The Delusion of the Polygraph | Hacker News Reader

The Delusion of the Polygraph

lithub.com

298 pointsbookofjoe312 comments

I awhile back I used to do work with a major DARPA contractor. If you're familiar with security clearance for these roles, you know that at the higher levels of clearance you eventually need to take a polygraph exam.

I was never interested in going the clearance route, but got into a conversation with a grizzled industry vet that seemed like a character torn from a hard-boiled detective novel.

At the time I had recently learned that polygraph exams were "fake" and when the topic of the exam came up I was quick to point this out. His comment surprised me, and, in a sense, demonstrated to me that saying a polygraph is "fake" is akin to saying WWE wrestling is "fake". Of course it is, but that is a misunderstanding that what you're watching is a real performance.

He said the polygraph itself is just a tool for the interviewer. The real value was in someone who knew how to use the machine to convince the subject that they knew the truth. He continue that in his time he knew some mighty good interviewers who could easily extract anything they needed from you.

My father did go the clearance route, and when I asked him about the polygraph he told me he confessed things to the interviewers he would never have told my mother. "Fake" or "real" the polygraph does work in this sense.

Yeah, polygraphs remind me of those TV shows where the investigator pretends to use magic, voodoo, astral signs, etc. to solve the crime, but is really just using them to manipulate the psychology of the subject and see their reaction.

Put simply, the polygraph is a powerful tool if the subject believes it is a powerful tool.

That's incomplete. As the article points out:

> Even the United States government isn’t dumb enough to believe the polygraph works. The machine’s real purpose is symbolic, as an icon of the power of the state. Law enforcement agencies don’t use the machine to detect lies. They use it to coerce confessions. [...] It’s a fact, part of a story power tells itself to justify its power. Maybe you can beat the machine— they don’t detect lies, so it’s not that hard—but you can’t beat an entire country that believes in it.

Even if you know its nonsense, there's still something coercive of any system where it can be used as a pretext to punish you, or where you are punished for not pretending to believe in it.

> They use it to coerce confessions.

It boggles my mind than confession even counts as evidence but then again, so does any other testimony. Sure, it made sense when we had almost no forensics (and that's the times that shaped our legal systems) but today we do, don't we?

Living in civilised country I find whole confession, anything you say, can't lie in your own defence thing so absolutely abhorrent. To me it seems absolutely sensible that you should be able to decide what is your statement as answer to any question by state. And if you are on stand in trial as defendant you should be able to lie however much you want. The prosecution must prove you were lying, but the act itself cannot be illegal.

>Sure, it made sense when we had almost no forensics (and that's the times that shaped our legal systems) but today we do, don't we?

The CSI effect. The amount of forensics that people think will be presented in an average case is so much more than actually are. Finding and collecting usable fingerprints, DNA, shoe imprints, etc. does not happen in every case. Most cases are a lot of circumstantial evidence all pointing to the same person.

Fingerprints, DNA, shoe imprints, & other forensic evidence are circumstantial. Evidence is legally either circumstantial or testimonial, there's no other category. Most cases are a lot of testimonial evidence all pointing to the same person!

Even if CSI was real (which it isn’t even close), the vast majority of actions anyone takes leave no discernible evidence that isn’t immediately made useless through entropy.

And the most important element in almost every crime (intent) almost never leaves any evidence at all.

IMO the biggest subtle lie that CSI convinces people of is not that facts can be determined so easily and unambiguously - though that is a lie - but that the evidence found and any conclusions from it will fundamentally matter. Each piece of evidence is always some turn of the plot.

In real life, it usually doesn’t. Too much ambiguity, or inconclusive or inconsistent results. Or false positive/negatives. Or data which is useless in the vacuum of other missing information.

In real life, it’s a frustrated and depressing slog - punctuated by occasional moments of elation and/or terror - being a detective.

So what could be more compelling than someone telling everyone in their own words their intent and their actions, so everyone can stop guessing and ‘know for sure’? That’s what a confession is.

Which conveniently at the end of nearly every crime show the suspect actually does.

In real life, some do that - but many lawyer up, and you spend years dealing with every kind of bullshit and confusion game a professional can throw at you, instead of closure and a clear answer.

The polygraph is an attempt at bluffing folks into ‘we got you’ moments. Which does sometimes work! But the pressure and techniques applied can also result in people falsely confessing to things that never happened, or getting confused themselves and ‘failing/lying’ when they were actually relaying the truth.

A dangerous thought. There is no proof the interviewee tell the truth. It's easy to plant a fake memory to humans and make them believe it's "real".

Oh what am I thinking? The important thing is, the interviewer can produce a lot of "confessions" and "revealing of truth". They will be evaluated a good interviewer. Secure their job position. Sounds good. /s

Except there is no evidence it helps even “good” interviewers “extract” anything resembling truth. And there is lots of evidence it does not.

This comment is a perfect study of this almost uniquely American insane phenomenon.

But then I don’t question Koreans about fan death.

What do you make of the placebo effect?

The polygraph obviously has no basis for working, but while a sugar pill doesn't make a tumor disappear, it can be very good at pain management.

I still wouldn't use it in the context of the justice system, though.

So you mind can fix pain, but not tumors?

Where do you draw the line?

At things that only exist in the mind, like, say, pain.

Except that's not even close to true.

In what sense? Pain is a complex emotion triggered by various simple sensations (very hot, very high pressure etc). But remember that people with certain kinds of brain damage can feel these same sensations, but not pain. To them these just don't register as painful. Other people feel pain in limbs that they no longer have, so not triggered by any sensation at all.

Also, even beyond medical issues, different people perceive pain very differently. Hot peppers are perhaps the clearest example of this, where people accustomed to them feel the same heat, but not the same pain as someone unaccustomed.

In the sense that it's a physical process triggering mental reactions.

Pain is the mental reaction, not the whole process (of course, mental reactions are themselves physical processes, but that's a different discussion). When you take an opioid, the pain goes away, but the physical process that was previously causing it doesn't.

This is such a one sided dogmatic view of how the body works and heals that I honestly don't know where to start. I've seen enough evidence of bodies healing themselves just from changed minds. The truth is the synthesis of western scientific reasoning and eastern mysticism and holistic thinking. The problem was always blind faith, dogmatism, which is exactly what you're doing.

The problem is that the polygraph doesn't work on both levels. Obviously, it doesn't detect lies. But more to the point, it also doesn't extract useful information from most liars, and leads to fake confessions.

To stay in your metaphor:

- Not only do sugar pills not cure tumors, but imagine - 60% of recipients don't report decreased pain levels (no placebo effect) - 20% of recipients feel more pain

> What do you make of the placebo effect?

The placebo effect is measurable. If there is no measurable improvement, there's no placebo effect either.

Bear in mind that what most claims in favour of the polygraph measure is not truth but potentially-false confession. Extracting false confessions is relatively easy, it's also completely f-ing useless to wider society and massively harmful to the victim.

I see no issues with using polygraphs for hiring at intelligence agencies (I defer back to the comment about people missing the point of it), but as an investigative tool it's definitely a net negative.

You're assuming truth is the goal, which isn't correct. The goal of police is to close cases, and in this goal, polygraphs are quite effective.

They don't care if you actually did the crime, they care that they can extract a confession, or provide damning evidence to a prosecutor that lets them throw you in prison, so they can put a nice big checkmark on that case. Did they actually jail who was responsible? Maybe not, but who cares about that, apart from you?

Same reason for Forensics to exist. Don't misunderstand, some Forensic science has validity in many cases, but a lot of it is just straight up nonsense that isn't proven or peer-reviewed in the slightest, in fact many Forensic sciences that appear in modern court cases are completely, 100% debunked.

And like, why should they care? Even if you hire a crack lawyer team that gets you out of the court case, it's not like anyone involved in the investigation that almost threw an innocent man in jail is going to suffer an ounce of consequences. Or hell, even if you're wrongly convicted, worst case scenario you get a financial judgement after years of litigation, that's paid for by the taxpayers.

A sugar pill does not make tumors disappear. That's not what the placebo effect does; it changes your perception of pain and well-being, but not much else. (Of course, that can have a value in itself, but it's nothing like the magical healing effects found in urban legends.)

the healing effect comes from the fact that your perception of pain and well-being actually contribute to the healing process.

You must've misread what I wrote, since we both said the same thing.

