ames was in the process of getting a whole lot of US agents in the Soviet Union killed by ratting them out, he continued to pass his polygraph tests.

There's no such thing as a lie detector. Polygraphs are voodoo, and so is this.

-jcr

Re:Once more, for all the slow JBT's.

[geekoid]

Show's what you know, they're worse then voodoo. ;)

People look at voodo and see a doll. they're familiar with dolls, and know its not possible.

People look at a polygrapgh and see needles and paper and wires all being run by some clown who's "certified".

I find it amazing that people still rely on them, when over and over again, in lab conditions there shown to:
a)be "defeatable" with little training.
b)the results can be interpetted diferently by different "professionals"
c)return false results(I was a victim of this once)

Re:Once more, for all the slow JBT's.

[4n0nym0u53 C0w4rd]

This technique has been used for years in psychology studies (and by bad cops on naive suspects). It's called the "Bogus Pipeline"[1]. The basic idea is that, given a quick demonstration of its effectiveness (instruct the target to tell a couple of lies, and show that the machine detects them), along with an incentive to avoid being caught (if you're honest with us, we'll be easy on you), people are more likely to behave honestly.

[1] Jones, E. E., and Sigall, H., "The Bogus Pipeline: A New Paradigm for Measuring Affect and Attitude," Psychological Bulletin 76: 349-364, 1971.

Re:Once more, for all the slow JBT's.

[coyote-san]

Polygraphs are fairly successful at intimidating the generally honest, but the hardcore criminals and foreign agents know they're worthless.

Worse, there are enough idiots in high places who prefer to believe in magic than science that they'll redirect the efforts away from hard evidence that points to the real bad guys, and focus instead on the innocents who just had a bad day that first time around. Many people consider this the most damaging side-effect of the belief in lie detectors - few LEAs can afford to waste time chasing down false leads or ignoring real ones.

The best analogy isn't to seatbelts, it's to a flawed airbag that deploys during slow collisions (giving many drivers unnecessary black eyes and broken arms), but which regularly fails to deploy in high-speed collisions where you need it the most.

Re:Once more, for all the slow JBT's.

[GigsVT]

People look at a polygrapgh and see needles and paper and wires all being run by some clown who's "certified".

Like an IIS server farm? :)

Forget airports...

[Anonymous Coward]

...politicians around the globe should be subjected to lie detector tests at regular intervals :P

Schneier said it

[coltrane99]

On the impact of seemingly acceptable success rates on large-scale systems here

What happens next

[adamy]

OK, 1 in 10 get A false postive. In a plane full of 280, that means that 28 people are going to be detained....I think not.

I guess if this was used as part of comprehensive screening process it might be useful....anyone who fails the test has to walk past a bomb sniffing dog or something.

Of course, the terrorists are going to be training to pass the lie detector test, so it probably won't help catch them.

Re:What happens next

[Howie]

Who needs training?

Q: "Are you a member of a terrorist organisation?" (as it says on the green US INS Visa Waiver form [*])

A: "No" (thinking: "I'm a freedom fighter", and therefore telling the truth).

Even if it were 100% accurate, it may not help.

[*] also containing 'Moral Torpitude' - my all-time favourite phrase on a government form.

Re:What happens next

[jeffy124]

the equipment needs better testing. thus far they have formally tested 20 people (if you read the article, you'd know), which is far much too small a sample. Plus the fact that something like 8 of them were told to lie intentionally, which isnt easy for a person to do. 6 of those 8 were found to be lying, which is worse than the rate of false positives.

Good testing would involve a very large sample, probably on the order of 1000s, of people who were willing to face an odd question, to which the response remains confidential (or even destroyed). After questioning, the subject gives a card to the proctor telling of whether their answer was truth or a lie, to be compared with the imaging.

Odd questions should be things people are generally uncomfortable with, like "Have you ever visited a porn website?" Think about how you would react to such questioning face to face with someone.

In addition, the person(s) reviewing the tape of the subjects in front of the thermal imager should not have audio or any other clue toward the question/answer being asked/given. They would base their judgement solely on the person's facial temperature. The cards would be used to match the correctness of the reviewer's decision.

Other things like nervousness to enter the equation, which is normal. Someone may have temperature increases even if they're telling the truth. In polygraphs dummy questions like "Is your name _____?" come up to establish a baseline of the subject. Thermal imaging may also require that, which would then defeat the purpose of "rapid examination" for airports and border crossings.

Are some people complaining a bit too much?

[krogoth]

I understand all the complaints about privacy, but when was the last time you implmented a security measure on any computer that you expected to be invincible? Does that mean we should give up on computer security just because it occasionally inconveniences us?

Re:Are some people complaining a bit too much?

[digitalunity]

I agree with most of what you say. However, with respect to the amendment you quote:

This applies only to the government. You don't have a right to board a plane. It is a privledge they've granted you. Planes are private property. You also don't have to be subjected to any security measures they have set up for screening. They may deny you access to the plane, but that's your problem. They could say "We'll only fly passengers wearing pink socks". They wouldn't get much business, but they could do that if they wanted to. It's their plane, it's also their right. The airline industry is a private industry. The government mandated security requirements are only the minimum. There is nothing preventing the industry from responding with even more stringent standards than the government is imposing.

To determine how involved the government is, they weigh the rights of the private industry against the rights of the people that cross the path of the airplane with the security risks in mind. With the danger involved in operating an airline in hand, I agree that whatever security measures they choose to implement are fair and reasonable.

Re:Are some people complaining a bit too much?

[bnenning]

With the danger involved in operating an airline in hand, I agree that whatever security measures they choose to implement are fair and reasonable.

