Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How can you tell if someone is lying?

theatlantic.com

Post reply on HN
Print view View original

131–140 of 363 posts

Re: How can you tell if someone is lying?

#131

Post by randycupertino »

doublejay1999 wrote: post #105
Robin_Message wrote: post #91

Earlier quoted context omitted.

If it's a book by Malcolm Gladwell, I can't anti-recommend it enough. All of the Malcolm Gladwell books I've read have been tripe. Unless he has radically improved his writing and grasp of nuance in the last decade (and published retractions of his earlier books), his books are dangerous anti-knowledge which will make you feel smarter but actually be dumber. Gladwell takes an obvious, folksy thing, adds a bit of a tw…

Well, I'm pleased to read this. I say that as someone who fell for the schtick, and he made a couple of sales out of me. He tells a decent story, and you feel like you've learn something after reading him. Scratch below the surface, and it turns out you haven't.

Ahh! Thank you for finally articulating what has always bugged me so much about his books. I am reading along thinking, "yeah, yeah, this makes sense" and then realizing later that it actually had no depth.

Re: How can you tell if someone is lying?

#132

Post by fluxinflex »

bobmaxup wrote: post #127
hbarka wrote: post #124

Then there’s the cousin of lying. Bullshitting. It is outright practiced at work and job interviews. Elon Musk at one time was interviewed on how he spots it.

Isn't Elon Musk a great bullshitter himself?

Takes one to know one, perhaps that's how Elon Musk does it.

Re: How can you tell if someone is lying?

#133

Post by andi999 »

raducu wrote: post #98
foobiekr wrote: post #65

David Simon's book, Homicide, has an apropos section: """Terry McLarney once mused that the best way to unsettle a suspect would be to post in all three interrogation rooms a written list of those behavior patterns that indicate deception: Uncooperative. Too cooperative. Talks too much. Talks too little. Gets his story perfectly straight. Fucks his story up. Blinks too much, avoids eye contact. Doesn’t blink. Stares.…

I'd like to see some hot-shot detective try to get a random number from 1-1000 out of my head by using those techniques. Yeah, he couldn't.

That's actually easy:the number is 50. Answer:no it isn't. Reply: he lied, get him boys we have the number.

Re: How can you tell if someone is lying?

#134

Post by doubleunplussed »

sriku wrote: post #123

"As a Homeland Security official told congressional investigators, “common sense” behavioral indicators are worth including in a “rational and defensible security program” even if they don’t meet academic standards of scientific evidence." What does that even mean? ... like there is a separate standard of evidence and there is a "we can get by without evidence" standard of evidence?

It's pretty obvious that there are useful truths that don't yet meet rigorous scientific standards of evidence. Like masks being useful in a pandemic or parachutes working. Non-serious "randomised trials" have been done with parachutes, and found no measurable effect - because for safety reasons the participants only jumped from an altitude of 2 metres or something.

In sports medicine there are loads of things where the anecdotal evidence is fairly strong and yet there is no rigorous evidence in favour of something. Like foam rolling, or running on your toes instead of your heels. Sometimes these things are hard to study, or simply haven't been studied yet. In those cases it's OK to fall back on anecdotal evidence, particularly when you don't have reason to believe the evidence is biased (like it would be if someone was trying to sell you a product or an ideology). Some of these things I'm sure will turn out to not be real results, if and when "proper" science gets around to studying them. But in the meantime the anecdotal, uncodified folk knowledge is better than useless.

When folk knowledge is "debunked" by science I sometimes still don't rule out that it might be true, because so much science is simply of too poor quality to be able to properly make any such conclusion. See the replication crisis.

I'm a scientist with nothing but respect for the scientific method itself, but "has this been scientifically proven" can be a really weak way to determine what's true and what's not in many cases. In those cases you kind of have to do your own reasoning with the data you have, anecdotal or otherwise, and try to get closest to the truth that you can. Which I think can be a valuable process in practice that shouldn't be discarded merely because it's not "proper" science.

Scientists are also flawed and sometimes publish results in a biased way to conform to an ideology. For example, I expect folk beliefs about gender differences to be somewhat accurate (not perfect), and for the official science to be hopelessly muddled.

Re: How can you tell if someone is lying?

#136

Post by graeme »

Godel_unicode wrote: post #85
Clewza313 wrote: post #78

Earlier quoted context omitted.