I think OP is trying to say exactly what you’re arguing

The placebo effect itself isn't real (at least in the vast majority of cases where it has been claimed to exist), when people measure a "placebo effect" what they are actually measuring is simply a regression toward the mean, not a causal effect.

https://slatestarcodex.com/2018/01/31/powerless-placebos/

https://www.dcscience.net/2015/12/11/placebo-effects-are-wea...

https://pubmed.ncbi.nlm.nih.gov/6369471/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6707261/

I don't think this is right. Placebo effect definitely exists for conditions that are largely influenced by mental perception. The common example is pain. You can reduce people's perception of pain by deploying the placebo effect, e.g. giving them sugar pills that you convince them will reduce their pain. It extends to other similar conditions which are not generally (or possible to be) measured directly, but rather based on a patient's self-reported scoring. Like "on on a scale of 1-10 how would you rate your experience of this condition". Placebo effect can work for that. But not for other more tangible conditions.

Did you read any of the articles I linked?

> The most important study on the placebo effect is Hróbjartsson and Gøtzsche’s Is The Placebo Powerless?, updated three years later by a systematic review and seven years later with a Cochrane review. All three looked at studies comparing a real drug, a placebo drug, and no drug (by the third, over 200 such studies) – and, in general, found little benefit of the placebo drug over no drug at all. There were some possible minor placebo effects in a few isolated conditions – mostly pain – but overall H&G concluded that the placebo effect was clinically insignificant. Despite a few half-hearted tries, no one has been able to produce much evidence they’re wrong. This is kind of surprising, since everyone has been obsessing over placebos and saying they’re super-important for the past fifty years.

Using science to attempt to measure qualia sounds like a good way to produce whatever results you want.

The words "clinically significant" and "benefit" are not the same thing as the effect being real. To me it reads as if they are testing the hypothesis that a patient comes into ER with a sprained ankle and the doctor gives them this "new powerful prescription pain pill that just came out" and instead tricks them with a sugar pill. If this worked, I'm sure it would be used as much as possible. And the study you linked is simply confirming that PE is not an effective treatment for anything.

That's not the topic at hand here, which is "is the PE real?". For me it absolutely is.

This is what you were replying to when you said "The placebo effect itself isn't real":

> while a sugar pill doesn't make a tumor disappear, it can be very good at pain management.

And now you go to this version:

> There were some possible minor placebo effects in a few isolated conditions – mostly pain

So is it real or not? We're just saying the same thing, what's the point of saying it doesn't exist and then revert back to exactly what was said originally?

In my mid 20s I tried antidepressants for the first time. To me it was a big step because like many, I had a false perception of it having an unnatural effect on my personality, but I was finally ready to try them. The doctor said they will take at least 2 weeks to have any effect, and despite knowing that AND knowing about the placebo effect, I still "felt better" for several days after I started the regimen. To me, that was absolutely proof of the placebo effect, especially because after the 2 week window the effect was a backfire where I was in bed for a day and couldn't do anything. The pills backfired on me.

A single data point like that should never be considered anything close to "absolute proof" of anything - because you have absolutely no way of knowing that, for whatever reason (random chance, or the food you were eating at the time, or a compliment somebody paid you on the day you started taking them, or....), you might have felt better on those first days even if you hadn't started taking antidepressants at all.

Correlation is not causation, as they say.

(Hope your depression is gone or at least not too bad now days, regardless of what drugs or placebos may have played a part!)

Not only does a placebo reduce pain, naloxone will reverse the pain reduction just like it would if you'd given them morphone instead. Placebo effect isn't simple psychosomatic, rather something real and physical is going on inside the human body.

In a small but measurable percent of cases, the sugar pill does actually make tumors disappear though. [https://pubmed.ncbi.nlm.nih.gov/12509397/], and helps with almost every other factor of care in much larger percentages of the time.

Literally from your link :

> Conclusion: In randomized double-blinded, placebo-controlled trials, presumably with minimum sources of bias, placebos are sometimes associated with improved control of symptoms such as pain and appetite but rarely with positive tumor response. Substantial improvements in symptoms and quality of life are unlikely to be due to placebo effects.

My dude you are way over-thinking the polygraph. This is the more-usual setting it is applied in: https://www.youtube.com/watch?v=DgrO_rAaiq0

> no evidence

Saying "there is no evidence" is sloppy cable political TV tier rhetoric. There is absolutely evidence[1]. You and others may not find that evidence convincing, or otherwise think polygraphs shouldn't be used, but nevertheless it exists. A brief survey of the evidence suggests that the polygraph is probably slightly better than chance, but with high enough error bars that we should be very cautious about its use.

[1] https://nap.nationalacademies.org/read/10420/chapter/7

The point is not to extract truth, it's to extract behavior. It's the fact you can convince a judge or jury to take the evidence as evidence of truth that's a problem.

Polygraphs are usually inadmissible in court. It's unfortunate that "usually" applies.

https://www.hg.org/legal-articles/is-a-polygraph-test-admiss...

If the interviewee's behavior is not indicative of truth then the test serves no purpose other than allowing the interviewer or whoever commissioned the test (like a prosecutor or employer) to invalidly convince other people that the interviewee was lying

That's exactly the point.

That and to convince the interviewee that the interviewer knows they are lying.

Not if you view the interview as a process to expose reasons to not eg hire someone rather than to establish a list of facts or earnest perspectives. Literally just putting the candidate under stress. For this to apply in court you'd have to be suing them for not hiring out of discrimination over a protected class (i think, I am not a lawyer).

I mean maybe there are other civil suits you could file, but I suspect a lot of that would be signed away before the polygraph.

It's not about extracting facts, it's about establishing justification for the decision made by the interviewer. Sorry you "failed your poly".

Here's a good example: https://www.salon.com/2013/11/03/lies_i_told_to_become_a_spy...

What evidence do you have that there is no evidence?

For sure, you have:

- your opinion

- your knowledge (which is a subset of all that is known/"known", which is a subset of all that exists...though, it all typically seems other than this, such is culturally conditioned consciousness), have you something over and above this?

Just in case, please do not do this:

https://www.logicallyfallacious.com/logicalfallacies/Shiftin...

I can see how pressure would be applied when seeing the machine is leaning towards "lying", possibly breaking the subject's effort to lie.

But what would be the interviewer's strategy, if the subjects insist that they are telling the truth regardless of how the interviewer manipulates the machine? Wouldn't it immediately start discrediting the whole process if the subject is in fact telling the truth? I'm telling the truth here, and yet your machine says I'm not, hence it's broken, and hence I'll happily lie in the subsequent questions when it actually matters.

Most innocent people doubt their innocence when strongly accused even when they know they are right. Just a tiny bit, but in the right setting and with enough wearing you down, you can make innocent people believe they did it. I've seen it happen right in front of me.

That's why you don't talk to cops. Have a lawyer present and use the courts that us taxpayers pay for.

Also why narcissistic and psychopathic manipulators are so dangerous.

They don’t have to be cops. Most aren’t.

I wonder if this is related to people adding ambiguity to what they're saying.

Eg instead of saying "it's 20 degrees outside" they will say "last I checked it was about 20 degrees".

They change their phrasing because they want others to not think that they are wrong. By doing this they undermine their own credibility though.

Hmmm. People who make absolute statements like that generally* undermine their own credibility in my mind.

*I do it lots. Partly because the way my mind runs, I can nearly always (lol, can't help it) come up with possible conditions where I'd be wrong.

I am the opposite. I constantly add ambiguity even when I'm sure. I'm trying to learn to do less of it.

But what's the point of making innocent confess to false crimes in this setting? (i.e. requiring polygraph for job application)

I would imagine the entire point of doing a test would be to find out who is innocent and who is lying about being innocent. If you pressure the innocent into false confession, wouldn't it just make everything even more difficult?

In the context of a job application? Making up excuses to hide discrimination, maybe artificially limit the pool of applications to get around other hiring restrictions? Put the applicant on the back foot or make them share information they normally wouldn't? There are probably many ways to abuse it.

Good points. It never occurred to me that there are situations where recruiters intentionally try to remove legit applicants from the pool.

The polygraph would be used as the “bad cop” in the good cop, bad cop routine. After a line of questioning where the interrogator/polygrapher suspects lying, or is fishing for more information, they might say something like “everything sounded good but the machine is showing some deception. Is there anything you can think of that might be causing these readings? Anything you didn’t tell me? I want to get you out of here, but we need to resolve these results.” If the machine does show spikes during certain lines of questioning but not others, for instance about someone’s timeline on the day of a murder vs their relationship to the victim, it can be a reason to pursue further questioning in that area.

Given all the ways polygraphs can be misused or abused, the only real use I see is as in interrogation tool. But given the issues with false confessions in general, I think the interrogation should hold less weight, but that is a whole other issue.