Really? You preemptively agree that anything they do is ok? To take an extreme example, what about strip-searching all passengers? Rational people can and do disagree about which specific methods are reasonable, but handing a blank check to the authorities is never a good idea.

Re:Are some people complaining a bit too much?

[digitalunity]

what about strip-searching all passengers
Fine with me. If they wanna shoot themselves in the foot; they can. People won't ride the planes if they go too far. I'm only arguing they have the right to do whatever they think is necessary. But, another comment has brought attention to the federalization of security. Being a government run operation, they do have limitations now.

handing a blank check to the authorities is never a good idea
Agreed. It's good we're only handing it to the airline industry.

I only fly a few times a year. However, I'm getting happier with the security. Delays Shmelays. That's what I say. I think they should do whatever it takes to make flying safe.

I'm willing to relinquish my right to privacy to fly on an airplane. I only wish more 'rational' people would do the same.

Re:Are some people complaining a bit too much?

[bnenning]

However, I'm getting happier with the security. Delays Shmelays. That's what I say. I think they should do whatever it takes to make flying safe.

Fair enough. Like I said, people can disagree on the proper balance between privacy and security. But I'm concerned when people altogether stop considering the balance and assume that anything done in the name of security is automatically good. That sort of thinking is what allowed the Patriot Act to pass with virtually no debate on its several questionable provisions.

The test

[ocie]

Look into the lens, now please tell me in single words only the good things that come to mind about your mother...

This is frightening...

[Bagheera]

Considering the absolutley abysimal record of the polygraph in controlled testing (references are extensive) this is just something else we don't need. Relying on an "automatic" system is just asking for more "false faith" in a security system that doesn't work.

The article states that it's proven as effective as the existing polygraph - which is to say its reliability sucks.

Just what the world needs. Another knee-jerk deployment of a technology "to make us feel better." I suspect it'll be as effective as the National Guardsmen standing on the end of the big bridges - only far more intrusive if you happen to be one of those 10% false positives.

Re:This is frightening...

[jimbolaya]

I don't really think this is so frightening. Granted, it's record is not outstanding, but the technology is not going to be used to detain, charge, or convict anybody. If they determine somebody may be lying, they'll just be subject to additional search and scrutiny. For the 10% false positives, this will be nothing more than a minor inconvenience.

Re:This is frightening...

[Sc00ter]

I agree with you. And besides? What's the alternitive? To have Joe Loser "randomly" descide to harass people? I'm sure his reliability on judging to detain the correct people is great,,

Re:This is frightening...

[roman_mir]

I lived in Israel for a year. They have airport security second to none, they have professional people working in the gates, professional enough to see who is who, yes it is possible to do given enough experience. In 40 years they did not have a single incident due to a security failure either in airport nor in the air. El Al (air travel company with their own planes) has a number of agents on each flight, these people are trained and are carrying weapons that can be used in the airplane. Now, imagine, if their security was like in something in the States or in Canada, what would happen to their planes, I mean in Israel there are suicidal bombings at least once every couple of months...

Re:a sad state

[erroneus]

clear sign of a highly paranoid society.

It's not just a product of a paranoid society... in fact, I would go so far as to assert that your assessment is rather incorrect when speaking in generalities. Reality seems to suggest that the public's paranoia is directly linked to the intensity of the situation as portrayed by the news media. (The people are mostly "Amber Grain" and the media is the wind... and together, we get Amber waves of grain...)

The reality is that we live in an extremely LITIGIOUS society and if it can be shown that "they didn't do enough" to prevent this or that from happening, it makes one or several rich and happy widows/ers... not to mention rich and happy lawyers.

I believe that is where the ridiculous measures are coming from -- legal CYA activities, not paranoia or an interest in protecting public interest. Since WHEN has any corporate entity ever been interested in public good?

The marketing dept. wants one

[maladroit]

In Interface, by Stephen Bury (aka Neal Stephenson and his uncle), an inventor's small, portable polygraph is put to a truly noble use: Marketing.

Stephenson's got the right idea about how something like this would be used - marketing droids would flip over getting 80% honest responses in their focus groups - it beats anything they see currently. Somebody's probably making plans for the mall kiosk right now.

25% and 90%?

[anthony_dipierro]

Shouldn't be a problem that 1 out of 4 liars will get away and 1 in 10 innocents will be incorrectly nailed.

Most metal detectors probably let 1 out of 10 get away and incorrectly nail 1 out of 4. Hasn't stopped them from using it as one of many screening methods.

Time for a new Continental Congress

[perdida]

It's time for a new Continental Congress.

That is a gathering where citizens decide on a new constitution. Sort of a constitutional convention.

The government is, in this case and many others, taking responsibility for things it has no right to control.

Either we must stop the government from violating the SPIRIT of the 1st and 4th amendments, or we make a new Constitution without these freedoms.

We do have the right to abrogate these freedoms, to voluntarily give up our right to free speech and against search and seizure, but we can't give them up and "swear to uphold and defend the Constitution" in the same breath!

Re:Time for a new Continental Congress

[_Mustang]

err - no.

That whenever any Form of Government becomes destructive of these ends, it is the Right of the People to alter or to abolish it

It says abolish, which means a hugely different thing from overthrow.

Overthrow is definately NOT abolish.

Re:Time for a new Continental Congress

[shogun]

True, you would probably have to invite the Canadians and Mexicans to make it a continental congress...

Will anyone else care?

[Raetsel]

Yes, they will... but how?

• Jon Katz will love it for all the material it'll generate -- just think, a whole new "Hellmouth" series!
• The (FBI | CIA | NSA) will love it because it'll allow them to assemble a biometric database of iris/cornea patterns.
• The average "Joe Citizen" will accept it because it's for protecting him from those nasty, evil terrorists.
• The (Taliban | Hezbolah | someotherfoamingidiot) will practice so they can defeat it.

