Detecting lies, how hard can it be?

[Noel_Prosequi]

One of the issues about studies trying to find if people are good at spotting liars is that the studies have to use standardised “lies” rather than real life ones in which the actual content of the lie itself is useful information. The example in John Cleese’s show had people watching one of two videos (a nature documentary or a horror movie, for example) and then either describe what they were actually watching or pretend to describe the other. People who could only see the TV watchers, not the screens, were asked to pick who was lying. The “lie” was necessarily entirely plausible in context.

It is clear that these tests are designed to see if people can spot liars using only meta information like breathing, eye movement and so on. But most judges look at the actual content of the story and its likelihood to judge lies as well.

For most liars, there are details of the story that jar, or don’t sit well with the way the world works, or are improbable or seem contrived, etc. The identification of that sort of thing is overlooked in these studies, but it is a very powerful part of the process.

Similarly, it is very difficult to test for lie detection in circumstances where the people are truly invested in the outcome. You can offer those who “beat” the test $100 or whatever, but you can’t standardise a real child claiming she was molested, for example, or a rapist claiming he was innocent. Reality has features that are too contingent on the moment that can’t be standardised away.

To add to the problem, whether or not something is a lie is not a light switch, yes-or-no issue. When my wife (hypothetically) asks “Does my bum look big in this?”, the problem I face is that I don’t thinks so, but I know that others might, and I also know that “big” is a relative term and I don’t know what parameters she is using. Big relative to a super model? Big relative to a normal good looking woman of her age? Even those are pretty loose criteria. And if I ask these things for clarification, I know I will look like I am prevaricating which will itself be a kind of answer. I also know that her question is probably just seeking reassurance rather than actually asking for factual information. So what does my answer mean? Is it a “lie”? Whether something is “true” is very commonly not a trivial question, and you have to idealise these sorts of problems away when you do studies, yet they are central to the process of lying.

Whether or not something is a lie is a deep question. The idea that there can be easy technological fixes for existential questions that deeply touch upon the nature of being human is a peculiarly 20-21st century one, I suspect.

[anson2995]

@FlyByNight512

That’s what I want to know. I’d like to do high-level security work at some point in my life, but I have an overactive conscience and could easily see myself getting a false positive on one of those things. Why in the world are they still using them?

Because polygraphs have a placebo effect. People who think they work may be more likely to make admissions. Police investigators know that going through the exercise can induce a confession because the suspect assumes they’ll detect every attempt to be deceptive. Or the police can tell a suspect that he has failed the lie detector test (whether he has or not) to exert more leverage. This has proven to be a very technique for extracting confessions, both true and false.

The idea that “sociopaths can beat the polygraph” really boils down to the fact that some folks aren’t afraid of lying. Convicted spy Aldrich Ames talked about this in a letter he wrote from prison.

[pravnik]

@Der_Trihs

I recall reading about a study some years ago where the ability of members of different branches of law enforcement were tested for their ability to detect lies. The only branch that was better than chance at detecting lies was the Secret Service; the speculation was offered that it might be because their job centers so much on watching people, observing them closely to see if they are a threat.

That was a study done by psychologist Paul Ekman, mentioned above by Thudlow Boink. Ekman tested federal polygraphers, judges, police, students, and U.S. Secret Service agents on their ability to detect lies based on facial expression. The only group able to perform at above-chance levels were the Secret Service agents, and only 29 percent of them were able to do so.

[Gary_Robson]

@Noel_Prosequi

One of the issues about studies trying to find if people are good at spotting liars is that the studies have to use standardised “lies” rather than real life ones in which the actual content of the lie itself is useful information.

Not so. The study I mentioned in post #17 used video recordings of people who were told to answer questions truthfully, and then told to answer the same questions using improvised lies. They were not yes/no questions.

[Peter_Morris]

[quote=“Gary “Wombat” Robson, post:44, topic:602287”]

Not so.
[/QUOTE]

Yes, so.