That just means the interview process is selecting for only completely uninformed idiots. What does that say about the resulting organization?

That its a cult.

Commenter subjectsigma understands this. In a former life I had jobs that required a polygraph, and I was not in a cult, nor does the interview process select for uniformed idiots. Both of you are reacting to commentary that the author barely understands and that neither of you clearly understand.

Nobody in my circles seemed to think the polygraph was anything but a tool. In fact, the sibject is sometimes gossiped about in thise circles about the "relevancy" of the polygragh today anyways. The thing about buearacracy though is that change happens incredibly slow. If everyone decided to get rid of the polygraph alltogether today, it would still take some years to actually happen.

>Nobody in my circles seemed to think the polygraph was anything but a tool.

Just like the e-meter.

>buearacracy

Yeah, that's a cult.

You don’t understand what you’re talking about. It’s an open secret in CDC circles that the polygraph is not effective at catching trained liars, more of a ritual than anything. The polygraph is not a test for how gullible or misinformed you are. It is mostly a test for two things:

1) are you willing to play by the rules and follow orders, even if sometimes they don’t make sense?

2) if you are being lightly interrogated, do you immediately freak out and tell the interrogator everything? Do you have a really bad reaction to pressure?

If you don’t match these criteria then you probably aren’t fit to know extremely sensitive government secrets. But like I said, it’s more of a ritual than anything, the value for even those two tests is unproven.

Even smart and informed people who know exactly what a poly does can say and do things they wouldn’t normally when they’re strapped to a chair, hooked up to machines, and being yelled at for hours

I think that is true, but not the whole truth (staying on theme). It is an interrogation, but it isn’t meant, or likely, to catch a trained, hardened spy or someone that can stand up to interrogations. It is to attempt to find if there is information that would make someone a bad candidate for a clearance, the same as the general background check is doing. If you are in massive debt, you are at much higher risk of being bribed. If you are cheating on your wife, and attempt to hide it during your polygraph, you are at much higher risk of being blackmailed.

It isn’t going to “catch” everyone but it is another way to reveal people with vulnerabilities that could be exploited. I think the real issue is people that “fail” the polygraph, since it isn’t actually a lie detector in any sense. It would be better if they just considered it a polygraph assisted interrogation.

Yep, IIUC blackmail or coercion is one of the primary concerns. Looking at people who sold out to foreign actors I think financial troubles are the #1 reason

>strapped to a chair, hooked up to machines, and being yelled at for hours

I've done periodic polygraphs, both lifestyle and full-scope, every 5 or so years since 1997. None of mine have ever lasted longer than 30 minutes.

You just sit in a chair while wearing some straps and there's never been any yelling involved.

It's all quite prosaic and relaxing actually.

It has been my experience that the clearance investigation process is quite simple, although I lead a very boring and law-abiding life.

I know of some people who have had quite long polygraphs and failed them repeatedly but my hunch is that the examiner has the findings of the background investigation in-hand and is trying to clarify some findings.

Many people with past financial, drug, or legal problems have gotten through the process with no issues just by being open and honest with the investigators and polygraphers.

So yeah, if the background investigator interviewed a friend of a friend of a friend and was told that 20 years ago you used to get stoned in college and whip your dick out but when asked during the polygraph about past drug use you go "I've been a squeaky clean boy my whole life" you're gonna have issues.

The annual financial disclosure is much more stressful just because of all of the damned paperwork.

That’s a little comforting at least… only met one guy IRL who ever said “Yeah the polygraph was fine.” Everyone else said it was a miserable experience for one reason or another, even people who were very straight laced

I have watched quite a few (American) police interview videos lately, and regardless of tools (polygraph or Reid(tm),) I wonder how many interviews start with the perpetrator really having no rationale, and ending in them simply back-rationalizing their emotions. A fit of rage might not have a rationale, if you're that predisposed. But being pushed to explain yourself will make the brain do what it's constantly doing: retroactively explaining your emotions. Especially if you've been promised a reduction in stress if you do.

And then there's the opposite, when the subject continuously makes no sense, because they have brain damage: https://www.youtube.com/watch?v=_c_lmx4LdNw

> I wonder how many interviews start with the perpetrator really having no rationale, and ending in them simply back-rationalizing their emotions

That's the core of the Reid technique, isn't it? Here's two rationales, one socially acceptable and one socially unacceptable. Pick one.

We don't actually care which one you pick, because a confession is still a confession, but we're handing you a convenient narrative you can use to justify your actions (a carrot), and an alternative people might believe about you if you don't pick Box A (a stick).

Something similar happens in high-pressure sales. I bet those guys would make great interrogators.

Wasn't familar with the "Reid Technique" so I had to look it up. One of the opening wikipedia paragraphs is just perfect.

>In 1955 in Lincoln, Nebraska, John E. Reid helped gain a confession from a suspect, Darrel Parker, for Parker's wife's murder. This case established Reid's reputation and popularized his technique.[3] Parker recanted his confession the next day, but it was admitted to evidence at his trial. He was convicted by a jury and sentenced to life in prison. He was later determined to be innocent, after another man confessed and was found to have been the perpetrator. Parker sued the state for wrongful conviction; it paid him $500,000 in compensation.[4]

It's the same story with: bite mark analysis, police dogs, hair comparison analysis, firearm toolmark analysis, many arson analysis techniques, bloodstain patterns. Discredited or unproven, yet still used in court.

Having served on a jury I implore you: Stay as far away from the criminal justice system as you can. Once you're in that courtroom your life is a coin toss away from effectively ending.

footnote: and regardless of innocence, you will be running from the arrest the rest of your life.

I will add this: serving on a jury is - IMO - a very valuable experience. Or, that is, IF you think you might ever find yourself on trial, I think it would be very valuable to have served on a jury yourself. You'll understand a lot about jury dynamics, and how juries make decisions (hint: it's not always as cut and dried as the facts and evidence presented). You may have insights that even your lawyer won't have, depending on whether of not they themselves have ever served on a live jury.

Will this information be useful to you? I believe it well could be. It's hard to explain exactly how/why without actually going through the experience though. And the verdict will still depend on many factors, many of which will be out of your control. But a few insights into the deep inner details of the process might be enough to tip the odds in your favor if things are close to begin with.

Another thing I'll add: a good lawyer really helps. The case I was on, the defense attorney was just totally on-point and absolutely "nailed it". Every single time - EVERY time - a witness for the prosecution said something questionable, or contradictory, or that in any way exposed a possible hole in the prosecution's story, he was all over it. By halfway through the trial I was rooting for the guy because watching him work was like watching a maestro in action. Hell, in the jury room during deliberations, we were all joking about how "If I'm ever on trial for a serious crime, I want this guy defending me."

That said, there's always an element of luck involved. In this case, I'm pretty sure the defendant was guilty. But I mean "pretty sure" in the sense that my subjective Bayesian posterior for "guilty" would be more than 50%... but the prosecution definitely did not prove he was guilty "beyond a reasonable doubt". And aside from the skillful performance of the defense attorney pointing out holes in the story being told by the prosecution, was the simple fact that there were holes. And part of that was because the investigators were bumbling and incompetent, borderline "Barney Fife" or "Officer Barbrady" types. To the point that in the jury room during deliberations, we were joking about how "If I'm ever under investigation for a serious crime, I hope this guy is leading the investigation." A better investigation might well has resulted in a guilty verdict, but as it was we acquitted the guy.

Another observation: the defendant took the stand to testify. Usually not recommended for criminal defendants, but it worked for him. Why? Well, not so much because of what he said or didn't say, but because the prosecutor acted like a dick towards the guy, belittling him and demeaning him, talking down to him, and just generally being a prick. Now strictly speaking, none of that had anything to do with whether the defendant was guilty or not. But it created sympathy for the defendant, and the prosecutor basically turned himself into the villain. Did if affect the outcome? TBH, yeah, I think it kinda did. Things were close as it was, and that little bit of extra sympathy might easily have been the deciding factor. But the thing is, this is one of those "things that are outside of your control" if you're ever on trial. Maybe your prosecutor will be more professional and under control. Not much you can do about that. But just seeing how emotional aspects like "sympathy for the defendant" CAN play into a decision can be valuable, I believe.

Anyway... sorry for the long rant. I'll just say that if you ever get invited to jury selection, I'd encourage you to NOT try to "get out of it".

Now I’m curious if anyone has studied criminal activity by jury pools.

Are jurists more likely to commit crimes that they’ve presided over? Do they absorb any knowledge from the cases they decide that enables them to live a life of crime?

It's an interesting question. Speaking only for myself though, I would never do something like that!