• and...

• Everyone reading this comment will worry about the consequences of a false positive happening to them.

This comment has been a knee-jerk reaction. We now return you to your normal thread.

Most of the tagged people will be innocent.

[enkidu]

10 million passengers. 10 bombers.

"Hello, do you have a bomb?"

"No."

Result: 1,000,000 innocent people incorrectly tagged as "liars". 8 bombers correctly tagged as "liars". Even with an order of magnitude improvement in accuracy, 100,000 innocent (easy blushing) people, 10 bombers. Of course, if they just use it to pick out people to do a detailed x-ray/explosives inspection of the bags, then it might help, supplemented with additional random searches of course. Unfortunately, most airports don't have any bomb-detection equipment installed yet, so only a hand search by incompetent security is available.

Let's face it. To get real aircraft security is going to cost a hell of a lot of money. Current airport security is a joke. Poorly thought out rules being implemented by semi-trained personnel with the cheapest possible equipment. We can't incrementally improve the existing security structures and expect that to work. In the end, we're all going to have to pay for bomb-sniffing machines/dogs properly trained security personnel, and have the whole thing organized and tested (continuously). Then I think it would be possible to make getting a bomb on board a plane at least two orders of magnitude harder than it is now. Of course, I'm among the few that think that flying is still safer than driving to the airport.

Re:Most of the tagged people will be innocent.

[mosch]

that'd be retarded. fatalities per mile is what you really want.

Airplanes have 0.7 deaths per 100 million aircraft miles.
Cars have 1.7 deaths per 100 million vehicle miles.

Re:Most of the tagged people will be innocent.

[streetlawyer]

No, fatalities per mile is a statistic rigged to favour aircraft, which only make long journeys. Fatalities per journey would be the more relevant statistic, as it answers the question that people want to know; how likely am I to die if I get into this thing? On that basis, aircraft look significantly worse than cars.

Re:Most of the tagged people will be innocent.

[Puk]

No, fatalities per mile is a statistic rigged to favour aircraft, which only make long journeys. Fatalities per journey would be the more relevant statistic, as it answers the question that people want to know; how likely am I to die if I get into this thing? On that basis, aircraft look significantly worse than cars.

You can keep getting more and more detailed in how you break down the question, and get different results. If you're going this far, you should take the next step and break it down by distance.

In fact, the safety swings with distance. The longer the trip, the better aircraft look. On longer trips, taking an airplane is safer than driving. For shorter trips, the reverse is true (imagine flying to and from work every day). I don't know at what distance the switch is made, but I'm fairly sure it falls in the area in which airplanes do make common flights (which ranges from intercity to intercontinental).

-Puk

Re:Most of the tagged people will be innocent.

[enkidu]

Mr. Atta was not a starving sheep farmer from Afghanistan. He was a middle class student from Egypt, whose parents were more than able to pay for him to go to Hamburg, Germany, Europe and get an education ... another 14 of the hijackers were from Saudi Arabia, not the poorest country in the world.

I didn't say economically and politically oppressed families but countries. Saudi Arabia is one of the most politically repressed countries in the world. And Egypt isn't Singapore by any stretch of anyone's imagination. In fact, Saudi Arabia actually encourages (or maybe that should be encouraged) anti-american fundamentalism to flourish because it took (some) of the heat off of the corrupt government.

I stand by my statment.

Fine, deploy the tech, but...

[e40]

There has to be serious compensation when the test fails. That is, when I go to the airport and I fail whatever "test" they give me and I'm "detained" for a few days, I want some serious cash as a result. Let's say $100,000 or more. That'll make Big Brother think twice about testing me.

Comment removed

[account_deleted]

Comment removed based on user account deletion

How Dumb

[mESSDan]

Quick, someone patent "Ice glasses" and make a mint selling them to terrorists, or people who want want to keep their privacy! Hmm, or what about skin colored patches that you put on your temples that slowly release chemicals that bring the blood flow to your eyes DOWN.

Or gee, the poor man's way of getting around this, buy a soda before you go, right before you go, you unobstrusively rub ice from your drink around your eyes and face, then wipe off the excess moisture.

Re:How Dumb

[roman_mir]

OK, this would be really dumb. If you had patches like that then right from the start, without asking any questions at all, do you know how ridiculous your face would look on a thermograph? HA! You would have blue, green or black circles instead of eyes. That alone should set off the alarms!

Consider the math -

[(void*)]

Most slashdotters will come to the conclusion that if they caught someone, the chance of actually catching a liar (not a terrorist - just someone with a reason to lie) is still vanishing small. This is true, assuming that the number of liars is small.

But there is another more interesting possibility to consider - if you want to catch a liar in this circumstances, there would be a better chance of catching a liar if you look for those that pass the lie-detector's test.

Just think about that.

Re:Consider the math -

[roman_mir]

Right, so that you would not care about 8 bombers out of 10 because they look like they are lying but would only apprehend the other 2, and out of 100 innocent people, 90 will be detained for questioning :)

Re:Consider the math -

[(void*)]

Don't argue with the math. I know it is silly. Thank goodness there are OTHER ways to check for criminals, besides using a polygraph.

This is only a test.

[exceed]

Folks, let's calm down here before we get too rational. This method is only a "field test." What makes us think we will be prosecuted based on a blush? There would be further interrogation, testing, and harrassment (if it even goes that far) before charges were brought up on anyone using this method.

While I don't think this is very reliable; polygraphs give MUCH more "feedback" based on factors other than a blush, I don't believe this system is going to be used as a sure sign someone is guilty (especially with it's accuracy ratio).