There is a difference between being told to lie in a test, and genuinely lying in a real situation. The first, somebody describes what they had for dinner last night, are they speaking truthfully, or making something up? The second, a suspect in a criminal investigation claims he was at home on the night of the crime, is he telling the truth or not? There’s a pretty big difference there, right? And while you can apply controlled tests in the first type, it would be extremely hard in the second type.

[bleibtreu]

@moejoe

I met a guy once who administered polygraph tests for a living. He said the key when you thought a person was lying was to convince him that the machine proved he was lying, and then use that to get him to confess. Once you got him to confess you would have him write out and sign the confession and once you had that the polygraph didn’t matter anymore.

In both HBO’s series The Wire and the earlier program Homicide: Life on the Street the fictional detectives did just that by convincing suspects that a photocopier was actually a lie detector.

Why Homicide: Life On The Street is amazing.

The Wire - Bunk's interrogation techniques

[Noel_Prosequi]

[quote=“Gary “Wombat” Robson, post:44, topic:602287”]

Not so. The study I mentioned in post #17 used video recordings of people who were told to answer questions truthfully, and then told to answer the same questions using improvised lies. They were not yes/no questions.
[/QUOTE]

That’s exactly what I mean. No real investment, it’s all just a game in the tests. Improvising a lie about something emotionally neutral is vastly different from doing it under pressure in an environment where you don’t know what the questioner knows and you are dealing with events at the extreme end of the spectrum of human behaviour like rape, murder or robbery.

A typical lie to police by a robber who was one member of a gang might be that he went along with a group of people to the bank but didn’t know they were going to rob it. He was just as shocked as anybody. He was only waiting outside in the car until they got back, he wasn’t a lookout, etc.

The problem with this lie is that it doesn’t account for the real motivations of the other robbers. Why would they imperil the operation by bringing someone along who was not in on the plan? And the robbers need a lookout and a reliable getaway driver. Why would they rely on someone who might not cooperate when the fact that this was a robbery became plain?

So without seeing anything in his face or voice, it is possible to tell from the content of his account that he is not telling the truth. That doesn’t really arise in laboratory conditions, even if the liar is doing more than telling yes/no questions.

The studies purporting to test police, judges etc aren’t really testing if such people can spot liars. They are testing if they can spot liars with one hand tied behind their back, using only artifically restricted parameters.

[AndyLee]

This is one of those, I can always spot a toupee arguments. You know when people say they always can tell when a man has on a toupee, and the counter argument is you can’t spot the good ones.

The real issue is if you can’t tell 100% of the time, how good is it?

How would you feel if you went to a doctor who treated you for mumps and he said, “well I was 99% sure you had mumps” but you didn’t.

Or the woman who gets pregnant .01% of the time when she’s on the pill

If someone says to me, “I can always tell when someone is lying,” they’re simply wrong. That would mean I could answer a thousand questions wrong and he’d be 100% of the time be able to tell me the whether or not I was lying. I can guarantee you no one could be that accurate.

You can make educated guesses, but what if you’re wrong. Even if you’re right 99% of the time, that’s not good enough

[md2000]

@Whack-a-Mole

I did this once.

…when the questioner asked me the beginning control questions I intentionally held my breath or clenched or paused. Apparently he could tell he was getting readings that would throw it off so he kept repeating the control questions (e.g. what is your name, how old are you, where do you live and so on).

…

I suppose someone who is not the least bit knowledgeable about how the thing works and is honestly sitting there answering might produce better than coin-flipping results but a modicum of knowledge will ruin its reliability. They are not hard to beat.

The question is not whether you can make the needles shake when you want, since that indicates lying; it’s more about whether you can keep them steady when you lie. That too is possible, but it’s a bit more tricky.

And the other key question, can you keep them steady when you know that you are answering the key question - when obviously you are nervous and stressed, but are not lying.

And finally, can even a trained polygraph reader tell the difference between the last two situations, or is he being influenced more by external factors, by what he knows “should be” the truth? Reading polygraphs is more art than science.