Muwahahahahahaha...

Based on your username, I’m… skeptical :-)

What was the crime being prosecuted?

> What was the crime being prosecuted?

Stealing a flashlight and a radio from a cop car, basically. The funny part is, that's the bit that I'm "fairly sure" the guy was really guilty of. But they also threw in some hinky claim about him stealing a chainsaw and a shotgun and some other stuff from a truck that was parked nearby (this was all at the officer's house). And for the most part, none of us on the jury believed he stole the chainsaw and the shotgun... in fact, we mostly think there never was a chainsaw or a shotgun. Because this guy was on foot, drunk, in the middle of winter, and was somehow able to carry a flashlight, a radio, a shotgun, and a chainsaw off to some safe hiding spot? From which they were never recovered? But he was found wandering on the side of the road later that morning with the radio and the flashlight? Hmm... so yeah, where did the gun and the chainsaw go???

Anyway, basically we thought they tried to "pile on" this guy with some made up bullshit, and that also probably helped us make the decision to acquit him, even if he was guilty of the radio and flashlight bit. And that played in with the prosecutor acting like such an ass to the guy and made the defendant seem like the victim.

Funny thing: after the case was over, they asked the jurors to hang around for a while (voluntarily) and talk to both the defense attorney and the prosecutor about how we reached our decision. And I actually told the prosecutor "dude, you were a total dick to that guy and made yourself look really bad by doing so". He said it was kind of personal with him and that guy, because the guy is a repeat offender and he's tried him like a dozen times. Now whether or not that supports a belief that they tried to set him up on some extra bogus charges I'm not sure, and I didn't ask about that. But man, oh man, did I learn that weird shit goes on that most people never see or hear about.

And of course there was the mysteriously malfunctioning dash-cam that didn't record any audio when the first officer approached the guy when he saw him on the side of the road, and the mysteriously missing 911 audio recordings of the guy supposedly keying up the radio and talking into it, etc., etc. And the fact that the cop car that the stuff was stolen from was left unlocked, and was un-monitored for like 11 hours from the time the officer got home from work until he got up to leave for work the next morning and found the radio and flashlight missing. And the absolute lack of physical evidence (fingerprints, footprints, eyewitnesses, anything) outside of him being seen carrying the missing radio and flashlight (he claimed he found both lying on the side of the road).

Like I said, there were serious holes in their case to leave room for "reasonable doubt" even though the guy was "probably" guilty.

"He said it was kind of personal with him and that guy,"

Completely unethical and possibly malicious prosecution. Amazing.

Yeah. The whole process was rather eye-opening.

If you think this is an isolated case, you vastly overestimate humanity. Stay away from the criminal justice system.

> they asked the jurors to hang around for a while (voluntarily) and talk to both the defense attorney and the prosecutor about how we reached our decision

My jaw is on the floor.

Here in Blighty, a jury deliberates in absolute secrecy. Revealing jury deliberations, even after the verdict, is a serious offence.

Here in the US that kind of thing generally varies from state to state. But as I understand it, in most states, once the verdict is rendered and the jury has been discharged, the jurors are free to discuss the case with anyone.

> We don't actually care which one you pick, because a confession is still a confession

Yes, but as you note, the point is to make it sound like barely a confession, the minimum possible confession. This makes it just as attractive to the innocent as for the guilty. It's offering a minimally painful way out of a deeply stressful situation.

And as you say, the punchline is that not confessing would lead to minimal or no pain, and every level of confession will equally turn out much worse. An innocent person is just trying to escape from that room, and is being conned into agreeing to a long prison sentence in order to do it.

Totally. I remembered the phrase I should have used: a false dichotomy (https://xkcd.com/2592/)

If you were going to buy a car today sir, would your budget be above or below $40k?

"$40k is way too much on the low side. Sorry, I'm looking for something a lot higher end." And walk off never to come back. ;)

> I bet those guys would make great interrogators.

Maybe on the context of US criminal interrogators, where discovering what actually happened isn't one of the goals.

> Something similar happens in high-pressure sales. I bet those guys would make great interrogators.

Without question. The best, most accurate, and most comprehensive texts I ever read about how to go about manipulating people was a series of books aimed at car salesmen. To the degree that "mind control" is a thing, these books were clear how-to guides.

And they convinced me to never talk to a car salesman.

Really important similarity to cops is that car sales people try really hard to control the physical space you are in. Can be simple, like, offering coffee, through shady, like taking your license on some pretext, to games like "wait here while I get the manager". The point is to lock you in a space and control the pacing and narrative.

This is why if you detect any nonsense like this, you should always remember that at any time in a car buying transaction, you can just walk.

> you should always remember that at any time in a car buying transaction, you can just walk.

Not just with cars!

I’ve watched police interrogations too and they are both fascinating and horrifying. Those interrogators have a lot in common with shady used car salesmen. They twist and contort the truth to get whoever they want out of the person being interviewed. Except unlike the car salesmen the good interviewers really know their shit and can “corner” a person in their own lies. (I suppose a good salesman is just the same though)

It’s a weird place to be when you are rooting for the child molester/arsonist/muderer hoping they’d come to their sense and fucking CALL A LAWER YOU FUCKING IDIOT and SHUT YOUR PIE HOLE!!!

But oddly, I guess maybe it’s a good thing “rape the kids and wife, shoot them point blank and burn the house down” criminals are too stupid to exercise such a basic right. Even the scummiest of police investigators give these people the option to shut the fuck up and call the lawyer. Sure the person might have to wait an obscenely long time before the lawyer shows up but they still are given the out yet these moron criminals think they can outsmart a highly trained police interrogator and choose to dig their holes.

Usually these people already dug their hole well before they are drug into the police station though. The on the ground evidence usually pretty much points to them already. All the confession does is save the state millions of dollars taxpayer money with courtroom proceedings.

So yeah… really mixed on the whole topic. Of course these videos on YouTube are selected to be the most interesting of the bunch. There are probably a hundred more mundane interrogations that go unseen for every one that makes it to a widely subscribed YouTube channel.

> They twist and contort the truth to get whoever they want out of the person being interviewed.

It's possible I just haven't watched enough of them, but I've never seen that. Usually the interviewer is a calm and dispassionate Columbo type who asks clarifying questions and then lets the suspects slowly trap themselves in their own web of lies. It is fascinating to watch.

That said, even though I have a healthy respect for the criminal justice system, I will still NEVER talk to the police without a lawyer. (Whether or not I've committed a crime.)

>Usually the interviewer is a calm and dispassionate Columbo type who asks clarifying questions and then lets the suspects slowly trap themselves in their own web of lies.

This loses its ability to inspire awe when you watch a video of one of these where you know the subject is actually innocent and the interviewer manages to also catch them in their own web of lies and make them look and sound guilty.

Academia has reproduced this result a fair bit: under interrogation it's very easy to convince people to admit to events happening which didn't. Eyewitness testimony is thus intensely unreliable - and psychologically as far as can be told, the act of remembering something makes the memory itself labile - i.e. you rewrite a memory as you recall it.

So rounds of intense questioning on the same subject infect any recollection of the experience someone has - amplifying or even adding events which didn't exist due to the focus of it.

And this has very real consequences - it was one of the driving forces of the satanic panic in in the 1980s[1] - for which people went to jail on the basis of "recovered memories".

[1] https://en.wikipedia.org/wiki/Satanic_panic

> Sure the person might have to wait an obscenely long time before the lawyer shows up but they still are given the out

This is not to challenge anything you said, but you have two rights here: the first is to have a lawyer present while you are being questioned, the second (and more important right) is the right to remain silent.

Talking to a lawyer is obviously a good idea if you are in any sort of legal predicament. Even if you were not arrested or charged, but have certain legal obligations or are considering signing a binding contract etc.

But when it comes to police "interviewing" you, you don't even have to go as far as to ask for a lawyer. You can simply leave it at "I don't want to talk to you." Even THEN you don't have to go that far. You can literally communicate absolutely nothing. Catatonic non-responsiveness is not a crime. This applies whether or not you are under arrest or being detained. You can always refuse to speak. That is your constitutionally protected right.

I have seen these too and although I'm certain there are bad interviewers out there, I have to say I gained a lot of respect for the detectives who conduct those interviews. They are much better than I could ever be at remaining dispassionate, curious, and above all, extremely patient.

The bottom line is that these interviews are all recorded and the police are well aware that if they make a misstep and if the defense has a competent lawyer, they may inadvertently set a thief, killer, or rapist free if they are not extremely careful in their questioning and processes.