Suggestions:

[AnalogBoy]

1: Line planes with bacon, or, more humanely, put wilber the famous flying pig in the terrorist-class section of the plane. (Which raises an interesting, if tacky, question.. Since they won't be using those frequent flyer miles anywhere else.. do terrorists fly first class?]

2: Strip search everyone from young, suspicous Abu Bin Confused to old lady Theresa Boobsahangin.

3: Stun guns under every seat.

4: Seperate section for screaming, annoying kids and their apathetic parents. (Okay, I admit.. this is more for my sanity).

5: Bomb-sniffing dogs. Mean ones. With the metal-tipped teeth, inlaid with gold, "F" and "U" on each canine.

6: Corrolary to 2, Naked flights, (seperated by age class for sake of sight)

7: Alien-esque automatic weapon. Pilot puts plane on defensive mode, gun shoots anyone not seated and buckled. Not feasable, but a fun idea.

8: Did i mention naked flights?

9: Flood cabin with nitrous oxide, chloroform, ether, or some other anasthetic gas. Only fresh air comes through pilots mask - Pilot breathes or everyone dies. Not being a scientist, i have no idea how those gases would act at that altitude.

Re:Suggestions:

[sharkey]

Pilot's opening speech (heard on Bob & Tom):

"Welcome to United Filght 101. Just to reassure you on our commitment to your security, all flight attendants have been replaced by the starting offensive line of the Green Bay Packers. If a person does get out of line, rest assured that THEY WILL HANDLE IT.

Second off, we in the cockpit are in full communication with our attendants at all times. If a terrorist does stand up, they'll let us know up here, and we'll put this baby into a nose-dive, pinning the him to the back of the cabin, then let our flight attendants "deal" with him.

Third, our snack today is bacon and beer. If the person sitting next to you does not eat all his bacon, and drink all his beer, he is a terrorist. Please let our flight attendants know about him.

Thank you, and enjoy your flight!


(Best as I remember.)

Re:Suggestions:

[roman_mir]

I've being a vegetarian for the past 7 years (only vegetables, fruits, nuts. Raw) and I don't drink alcohol, I mean ever. So by your defenition, I am a terrorist! Yahoo.

Phil Dick Lives

[glassware]

Tyrell: Is this to be a capillary dilation test? Involuntary reaction of the iris? The so called Blush Response?

Deckard: We like to call it Voigt-Kampf for short.

The correct way to use this technology

[MBCook]

If they're smart they will use this the way that do/should other technologies. It should be used to help them spot POTENTIAL liars. It should help them figure out who to watch more closely. It should not be the be-all-end-all test for such a thing. This way that one liar will still (hopefully) get suspected, and most of those wrongly flagged "innocents" will be realized as such. But to use this without thinking would be like giving random people drug tests where everyone ate poppie-seed muffins before hand and then watering down ever sample with a gallon of water. It just wouldn't be usefully accurate. But to use it to aid judgement instead of replace it would be the only correct way to do it.

It's sounds better than a polygraph.

[AnotherBlackHat]

Polygraphs are considered about 75% accurate, which sounds good until you consider that flipping a coin is 50% accurate.

If they can really catch 3 out of 4 liars, and "avoid" 9 out of 10 innocents,
(which is what the article claims inventors claim) then it's much better than 75%.
If 1 in 100 people are "liars" then this would be nearly 90% effective.
Which again sounds good until consider that identifying everybody as innocent would be 99% accurate.

On the plus side, this might make wearing eye shadow a crime under the DMCA.

Polygraphs can be beat simply by putting a thumb tack in your shoe,
and stepping on it during the "little bad" questions and not during the "big bad" question.
(saying that probably makes this post a violation of the DMCA ...)
I'd bet that this device can be beat by a similar method.

Re:It's sounds better than a polygraph.

[Saithier]

If they can really catch 3 out of 4 liars, and "avoid" 9 out of 10 innocents, (which is what the article claims inventors claim) then it's much better than 75%. If 1 in 100 people are "liars" then this would be nearly 90% effective. Which again sounds good until consider that identifying everybody as innocent would be 99% accurate.


Using those numbers it is actually far far less than 90% effective. If 1 in 100 are liars, that means 10 in 1000 are. You have two populations to work with 990 innocent people, and 10 who lie. The test is 90% accurate on the innocent, which means it false reports 99 people as liars (10%). The test is 75% accurate on liars, so (rounding up) it reports 8 of the 10 liars as guilty, and 2 as innocent. This gives us a total of 107 people reported as liars, when only 8 of those actually are a rate of 7.47% accuracy. And that doesn't count the 2 liars it missed!

Cheers!
**Saithier

Old Joke

[Detritus]

It reminds me of the old joke about the psychiatric patient who flunks his polygraph test when asked if he is Napoleon Bonaparte. He answered "No".

What is the machine really measuring?

Fun with mathematics

[BlueWonder]

Let's assume that one out of a million people is a terrorist and will lie when tested with the device. This means that in a group of a hundered million people, 100 liars exist.

If the device identifies a liar with 75% success rate, 75 out of the 100 liars will be found. On the other hand, if the device misidentifies 10% of the truth-tellers as liar, 9,999,990 out of the 99,999,900 truth-tellers will be misidentified.

Therefore, under these assumptions, if the devices indicates someone as a liar, the probability that he's actually lying is 75/(75+9,999,990), roughly 0.000749995%.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Malingering

[webword]

If you are interested in this topic, I suggest that you look into malingering. Detection of deception has been well-researched in cognitive psychology. I actually spent a few years working in a laboratory where people we researching this topic.

One of the coolest things I read about was a study where people would be hooked up so that event-related brain potentials (ERPs) were detected for malingering. In effect, your brain gives you away. For example, if you saw a video with some information and then you were asked about it, your brain does a little "hop" which can be detected with ERPs. It didn't matter how well you lied or how convincing you were, you would be detected. Supposedly, the methods works extremely well. However, you can't expect people to accept this. Would you like to have an electrode cap put on your head?