Most real life lies are also pretty low stakes. "Oh, I'd love to come to Aunt Mabel's party but I can't." "We should totally do lunch sometime."

Which is, of course, totally irrelevant. Nobody cares about detecting those lies. Where's the evidence that big lies are in any way related to little lies?

Everyone tells little lies. But it someone, in a game for example, is so good at lying that you have no idea whether they’re telling the truth, that raises some questions.

How can you trust your judgement on whether they’re truthful about bigger things?

Re: How can you tell if someone is lying?

#137

Post by Nitramp »

zakember wrote: post #116
raducu wrote: post #98

Earlier quoted context omitted.

I'd like to see some hot-shot detective try to get a random number from 1-1000 out of my head by using those techniques. Yeah, he couldn't.

That is because the stakes are too low to make you feel unsettled (in getting a random number). But if you knew you did something that could land you in prison and you were supposed to hide it, that would be a completely different scenario.

The article linked explicitly cites a study that disproves that assumption. People cannot tell liars apart better than chance even when watching videos of murderers' interrogations.

Re: How can you tell if someone is lying?

#138

Post by Veen »

tarsinge wrote: post #118
foobiekr wrote: post #65

David Simon's book, Homicide, has an apropos section: """Terry McLarney once mused that the best way to unsettle a suspect would be to post in all three interrogation rooms a written list of those behavior patterns that indicate deception: Uncooperative. Too cooperative. Talks too much. Talks too little. Gets his story perfectly straight. Fucks his story up. Blinks too much, avoids eye contact. Doesn’t blink. Stares.…

Countless times when I was a kid (and sometimes now) people thought I lied when I was innocent because of timidity (and not diagnosed but maybe on the spectrum), so yeah I was unsettled. Uncontrollable smiles were the worst ("that makes you laugh!"). I can’t blame neurotypical people for having heuristics, but I’m still afraid to be interrogated in something serious with my reactions analyzed.

The article shows neurotypical people should have the same concerns. Interogators who think they know how to spot lying are usually wrong, regardless of whether they are interviewing neurotypical people or people on the spectrum.

I'm not on the spectrum, but I was a rebellious kid who didn't like to take shit from teachers. My defiant and sarcastic reactions to accusations were often considered evidence that I'd done wrong when I was innocent.

It's not that lie-spotting heuristics work better for neurotypical people or worse for people on the spectrum. They don't work at all for anyone. We're all in the same boat.

Re: How can you tell if someone is lying?

#139

Post by noiseman »

Godel_unicode wrote: post #85
Clewza313 wrote: post #78

Earlier quoted context omitted.

Most real life lies are also pretty low stakes. "Oh, I'd love to come to Aunt Mabel's party but I can't." "We should totally do lunch sometime."

Which is, of course, totally irrelevant. Nobody cares about detecting those lies. Where's the evidence that big lies are in any way related to little lies?

Pathological lying is one of the indicators of psychopathy. Not quite the same thing as what you’re talking about but one can’t be too careful about those things.

Re: How can you tell if someone is lying?

#140

Post by addicted »

doubleunplussed wrote: post #134
sriku wrote: post #123

"As a Homeland Security official told congressional investigators, “common sense” behavioral indicators are worth including in a “rational and defensible security program” even if they don’t meet academic standards of scientific evidence." What does that even mean? ... like there is a separate standard of evidence and there is a "we can get by without evidence" standard of evidence?

It's pretty obvious that there are useful truths that don't yet meet rigorous scientific standards of evidence. Like masks being useful in a pandemic or parachutes working. Non-serious "randomised trials" have been done with parachutes, and found no measurable effect - because for safety reasons the participants only jumped from an altitude of 2 metres or something. In sports medicine there are loads of things where…

Your examples of parachutes and masks are not examples of useful truths that don’t meet standards of scientific evidence. They are examples of useful truths that have not been tested through double blind randomized control experiments.

But they both have scientific evidence behind them, and by any reasonable definition of meeting standards of scientific evidence, they do. The efficacy of parachutes can be derived from first principles. We know parachutes slow descent. And we know the risk of injury and death caused by force of impact. We can calculate how much a parachute slows descent snd therefore reduces force of impact snd prevents death.

The benefits of wearing masks in reducing the spread of COVID and serious disease through COVID can similarly be derived from first principles but more importantly, there are many comparative population studies that meet the standard of scientific evidence.

Post reply on HN