Everybody always thinks this about stuff that doesn't work. It turns out it's total bullshit.

There was someone I knew who claimed that dowsing rods worked a similar way. A good practitioner of the dowsing rod was really using the information from his audience and was Hans the Mathematical Horsing his way to water. The audience of course had this information from the subtle way they picked up on cues of water, ways even they didn't recognize until the dowsing rod was in action.

They all think this. Some can even bend spoons.

Dowsing rods are an excuse to trust common sense; that dip in the land where plants are greener is probably a good place for a well. But digging wells is expensive so people want something more than a guess, but also something cheap.

When I was a kid, a new house was being built downhill from my parents and they dowsed out a "good" place for a well. They were all set to drill before my dad went out and told them they were directly downhill from our septic tank.

What's "downhill from a septic tank"? Isn't a septic tank a container that slowly fills up and is periodically emptied by a truck or something? There is no "flow downhill", is there? I thought the whole reason to have a septic tank is that you have no place for the stuff to go.

Septic systems are more interesting than that: https://en.wikipedia.org/wiki/Septic_drain_field

A straight tank would fill up in no time.

Unless you're in a location where you absolutely can't leach the liquid waste out into a leach field (like right on a lake), the tank usually just settles the solids and give the liquid some time to mingle with whatever biological processes are happening in the tank.

When the ground doesn't perk naturally, it's common up here to build a mound system where you have a mound of soil that does perk and vegetation (grass) on it to take up the liquid.

A fully closed tank is basically the last resort. We have friends with a house right on a lake, and their tank had to be pumped every three weeks before they had a kid. I can't imagine what the interval is now.

Thanks for explaining, I had no idea.

The best brain power a human possess is subconscious-- its all the perception that allows you to run fast around moving people and catch a ball. You can do it but you can't explain how. You can probably train yourself to be good at finding water or reading people but you might be bad at explaining why. Worse-- trying to explain yourself might be forcing you to use the feeble symbol processing power of your brain. So a prop helps you feel your way.

Still-- total bs.

A friend bought land to build a house on where dowsing was culturally pervasive. He knew it was bogus but it was cheap compared to the land price and everybody around was heavily pushing it. An old guy came out and, during the performance, told him the total history of the land parcel, including stuff that would be inappropriate in formal disclosures ("those neighbors are assholes" etc). They did hit water, but the whole area is pretty verdant.

As someone who used to be a cop, this is absolute peak cops justifying the evidence not supporting their intuition by just making bullshit up.

If it actually worked, they'd have data and results. But, spoilers, they don't.

What's your opinion of the copy machine polygraph scene from the opening of The Wire's final season?

That was a pretty accurate presentation of how polygraph machine testing works. The machine isn't doing anything useful in terms of determining if what is being said is true or not.

Doesn't the polygraph machine just play a role similar to "the manager" in a used car negotiation? Just like the salesman can leave the room, get a coffee, never actually talk to anyone, and come back in and say "Sorry the manager says I can't go that low" and lots of people will buy it - the polygraph interviewer is "saying" to the interviewee (in not so many words) "sorry pal, I'd love to believe you, but the machine says you're lying - my hands are tied"

Seems like it's a useful prop for manipulating people, and in that role it really is effective.

> Seems like it's a useful prop for manipulating people, and in that role it really is effective.

Similar to e-meters in Scientology.

I’d compare it to the sobriety test of being asked to walk in a straight line. someone who’s drunk will put a lot of effort in it, which is the giveaway.

Or someone who's nervous because the police are accusing them of a serious crime...

Or... it doesn't really matter.

By the time the police are commanding you to do a field sobriety test, they have already decided to charge you with DUI. There is literally nothing you can do during the test to change their mind. It's a formality, and the "test" is vague enough that the officer can cite any little twitch or misstep as "evidence" that you failed the test.

If you pass a breathalyzer test (blow under the limit), and they still want to charge you with DUI, they will likely do a field sobriety test because the results are non-numeric and are subjective. Heck, you can blow 0.0 and still get arrested for DUI[1][2], and if you tell the world about it, the police will sue you for defamation[3].

1: https://reason.com/2024/02/14/iowa-cops-arrested-a-sober-col...

2: https://www.youtube.com/watch?v=QGWSbAHaHUw

3: https://iowacapitaldispatch.com/2023/10/02/after-traffic-sto...

Complete nonsense. They need evidence. If you blow 0s, you're not getting a DUI. Just because you have an example or two doesn't mean it's common. I've been pulled over, blew 0s, passed the tests and let go.

With the wide availability of countless drugs that impair driving (that may not even be detectable on a urine/blood panel) and obviously don’t register on a breathalyzer you absolutely can be charged with and convicted of DUI based on behavior, FST performance, officer observations, driving pattern, etc alone. Stumbling over a word or two like I do on conference calls everyday could be considered “evidence”.

Just like you can also be charged and convicted of DUI even with zeros or being under the legal limit. If you’re traveling to/from/around a bar area at 2:30 AM your driving pattern and behavior is going to be heavily scrutinized.

Just because you have an anecdotal example or two doesn’t mean it’s common either. The FSTs are also completely stacked against you. Take a high-pressure scenario, less than ideal conditions (side of the road with passing cars, dark, cold/hot, precipitation, flashing lights, etc), and ridiculous/conflicting/confusing instructions and even people who are completely sober end up providing what could appear as damning evidence.

Even professional athletes have bad days where they just can’t land a shot they’ve nailed thousands of times.

I know at least a few cops who openly admit they struggle with the tests (to the point of “failure” in some cases) in no-pressure ideal classroom training environments.

Of particular curiosity is a lot of police body/dash cam footage where the officer struggles to demonstrate/explain the tests, stumbles over reading Miranda cards, etc. Evidence where if the same observations were applied to them they could be scrutinized as being “impaired”.

Of course I’m not advocating for impaired driving, just highlighting that it’s a tricky situation overall.

Are you white?

Yeah... Cause it's hard to do when you're drunk.

Sobriety tests are notorious for being very subjective and not having well defined criteria and cops failing people even if they aren't drunk. The subjectivity in the test is a feature that allows cops to justify their arrests or uses of force.

Obviously once you hit a certain point of drunkness obviously maybe a test like walking in a line can demonstrate something useful. But so would a breathalyzer. The false positives is the problem because they're being used to illegitimately subject innocent people to criminal charges.

I'm reminded of the story about the cops who constructed a "polygraph" by attaching wires to a colander that ran to a photocopier in which a piece of paper that had "You're Lying" written on it was placed. It was enough to intimidate the suspect into singing like a bird.

I believe that, or something very similar, happens in _The Wire_.

Scene from The Wire doing roughly this: https://www.youtube.com/watch?v=bg13hyfm2Ik

Great point, it reminds me of when I read about how "trials by ordeal" sometimes worked.

Consider the boiling water/oil thing: If you're innocent, you can "stick your hand in and not get burned."

What they did was, they faked the water being hot by blowing bubbles in it. All then that was needed was for everyone to "believe it worked," The innocent sticks in, and then the guilty confesses.

A perfect description of how "trials by ordeal" don't work. This phrase is doing some HEAVY lifting:

> All then that was needed was for everyone to "believe it worked,"

In an era where mass media basically didn't exist, most people couldn't read, and information about how these things work could not easily spread, it might have been easier to convince people that a fake test was really what you were claiming it was. If people could google it they would instantly find out it was fake.

So guilty people were convicted, and so was anyone who knew how oil works.

Unlike now, where only innocent people are convicted and everyone knows all the science that always works, like DNA testing?

Oh, correct. I think by "work" I merely meant "here is the mechanism," not "this is why they are successful."

I knew a professional polygraph examiner who told me the exact same thing.

It means that the polygraph works as well, and in the same way, as the ancient Roman(?) method of having a tent sealed off from light, with a donkey in it. The examinee is told that he is to hold the tail of the donkey and if the donkey brays while he says the thing he's being tested for, then they know for a fact that he's lying.

The actual test, though, was that the donkey's tail was covered in soot. If the examinee comes out of the tent with clean hands, they know that he didn't hold the tail and so is deemed to be untruthful.

What if the donkey kicks you?

Means you were about to say something the donkey didn't want to be made public.

It means you stood in the wrong place. Don’t stand behind, stand next to the donkey.

You kick the ass in the ass.

There's a similar "actual test" scene in By The Great Horn Spoon!, a fun kid's book about the California Gold Rush we read in elementary school.

Sounds like the premise then is that everybody has got some shit they are hiding but it’s better to employ a criminal than a lier.