(Ah, you have to love science.)

By the way, you might want to check out these resources:

The Journal of Credibility Assessment and Witness Psychology

Forensic Psychology and Forensic Psychiatry

Polygraph Law Resource Page

Re:how does this do anything?

[digitalunity]

Would a Lunatic pass a Polygraph if he really believed what he said?

Yes. The causes for which a lie is detected are things like respiration and heart rate. These things wouldn't be affected by someone who isn't nervous about what they're saying.

And so, in the airport:

[tunah]

Security guy: Are you carrying a bomb?

Terrorist: Yes

Security guy: Well, the machine says you're right, but it would say that for 25% of liars, so i'd better double-check. Are you a terrorist

Terrorist: Yes

Security guard: Thanks sir, move along.

Re:And so, in the airport:

[KjetilK]

Actually, this isn't too far off... When I, I guess as a foreigner, enter a flight to the US, "Are you, or have you been involved in terrorism?" is the type of question you've got to answer.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Same Theory as Polygraph

[xonker]

This sounds like it works the same way as a polygraph test -- if you have a physical response to lying, basically caused by a panic response then the machine can detect it. If you're a sociopath that doesn't have that response...nada. Polygraphs don't detect your guilt or innocence, they detect your reaction to the question. I'm reasonably sure that if I shot someone, but felt good about having done it, that I'd be able to sail through a polygraph (or this test) with flying colors.

If the paranoia continues we'll all be flying naked without carry-ons in a few months. (Perhaps the airlines would issue something similar to hospital gowns...) On the plus side, being surrounded by naked people might help me with my phobia of flying. I've flown since 9/11 and I'm still more worried about a wing falling off than I am about terrorists...

Cool idea, dude!

[nytes]

Wouldn't it be great for a candidate to show up at a press conference to find one of these things, perhaps along with a breathalyzer, sitting on the podium?

Re:how does this do anything?

[damiam]

Probably so, but even a religious fanatic isn't going to believe that there's no bomb in his bag when he put one there a few hours ago.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Small numbers ~= B.S.

[mark-t]

Exactly... 20 people in a control group is utterly inadequate to demonstrate even proof of concept, let alone actually be ready for real world use. Those same percentages might be marginally more acceptable if they had a control group a hundred, or preferably a thousand or more times that size.

What does ANNOYANCE look like

[symbolic]

to the camera, as in, "I'm so FREAKING annoyed that I'm being subjected to this nonsense."

Why do I think it's nonsense? Have they tested this on any Al-Qaeda members? Remember - these people have gone through extensive training, and there's no reason tha such training in the future can't include something that addresses this kind of interrogation. Another possibility- what if there was a drug whose effect could render this test completely useless?

Some theories on how to beat systems like this.

[AgTiger]

No, I don't know exactly how well some of these would work, but I figure if I can come up with more than 5 ideas off the top of my head in a few minutes, then how hard would it be for anyone seriously bent on beating this system to find and perfect a way of doing so?

Here goes:

1. Buy a thermal imaging camera of sufficient sensitivity so that you can see your own reactions, and learn how to modify them (feedback, negative or positive, does allow one to learn).

2. Know the questions being asked in advance. Practice giving rote answers to them so you're no longer thinking about the meaning of the question when it's actually asked, much like we no longer think about how exactly we tie our shoes.

3. Practice lying and learn not to give a damn about the fact that you're lying. In essence, practice becoming somewhat sociopathic. (Gee, shouldn't be too much of a stretch for a terrorist!)

4. If the expected answer is "no" (are you a known or suspected terrorist?), before answering, think of a question in your own mind to which the correct answer is no, and ask it of yourself before audibly answering "no".

5. Throw the baseline off before you even get close to the camera - get drunk enough to bring a flush to your entire skin but not so drunk that you are obviously impaired.

6. Like 5, get drunk, but don't stop at 'non-obvious'. Make it very obvious that flying scares the b'jeez out of you, and the only way you (especially after 9/11/2001!) and the only way you're getting on one now is if you're suitably numbed/happy.

7. Inhale a little powdered black pepper up the nose just before walking up to answer questions. The sneezing fit should throw off your reactions nicely. Blame it on allergies or a cold.

8. Take an emotion levelling drug before you get anywhere near the airport - the type that leave you not really caring about much. Surely you know a friend or two who has some psych-based drugs in their regimen of prescriptions...

9. Make like you have a toothache. Dig something sharp into your side through a pocket (a sliver of sharpened wood? A pencil?) to cause pain while being asked the questions such that your body's reactions are different.

10. Make like a person with a mild (or severe) disability, either mental or physical. Our social training has engrained that these people are "invisible", and that they CERTAINLY should not be unduly hassled, as that's cruel. An interesting physical choice might be "deaf mute", where you hand over a card asking the person to write what they want to say or ask on the small pad of paper you conveniently have with you. You write your answer as a response. So much for the instantaneous flush of heat from the eyes... You'll be looking down at a piece of paper, and will have time to "cool down".

Okay, not only five points, but ten. Much like physical locks only keep non-determined innocent people out of where you don't want them, this method will only catch nervous, embarassed, unprepared people, and thus is nothing more than the illusion of security. *sigh* It doesn't stop the really determined people, and those are the ones you wanted to catch, darn the luck.

Re:Some theories on how to beat systems like this.

[vorgriff]

"10. Make like a person with a mild (or severe) disability, either mental or physical. Our social training has engrained that these people are "invisible", and that they CERTAINLY should not be unduly hassled, as that's cruel."

Your comment brings up a side point: what about all the people who have disabilities who are on medication? The number of medications total that could cause problems for this machine is huge.

Re:Some theories on how to beat systems like this.

[zulux]

10. Make like a person with a mild (or severe) disability, either mental or physical.

This would work now! I was waiting for 45 minuets in the Tampa airport security line this holiday, and over 5 obese people we weal chaired to the front of the line and allowed though after a helper stuck their crary-ons though the detectors. The fatties themselves weren't even hand checked with a wand because their wheel chairs were metal. These people we so fat that they could have stuck a bomb in one of their fat-rolls and nobody would notice.

Re:Some theories on how to beat systems like this.

[Monkeyman334]

2. Know the questions being asked in advance. Practice giving rote answers to them so you're no longer thinking about the meaning of the question when it's actually asked, much like we no longer think about how exactly we tie our shoes.

I already do this one.

Did your bag leave your possesion after you packed it?

No

Did anyone ask you to carry on their bags?

No

They're going to start having to ask questions like "Are you not not a terrorist?", "Uhhhh, no?"

Comment removed

[account_deleted]

Comment removed based on user account deletion

Real Stats

[FFFish]

Here are real stats.

In summary, accidents -- fatal and non-fatal -- are on the decline in the airline industry. There were six accidents for every 100,000 hours of flight time... and that includes all those piddling little one- and two-seater private craft.

Take a look at real aircraft, those that operate on schedule and carry more than a handful of people, and the rates are very impressive: 0.4 accidents for every 100,000 departures. (It is a little unnerving that the rates are on the increase, though!)

Finally, at the bottom of the last table, we see that there were only five suicide/bomb crashes during the eighteen years between 1982 and 2000. There were 147,577,440 departures. That's an attack rate of sweet fuck-all (0.00000339% for those that really need the number).

In short, there appears to be no real good excuse for spending a pile of money on increased security measures. The risk-cost factor just doesn't justify it. Yes, there should be better security measures; but, no, they shouldn't be costly.

IMO, YMMV, IDFM (I don't fly much).

Re:Real Stats

[dragons_flight]

The biggest flaw with your analysis is that the historical trend is unlikely to describe the future trend in this case.

Given the number of people in and sympathetic to terrorist organizations, I would bet they could get enough volunteers to bring down several planes a year, assuming security couldn't stop them. The direct cost of such a situation would be horrible for us, and secondary costs on morale would be worse. That is the risk one has to counter, not the historical legacy that it doesn't happen. After all, prior to 9-11, history would tell you that jumbo jets are never used as missles.

In my mind the potential risk does justify considerable expense. Now that expense should make sense, no argument there. After all at some point it would be cheaper simpler to tell people that they can have no carry-ons and must strip and put on airline issue clothing. A step even further, you could have two planes fly every route, one for luggage, one for people. Then even if there is a bomb in someone's luggage, only the pilots die. (Of course, who wants to live in that world?)

We can make flying safer, and there is IMO some justification for spending a significant sum to do so, but I agree that it has to be real security and not the illusion of security that we pay for.

Alone it might not be worth much...

[Greyfox]

But combine it with microexpression detection and voice stress analysis (Which your financial institution may already use) and you might just have a winner.

Been There - remember the "voice" lie detector?

[Seth Finkelstein]

The article says:

"This is the first technology that allows lying to be measured or lying to be detected without any contact with the subject whatsoever instantaneously, in real time," said lead researcher James Levine, an endocrinologist at the Mayo Clinic in Rochester, Minn. "You don't need to hook them up to anything -- you don't need any sophisticated experts to analyze the data."

Everyone seems to have forgotten Voice Stress Analysis which was once similarly hyped as real-time, no-contact, super-duper lie detection. And where is it now? In fact, it was better, since you could supposedly apply it to a tape-recording, and there's even VSA freeware you can run on your own PC (have fun).

Remember, stress is a matter of the body, but a lie is a matter of the mind. They're correlated in many people, but by no means identical. Just think, do you know any smooth-talking liars (i.e. ones displaying minimal stress)?

Sig: What Happened To The Censorware Project (censorware.org)

Best Behavior

[cloudscout]

This will require everyone in the airport to be on their best behavior. Thermal imaging will be able to 'see' when you pass gas. No more blaming it on that poor sedated dog in the travel-kennel.

Re:No they DON'T!

[J. J. Ramsey]

However, if one sets off a metal detector, neither the security personnel nor the other people at the airport necessarily assume one is carrying a weapon. After all, it's a *metal* detector, not a weapon detector, and most everyone knows it. A so-called lie detector ostensibly tests for lying, though, and so security personnel and others at the airport may assume that setting it off means that one is lying, even if one isn't.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Comment removed

[account_deleted]

Comment removed based on user account deletion

Apprecation of Ethical Considerations

[Nathdot]

From the UPI article, the inventor has a good appreciation of the ethical considerations.

Ask him about his appreciation of the ethical considerations with the machine switched on...

"erm..."

:)

What numbers would you prefer ?

[tmark]

Shouldn't be a problem that 1 out of 4 liars will get away and 1 in 10 innocents will be incorrectly nailed.

So you don't like the S/N ratio implied. What numbers would make you happy ? 1 out 100 liars get away; 1 in 1000 innocents incorrectly accused ? Higher ? Because if you're looking for something with no Type I and II errors, you will be looking forever - ANY system you can imagine will ALWAYS falsely accuse innocents and miss the guilty.

Re:There is plenty of cost justification.

[FFFish]

I'd buy all that except that there hasn't been an increase in security anyways!

A lunatic with a C4 shoe-heel got on a plane just last week.