Which would mean that the polygraph is really good at filtering out genuinely good people from the recruitment process (in the case of this article people whose actually never done drugs or talked shit about their superiors).

This is a bit like saying good witch doctors can diagnose you without the voodoo accoutrements, which are just there to get you to open up about how you are feeling. It's still voodoo, and still produces garbage conclusions much of the time.

Thing is, there isn't an academic journal for voodoo. There is at least one for polygraphy.

Wait, what, Voodoo doesn't work? Has this been studied?

This reminds me of a study I read decades ago that showed patients have better outcomes from medical care if the doctor has credentials prominently displayed on the office wall.

> My father did go the clearance route, and when I asked him about the polygraph he told me he confessed things to the interviewers he would never have told my mother.

And this is exactly the problem, people make stuff up all of the time under stress.

> Of course it is, but that is a misunderstanding that what you're watching is a real performance.

This is not the value. The value is that the polygraph is that its an end-run around employment law. You can't use a polygraph on a general employee to fire them nor can you fire them for many of things that they ask in a polygraph interview. However you can revoke their clearance and fire them for not having a clearance.

Well I agree but also if someone makes stuff up that is incriminating while on “friendly examination” they shouldn’t get the clearance because they will absolutely incriminate themselves and everyone they met in their lives when on “unfriendly examination”.

If you mean when their tortured I thought the consensus was that everyone talked under sufficient torture and the actual problem was determine fiction from reality.

I know some govvies who have been through the polygraph circus.

One guy knew it was BS and could not get worried enough for them to come up with a baseline - he was too calm. So they took the tack of rescheduling it a bunch of times, making him go home after showing up so he would be good and pissed off when they actually did the test. He passed.

Another lady I know had the interviewer go so hard on her she was crying through half of it. Afterward, the interviewer told her the goal was to make every interviewee break down so they would reveal stuff.

Don't scientology go around with a similarly fake device?

Yes, the "E-meter" is a primitive form of lie-detector equipment:

https://rationalwiki.org/wiki/Dianetics#E-Meter

"Basically it is a simple ohmmeter that measures galvanic skin response (electronic resistance of the skin), somewhat similar to a polygraph; the user (the "preclear") provides one of the elements in a Wheatstone bridge."

Yep - this is like field sobriety testing, in my mind. Everyone will display some level of nervousness and inability to perform all the tests, and the officer thus has a baseline level of "cause". They can therefore do all kinds of tests or hold you until you do those tests. B/c "He failed his FST"

Same with poly. If they don't like something, they can just say "He failed his poly"

I don’t get how “field sobriety testing” in the sense of observing someone perform some coordinated actions is still a thing. In Ireland if stopped by a member of the police (Gardai) and they have suspicion they can require you to provide a breath or saliva sample. If you pass you are free to go, if you fail or refuse you are escorted to a police station and provide a blood sample or urine sample (I think this depends on if suspected of alcohol or drug offences) refusal to provide the sample is an offence

The purpose of polygraphs is to minimize the amount of bits of information one's adversaries can gain information OVERALL while maximizing the amount of bits of information one gains OVERALL—only (but not all bits are equally worthy which makes this partially an algorithmic problem).

Then they use the polygraph to obfuscate information (EG about the process of selection, knowledge gathering, and a considerable amount of other things); to have a highly spacious separate "stage" for the real sensors; and can NEVER retire it because that would reveal several bits of information.

Consequently they ALSO can never state this.

> But the machine remains useful for extracting confessions. [...] Despite a growing body of evidence, including hundreds of exonerations based on DNA evidence, most people don’t believe in false confessions.

Arguably the bigger/worse false belief right there.

> First the exam makes you doubt or forget your memories. Then, by forcing you to re-access them again and again under stress, it literally rewrites them.

To some extent this happens naturally, so if the questioner really wants accuracy, you won't force people to re-access the memory for no good reason.

There is an entire little ecosystem/subculture around "repressed memories" doing a lot of harm to vulnerable people. Basically you go to a "therapist", they do some sort of hypnosis/interview session where they ask you a lot of very leading questions, and then you leave having been convinced that your family or a satanic cult abused you as a kid (or in some cases that you have been abducted by aliens). The person performing this interview might not even be aware they are doing anything wrong, to them those leading questions ("Do you see anyone else in the room with you? Look closer, are you sure?") may just be how you get to the truth.

> (or in some cases that you have been abducted by aliens)

Really? There are therapists who will try to convince you that you were abducted by aliens?

I don't think anymore, but in the last century there were several cases of therapists using hypnosis to unlock "repressed memories" of alien abductions. They usually wrote up books about it for profit.

Not specifically, but they will follow the absurd path of asking you leading questions until you convince yourself that you were abducted by aliens, and then being quacks will decide that their methodology couldn't be wrong, and so they validate your own invented beliefs no matter how stupid.

If it keeps you coming back for another session, they'll keep doing it, even if they know the whole process is bullshit.

This is thoroughly studied by psychologists. I feel obligated to tell you that multiple therapists have used access to repressed memories to control their patients. A large, highly organized religious movement overtly claims that its alien teachings are supported by repressed memories.

Hey quackery or not, the Polygraph saved my heart. (Disclosure: A bit of an exaggeration.)

After hours of uncomfortable prodding, an interviewer came into the room and suggested I see a doctor for what looked like heart arrhythmia. I did shortly after and they were right! (It was deemed nothing too serious though.)

Yeah, the instruments work fine, but the interpretation of the readings is just modern day tea leaves.

Alternative take: lack of accessible, affordable and comprehensive healthcare almost cost me my life.

I'm actually curious. Do you get an annual physical? Does your PCP give you an EKG? That would be the appropriate way to catch this.

Are annual physicals provably worth doing? Or are they the medical profession's version of the polygraph?

I was young back then and was not getting annual physicals. That has changed, along with regular exercise!

Last night the polygraph saved my life…

From the article:

Even the telltale spike in the polygraph chart, itself largely a myth created by TV and movies, can indicate anything from a heart problem to sexual attraction. (Indeed, the machine’s inventors used it to detect both of those things. Larson married one of his first test subjects, and Keeler discovered a heart defect while testing the machine on himself.)

First sentence on Wikipedia:

>A polygraph, often incorrectly referred to as a lie detector test, is a junk science device or procedure that measures and records several physiological indicators such as blood pressure, pulse, respiration, and skin conductivity while a person is asked and answers a series of questions.

Many years ago, at the age of nineteen, I was forced to take a polygraph if I wanted to keep my job. Someone was stealing products from the store, we heard out the back door, and they required everyone to be tested.

Naturally, my co-workers and I discussed this among ourselves and we all agreed to test, we knew we were innocent. One of the men said all they're trying to do is see if anyone cracks under the pressure of the test. Being kind of a nervous type of person, I was concerned they might misread my domineer. I talked to my sister and she told me to try my best to control my breathing during the test.

For me the problem wasn't that I was guilty, it was they would think I was guilty. We all passed the test and went back to work, later it was revealed the thief was someone on another shift. Or were they just nervous?

This might have been illegal, depending on how many years ago this was: https://en.wikipedia.org/wiki/Employee_Polygraph_Protection_...

In either case it was complete BS and just shows the sorry state of labor in the past few decades.

Confess! Confess!

A LOT of businesses break labor laws on a regular basis, especially smaller businesses without a legal department telling managers "You can't do that".

You'd be alarmed to know how many restaurant owners tell the staff to clock out when the store closes and then finish closing duties off the clock, among other types of wage theft.

domineer → demeanor

I only post the correction because it took me a couple of minutes to figure out. Domineer is an uncommon word, so initially I thought I had a gap in my vocabulary, but couldn't find any definitions that made sense in this context.

Wish I had seen your post before I resorted to ChatGPT.

https://chatgpt.com/share/2400c631-fc68-4f95-b4e7-861324a3dc...

That’s neat

Oops, thanks for the correction.

When I was in high school, I had a similar experience at the place I worked. The manager had money missing from her purse, and someone (the actual thief, I assume) said they saw me take it.

On my own volition and expense, I took a polygraph test about it. On one of the control questions, I kept reacting in a manner the examiner said indicated untruthfulness. I was certainly being truthful, though. No matter how he reworded the question, I failed it.

Fortunately, it wasn't the "payload" questions. He swapped that control question out for a different one and declared me truthful. I presented the results to my manager at the same time as I quit.

That experience, though, got me very interested in polygraph examinations and started a hobbyist interest in the entire field and history of lie detection.

That led to me understanding that it's not a thing that is (currently, anyway) actually possible. What is done instead are psychological tricks that very much depend on the examinee believing that the whole thing is legitimate.