I drop my wife off for a business flight (same day as the jet in NY went down in the harbour, ugh) and as she's standing in a mile-long line-up... "Attention: all passengers for Flight 123, please proceed through the entrance gate immediately." The damn flight was getting held up, so they just waved (waived?) everyone through! Apparently, terrorists just have to wait for the final, ultra-final boarding call. WTF?!!

We have people being detained for reading Alex Haley's "Airport," FBI agents being detained for having dark skin, and no real security against Cessna's loaded with ammonia and diesel.

It's all a crock of shit.

The next terrorist attack against the USA isn't going to be done with a big ol' jet airplane. Too risky.

I'm sure that, were it not outright dangerous to speculate on terrorist tactics vis-a-vis getting one's ass nailed to the wall by the FBI under the new, draconian laws passed by a panicked Congress, you and I could come up with a dozen surprisingly effective attacks, none of which involve airliners, nor could they be prevented by any amount of security.

Re:There is plenty of cost justification.

[FFFish]

Oh - and a further note:

There were 63 airline accidents resulting in fatalities in the years 1982 through 2000.

Compare that against the five bombings/suicides, and one thing is immediately obvious: reducing accidents by a mere 10% will have greater effect than eliminating terrorism.

Achieving the former is both possible and relatively cheap. Achieving the latter is impossible, and to even partially achieve the latter is terribly expensive.

Even more so, reducing automobile accidents by 1% would save more people than any amount of anti-terrorist measures.

Let's deal with reality: terrorism isn't much of a threat against American life and property when compared to things that we accept every damn day -- driving, smoking, eating Cheetos, and walking downstairs.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Comment removed

[account_deleted]

Comment removed based on user account deletion

Startle responses

[coyote-san]

Oh joy. I guess nobody remembered that it's well documented that many people have an exaggerated startle response. It's not a matter of fear, or drug use, or fatigue, or anything else other than some messed up wiring in our brains.

Note the word "our" - I know about this because I have it. If somebody knocks on my door, and I'm expected them, I'm fine. If someone knocks on my door and I'm not expecting anyone, my heart is pounding (elevated pulse and blood pressure) for an hour.

I, and others, can often compensate for this by becoming hyperaware of our environment. If you catch movement out of the corner of your eye, you aren't as startled when they suddenly make a loud noise. But in a busy environment like an airport, we're always stressed out.

Once a false positive, always a false positive?

[Kris_J]

Those stats look bearable until you realise that it probably catches the same people as false positives every time. If implemented, it would probably make 5-10% of the travelling population's lives hell.

AOPA member eh?

[Robber Baron]

An AOPA member...might've known! Weren't you guys the bastards that railed against installing mode-C transponders because despite the obvious safety enhancement (preventing mid-air collisions between large jets and some fuckwad that wandered into restricted airspace) for no good reason other than you figured that it might prevent you from exercising what you felt was your god-given right to fly whenever and wherever the hell you felt like?

Leave your plane alone? I'm staying the hell away from all of you...idiot kamakazes and clueless techno-spook wannabees and taking the train!

crappy effacy never stopped urine testing

[Wansu]

Shouldn't be a problem that 1 out of 4 liars will get away and 1 in 10 innocents will be incorrectly nailed.

1 in 20 wee wee tests are either false positives or false negatives. So, if it's a false positive, they retest the sample with the more expensive gas chromotagraphy mass spectrometry to validate it. At least, they are supposed to. :-) However, we're talking about lies here, not pee pee. There ain't a more accurate test.

Re:There is plenty of cost justification.

[FFFish]

Then use a frigging Lear Jet. The point is that no one's likely to be using a commercial jetliner for that job ever again.

The new security doors, for starters, are going to see to that.

Whoa there, calm down a second...

[asteinberg]

Now I definitely agree with everyone criticizing this system, but lets not jump to conclusions. The Yahoo article definitely does NOT definitively say that these will be used in airports. Slashdot's summary says "... may be used ...". The word "may" implies speculation, and for once it seems Slashdot's summary isn't so far from the truth. To be sure, let's examine the exact wording from the article:

Scientists have developed an instant lie detector technique which picks up mini hot flushes around the eyes and could [emphasis added] lead to truth tests becoming standard at airport check-ins.

I could just as easily say "scientists have developed a new kind of extra-powerful gun, which could lead to people using it to kill each other" (err perhaps I should have chosen a less touchy example...) How about this: "Linux makes a computer so powerful that a user could use it to do malicious things." You get the point here. Moving along in the article...

Polygraphs have long been considered [emphasis again added] for increasing security at airports but current technology, which links tiny changes in heart and breathing rates to blood pressure and sweating, takes too long to process, making it impractical for large numbers of people.

Okay, so they have considered polygraphs before, and they opted not to use them. I don't see any quotes from high-ranking airport/security people saying that this is definitely going to be implemented. So, they might think about it, but hopefully they'll come to the same conclusion I think the majority of us came to - this is FAR too inaccurate to justify its use.

Of course, with all the post-9/11 hysteria I guess I wouldn't be too surprised to see them try to implement something this ridiculous, but lets just keep in mind that they haven't decided to do that yet (of course, I do not advocate that we stop arguing).

too sublte

[SlashDread]

"Shouldn't be a problem that 1 out of 4 liars will get away and 1 in 10 innocents will be incorrectly nailed."

Irony too suble for non-techies, so just imagine:
In Amsterdam Schiphol Airport, a medium sized european airport, there are 450.000 annual flights in and out. Say each plain carries about 250 passengers. An even 1m passengers in/out a year.

In Amsterdam alone we are going to point to 300 misidentified Lyars per DAY.

What do we do with them? Slap em on the wrist? Make em write "I shall not lie" 1000 times?

Gr /Dread

False positives

[KjetilK]

Shouldn't be a problem that 1 out of 4 liars will get away and 1 in 10 innocents will be incorrectly nailed."