No Polygraph discussion is complete without Moe being subjected to the Polygraph...

Eddie: Checks out. OK, sir, you're free to go.

  Moe: Good, 'cause I got a hot date tonight.  [buzz]

       _A_ date.  [buzz]

       Dinner with friends.  [buzz]

       Dinner alone.  [buzz]

       Watching TV alone.  [buzz]

       All right!  I'm going to sit at home and ogle the ladies in the
       Victoria's Secret catalog.  [buzz]

        [weakly] Sears catalog.  [ding]

        [angry] Now would you unhook this already, please?  I don't
       deserve this kind of shabby treatment!  [buzz]

So good, and I love this SMBC take on it: https://www.youtube.com/watch?v=mJS76Bf-ZYo

https://www.youtube.com/watch?v=VdIDwYW_JZg

One of the two best polygraph jokes in the Simpsons! The other, of course, being https://www.youtube.com/watch?v=Or2iZ-mOvf8

This article touches on it but to me the most offensive part of a polygraph is the presumption of guilt if you refuse to take one. This is very much on purpose, of course, given as the author accurately states, its purpose isn't to detect lies, it's a tool used for coercion.

There was a big robbery at a place I worked, luckily I was not a suspect but the FBI came and administered polygraphs to anyone who could have done it. I asked what happens if someone refuses, and the answer was basically "they won't."

Of course it ended up being someone that didn't even work there. The test was clearly for intimidation purposes, at least from the view I had.

Aren't polygraphs not acceptable evidence in the US courts?

Polygraph tests are not usually acceptable as evidence in court.

HOWEVER. Law enforcement can still use them as another tool in their questioning process to produce a confession when there is other evidence pointing to the guilt of the suspect. But of course is only effective against defendants who represent themselves, don't seek their lawyer's advice, or ignore their lawyer's advice.

Polygraphs can be used as evidence, but you would have to be a moron to submit to one, and refusing one cannot be used as evidence of guilt.

I’m very curious about your last sentence. Are you basing this on the fifth amendment, or something else.

FWIW, in PA, it can't be brought up in court as evidence that you refused to take a polygraph. Nationally, we have a right to remain silent, so I imagine it's not much different elsewhere, but IANAL, so there could be some tricks cops/prosecutors can play in other states. There's no connection to the fifth amendment.

This is in line with the joke, "How do get the NYPD to catch a rabbit?"

You ask them, and a week later they bring in a badly beaten bear, who shouts "Okay, I'm a rabbit! I'm a rabbit!!"

That used to be a joke about KGB, or so I thought.

It was

https://imgur.com/fbi-cia-kgb-5koAv3H

It has a bear in it, so probably.

It's (unfortunately) not an American problem. I have firsthand experience that courts are highly prestige-driven and the best way to make your case is have a ton of expensive-looking and official-sounding documents written by 'experts' that support your argument.

Judges are like every high-level decisionmaker ever - the thing they fear the most is publicly being proven wrong so they always go for the safe option where they can share the responsibility of their decisions with 'experts'.

To which countries and which cases are you referring to?

Watching Dexter this 'blood splatter anaysis' seemed beyond ridculous to me. I thought there were no way it would be a thing in the US in the way portrayed.

But then I learn truth detectors are? What more movie tropes are not tropes?

Plenty of techniques are based on junk science, much more in the past than now. Police/FBI don't actively try to find out if techniques are based on real science or not as they are useful in securing convictions. So called experts in these fields testify in trials and are paid quite a bit, it is in their interest to continue being paid so the fraud perpetuates. Unless you have money you have no way to put up a defense that can discredit expert witness testimony.

https://www.theguardian.com/us-news/2022/apr/28/forensics-bi...

Man it's shocking how bad forensic "science" is. If you watch Law & Order you may be convinced that police can detect gun shot residue on a suspects hands/clothing or match shell casings to a specific gun.

It turns out just throwing handcuffs on someone is enough to get a false positive GSR test. And matching a shell casing to a specific gun is essentially impossible.

The false positive rates on forensic "tests" are hard to study because no one has an incentive to, and if you go digging you'll find how bad they can be. Like for example K9 units have about a 50% false detection rate (dog indicated but no drugs/weapons/bombs found). If a cop told you they flipped a coin to decide whether to search a car then the search would be tossed out of evidence immediately!

The vast majority of "Forensic Techniques" used by cops are literally things an ex-cop made up / """observed""" and are now charging insane prices to go across the US, giving presentations to other cops or giving testimony in trials.

It's a factless and truthless system.

Fingerprints are less accurate and reliable than often portrayed in the media.

I'd describe American (govt. agency) polygraphy as a psychological hazing ritual, run by a pretty machismo crowd. Who don't like their beliefs and rituals questioned.

Lots of these. E.g. changing car engine oil every 20 minutes. Although this doesn't extend to the US government -- the army periodically samples oil and changes when it actually needs to be changed.

Situations like these https://www.thehulltruth.com/boating-forum/183583-fuel-tank-... could be the reason to test oil frequently.

The US Army changes oil in their trucks based on the schedule in the Technical Manual, not based on sample analysis.

I discovered the engine oil thing in the past two years.

In the US, I’ve always heard something like every 3 months or 3,000 to 5,000 miles. The cars even warn you in that timeframe.

In Europe, a mechanic was very confused when I talked about an oil change when one was done the prior year. He said it’s more every 2 years or 20,000 to 25,000 kilometers. Once again the cars were set up to warn at that distance.

I’d love to get to the bottom of it. Why such a difference? I can understand the garages wanting the extra business, but why would the car manufacturers go with it?

Free oil changes from dealers are often offered on new cars. Getting you in to the dealer more often may benefit them in some ways:

1) Opportunity to upsell other services,

2) Some people (quite a lot, actually) have alien-to-me car buying habits and might be convinced to trade in their new car after only a year or two and buy another new one if you can just get them into the dealer at the right time.

No clue if that’s why they do it, but maybe.

FWIW, every time I've had my oil changed at the dealership (dozens of times, but only 3 different dealers, so take it with a grain of salt) it's always been done by the service department, with no interaction from sales at all, and therefore no upselling or discussion about trading.

In this case I think the times have changed but old advice has stuck. From my understanding, cars until the 80s or so did need their oil changed this often, but newer cars with EFI and especially if you use synthetic, its no longer necessary to do so. Its been many years since I bought a new car, but IIRC even the mid 2000s you were supposed to get an oil change relatively soon after getting a car as fine particles that weren't entirely machined off should have been worked off in the first thousand miles or so and you were told to get an oil change then.

There is also the case of changing oil for hot and cold seasons- getting thinner oil in the winter and thicker in warmer weather to adjust. I think thats more or less a thing of the past as well, my Honda does not specify/recommend this in the owners manual but perhaps some cars do?

Old adages sometimes stick around forever though- especially when there is money to be made in keeping them alive.

Remember that the manufacturers recommendations are designed around keeping the car operational through whatever warranty period they sell the car with. It's not some sort of ideal program and some benefit exists from changing fluids more often than specified.

Specifically, transmission fluid is often considered 'lifetime' fluid that doesn't need to be changed, and, if you follow that advice, you end up replacing the transmission, an expensive endeavor, whereas if you don't, you can typically double its usable lifetime.

You also have to be careful about engine oil, because the margin of error is very small. Most modern cars have an oil minder, and those work well provided the oil itself is in spec.

>Old adages sometimes stick around forever though- especially when there is money to be made in keeping them alive.

Don't forget to make your swap space double the size of your ram!

Oh if only it was that easy.

Those extremely long intervals will only work for cars that travel long distances, where the engine has a chance to warm-up and burn off any residual gasoline that gets in the oil during cold starts. If said car spends most of its life in a city, doing short trips, the oil gets rapidly dilluted and loses the ability to lubricate the engine properly.

Most modern cars will take all this into account when trying to determine when the next oil change is due. Also, manufacturers nowadays usually specify shorter intervals (6 months or 7500km) for modern direct injection engines that only drive in a city.

Modern Toyota engines require regular oil changes because their tolerances and oil are very thin. You certainly can run them with far less but they won't last nearly as long

With the polygraph used as a universal filtering device for hundreds of thousands of employees in powerful agencies, we end up with a mix of either super honest ones who reveal everything to the polygrapher (i.e. interrogator), or psychopaths who lie through their teeth without showing any physiological signs and passing with flying color.

Next time when you deal with these agencies, as a fun mental exercise, try to figure out which one of the two types you've got in front of you. When it comes to climbing to the top through the ranks, which ones will get there more effectively?