Unless that person is you! Just think about it: With a plane with a 100 people on board, 10 will be incorrectly flagged as "liars", and what's worse, the person behind the desk who is going to decide whether or not to let you on board or have you put in front of a military tribunal and shot, will have no clue as to where to start. The only thing they have is that you blushed when asked a specific question. I bet you were just looking down her ..., you bastard! :-)

There is no way you can deal intelligently with all those false positives (contrary to a metall detector, where you can find out very fast exactly what caused the alarm).

Besides, take the scary option that they will actually record who were detected as liars, you'll get an incredible amount of data then to be cross-checked with a lot of other databases, and make a lot of people subject to criminal investigation wrongly.

Besides, I really doubt they will catch any terrorists this way. They only way you can achieve security against terrorists is to eliminate the desire to commit terror, and you can only do that by emphasizing human rights for everyone.

Re:False positives

[Artagel]

Sure, there will be false positives. You have to consider the value of a screening technique to the alternatives. Currently random spot checks are being done at the gate, and virtually all of those checks are "false positives" and will remain so.

If the spot checks are done on every tenth person, you have a 10% chance of checking a terrorist walking up. If you have a screening method that is 80% to pick out the terrorists, and will result in checking every tenth person then you have no increase in false positives, but a factor of eight increase in efficacy.

Public safety is not going to be founded on some amorphous public policy change. There is no conceivable change that would have appeased bin Laden. I don't see that the Unabomber or McVeigh would have been easily dissuaded by such things either. A free society is always going to piss someone off with the choices it makes. Rational, respectful, effective policing is essential to maintaining a free society, and as events of the past year demonstrated, essential to keeping a free society safe.

Re:False positives

[sporty]

There is no way you can deal intelligently with all those false positives (contrary to a metall detector, where you can find out very fast exactly what caused the alarm).

Of course there are ways. Most basic way is to do a background check, which can be done quickly provided the infrastructure is in place and a quick baggage check.

I think you are overreacting about recorded liars etc etc... If they are lying about say, murder, then there's prolly good reason it is on their record in the first place as well as the fact that they are lying.

Besides, its not expectant of eliminating terrorist acts, just reducing them as technology for terrorism surpass the technology of airport security.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:There is plenty of cost justification.

[Saeger]

It's psychology, stupid -- it's the rare person who thinks things through logically.

When death and destruction is concentrated in one big horrific event (like 9/11), it registers as a huge blip on a lot of radar screens; it becomes threat #1 because it's so visible. But when death and destruction is spread out over space and time (like auto accidents, liver disease, or falling down stairs), its easy to miss because it's so diluted and hence your risk SEEMS so much less.

So, the cost justification for useless "anti-terror" tools is in order to calm the irrational human beast down enough for business to continue as usual. NOT providing the illusion of security is the real risk here.

--

Re:Unfortunate baggage search

[RFC959]

And that's part of the problem. Even if the science is valid, to be useful by airport screeners, it has to be boiled down into a device that basically says "TERRORIST" or "OK". You lose all the subtlety that might be in the science of the device so that a high-school dropout can operate it quickly.

Re:There is plenty of cost justification.

[alexjohns]

You have some very good points. For some reason these terrorists had it in for the World Trade Center and (more obviously) for the Pentagon. If I was a terrorist, I wouldn't have bothered with the Pentagon. I would have taken out a nuclear power plant. Far wider reaching consequences. Way harder to clean up. Much more devastating to the continued health of a nation. (The consequences here would be far-reaching, including what if we decided to get rid of all nuclear plants to remove future threats? Greatly increase our dependence on foreign oil.)

Also, what's the average number of people killed in a plane crash? 100? Here we have 4 planes with 3000 dead. 750 apiece. And if we only consider the WTCs, 1400 apiece. That's 14 of the accidental crashes rolled into one.

We can decrease deaths in a whole bunch of ways: outlaw tobacco; all cars older than 5 years go to the scrap heap (so you always have the latest and greatest safety devices); outlaw all guns (90% chance that if you die by gun, it's someone you know, not some random criminal); mandatory healthy diets and exercise for everyone. :)

It's possible for us to live more safely, but there's a limit to how safe I want to be. Airlines have an interest in making their planes safe. If that 10% improvement comes at the cost of doubling ticket prices, are you willing to pay that? The odds already seem pretty good to me. What about a further 10%? Double it again? Pretty soon only millionaires would be flying in super-safe planes. Although, if it was possible to make super-safe planes, why wouldn't the millionaires have them now? Maybe planes are already about as safe as we can make them.

Re:There is plenty of cost justification.

[armb]

> There were 63 airline accidents resulting in fatalities in the years 1982 through 2000.
> Compare that against the five bombings/suicides, and one thing is immediately obvious: reducing accidents by a mere 10% will have greater effect than eliminating terrorism.

No. Eliminating 10% of the accidents would give you a bit over 6 fewer fatal accidents, which is more than the 5 suicide attacks. But how many of the fatal accidents destroyed large buildings full of people? How many of them killed hundreds of police officers and firefighters?

Auto accidents are another matter. Just think, if thousands of Americans got their fat butts out of their SUVs and on a bike occasionally, they'd be less likely to kill anyone, less likely to die of a heart attack, use less gas, and be fitter if they ever did have to help tackle a terrorist on an airplane.

Seems fair, actually

[MrResistor]

Well, compared to polygraph, anyway. IIRC polygraph has a 99+% chance of detecting a lie, but about a 40% chance of false positive (saying someone is lying when they aren't).

So, compared to that, a 1 in 10 chance of false positive doesn't seem so bad...

Blushing

[knuth]

Maybe they were embarrassed.

Did you stab a mannequin?

Um . . . .

And did you then rob that mannequin?