I wonder if it's simpler than that: The examiner is just choosing whether they like you or not.

Officially they have to follow their training, so to speak, and I am sure that's all about how polygraphs are 100% reliable and it's scientific and all that. I wonder if any instructors at some point close the door and tell them "listen, students, yeah, it's all bunk, but we just have to pretend, ok?".

They're apparently allowed to manipulate the subject ad lib.

> I wonder if any instructors at some point close the door and tell them "listen, students, yeah, it's all bunk, but we just have to pretend, ok?".

If online stories are to trusted, this has occurred in a lot of government clearance related situations.

Almost certainly.

Worth noting that the most detestable traitors in those orgs all passed. The one thing they all have in common is that they were granted higher clearance than tons and tons of fine applicants.

Good point. Both Ames and Hansen were passing their polygraphs while giving away secrets and disclosing information to the enemy. That should have given the agencies some pause and made them re-evaluate the usefulness of it all. But, I am sure the paused, re-evaluate, and decided their voodoo magic just needed more effort, so probably increased the effort to polygraph even harder.

Actually now I’m not sure Hansson took a polygraph. The FBI wasn’t keen on using it on their own people in the 80s and 90s. On one instance, he avoided it.

Ames was CIA and speaks from prison with some authority [0], hinting at internal conclusions polygraph is ineffective.

[0] https://sgp.fas.org/othergov/polygraph/ames.html

No wonder he heard from his brother that the Border Patrol were always hiring; they seemingly make it impossible to get hired.

There is a long history of junk science in forensics [1]. Often this junk science is used to reinforce established biases and give bloodthirsty juries a hook to hang a conviction hat on. Now polygraphs now generally aren't admissible in court. This wasn't always the case [2]. Still, law enforcement does use them to eliminate or confirm suspects outside of a court environment. This can just confirming existing biases.

And that's the deeper issue here: the American criminal justice system is mainly retributive or punitive. Long setnences for minor crimes. High conviction rates. Abuse of the power imbalance between prosecutors and defendants. Judges that essentially work for the prosecution (eg [3]). Juries that just want to convict. Of anything.

Yet there's ample evidence none of this is effective whereas something as simple as giving inmates cats to look after is extremely effective at reducing recidivism [4].

My point is that technology can be (and has been) used as a crutch to confirm biases and that criminals are still people and we should treat them as people.

[1]: https://www.propublica.org/article/understanding-junk-scienc...

[2]: https://axeligence.com/polygraph-admissibility-in-united-sta...

[3]: https://www.rollingstone.com/music/music-news/young-thug-ric...

[4]: https://www.indystar.com/story/news/local/indianapolis/2020/...

When we presented our pediatrician with our third child who could urinate in a sink on verbal command at just six months old, she remarked, "We should write an article for a medical journal!" We explained that such an article would never get published because it's not new information; most of Europe begins potty-training at around six months. Delaying this valuable skill until the age of 3-4 years is an enormous waste of resources - but still the whole country was insisting on doing it, don’t know about now.

> most of Europe begins potty-training at around six months

Where did you hear that? Admittedly here in the UK we've been doing our level best to extricate ourselves from the continent, but I've only ever heard of one mum even thinking about it before 18 months. Ours is just over a year and we haven't thought about it at all yet, same with our ante-natal group and friends with slightly older babies.

I googled around a bit and this reddit thread has a lot of Europeans with similar experience to me: https://www.reddit.com/r/Mommit/comments/tdb1f2/what_are_non...

The German term is "Abhalten" (from "halten = to carry") or ("windelfrei = diaper free") and you'll find quite a lot:

https://www.google.com/search?q=baby+abhalten+*.de

Our baby was born potty-trained (which actually means the term is misleading in our case) and a relative started at their childrens' birth.

This sounds different to potty-training, which is where the child knows that they need to wee or poo and tells you or goes straight to the potty to do it. "Abhalten" sounds like the parent training to know rather than the child.

In English this is called "elimination communication" or EC. The parent subconsciously figures out when the baby needs to go based on schedule, observation of milk/water intake, and subtle behavioral cues (facial expression change, posture, etc).

It's a system that works quite well if the baby is exclusively cared for by a stay-at-home parent or a long-term nanny (and no other babies or toddlers in the house to distract the parent/nanny). But try to leave the baby with a new sitter or at a daycare - they have no idea what your baby's cues are and cannot be bothered to learn them; so back in diapers the baby goes.

We have 1 year of paid parental leave (although only ~60% of your salary) in Germany, so almost everyone can do it.

What happens after a year when the parents go back to work? The baby goes to the potty by themselves at that point?

> most of Europe begins potty-training at around six months

Not sure how this relates to the article, but this is news to me (European). We slowly began potty training somewhere between 1 and 2 years. I have never heard of anyone doing potty training at six months. Babies are just barely able to sit upright at that age.

> most of Europe begins potty-training at around six months.

Hailing from Poland; first I hear of this. I know of total of two people who said something like this before - one person is saying a lot of other borderline insane things about parenting, and the other has a business selling webinars around the idea of potty-training kids less than a year old.

>We explained that such an article would never get published because it's not new information; most of Europe begins potty-training at around six months.

Funny. I am European and we have this myth about Asia.

>>urinate [on] command

Wait, what do you think potty training is?

It reminds me of the GK Chesterton short story "The Mistake of the Machine" which suggests it was a very American idea even back then.

https://en.wikisource.org/wiki/The_Wisdom_of_Father_Brown/Th...

The evidence it "works" reminds me of alternative medicine.

I took a polygraph about 30 years ago a Ft. Meade during a background check for NSA. Honestly it took like 45 minutes, guy was totally pleasant and he treated it seriously, but it seemed mechanical, strictly scripted, and more like a box checking exercise.

However, I also had an interview with a psychologist which was very intense and manipulative, but effective, in eliciting true information about how much I was drinking at the time (too much, quite frankly). It was only years later that I realized that was the information they really wanted and all the other embarrassing sex questions were just to set me up.

The sex questions are often the setup. You fill out a ton of background material with one or two sex questions, then the poly has hours of it. People use the word “ambush” and I think it’s useful when they’re trying to build a baseline profile.

The writer is obviously an intelligent person with good critical perspective, so it is surprising that he missed some important material about known countermeasures that can be effectively used to defeat the polygraph. A quick HN search with keywords "lie detector" offers a number of excellent articles on the subject, including discussion of the book "Lie Behind the Lie Detector" (available for free online) which provides detailed information on polygraphs and techniques to subvert the tests.

Sadly, not All-American anymore. The US police force doesn't exist in a vacuum. There is a whole industry of pseudoscientific interrogation techniques that has set itself up in other nations and regressed their policing by decades. Several states in Australia, and I believe parts of the UK, removed the polygraph as a discredited technique and now accept it as evidence once more.

UK here, just looked it up and although it's not admissible as evidence in court, since 2021 it can be used as a requirement for release from prison for domestic abusers.

Bloody hell, that is scary.

That is scary! Well, we've recently got a new prisons minister, this might actually motivate me to write to them about this.

I looked them up: https://en.wikipedia.org/wiki/James_Timpson

Woah. It's actually the MD of Timpson, whom Starmer has parachuted into the role along with a peerage. So an unelected Minister, but apparently he's very into prison reform, so there's that.

Indeed, Timpson's are a leading employer of people who have served their sentence:

> The company is well known for its policy of employing ex-convicts, who make up over 10% of its workforce

Where in Australia?

I've been out of Aus for a few years, but in NSW I understand they're still largely banned (Lie Detectors Act) for employment, courts / evidence, insurance etc (and if they weren't by legislation, they would be for evidence by precedent, which is how the Act came about). As I understood it, this and the Canadian precedent that NSW relied upon, have basically made them a non-starter for courts in Australia ever since.

It's a bit horrifying if they're making a comeback, but then our politicians have always been a bit prone to right wing shock jock rhethoric around election time.

> Why should I be banned from a job for being a small-time delinquent twenty years ago?

Fair question. I'm not sure that's 1) been shown to be true 2) the fault of a polygraph; it's a policy decision not a mechanical one.

> As he pumped up the cuff, Kevin asked if I was comfortable. He didn’t seem to be joking.

Yeah, he probably was serious. Kevin has a job to do and it's to ask you questions and record your responses; not to torture you. Sometimes people really forget that not every examiner thinks you're the next Aldrich Ames.

But I mean also if you're in a very uncomfortable situation and somebody asks you how you are, maybe being honest will help resolve that situation? Like clearly you can't resolve it on your own otherwise you wouldn't be in such a panic.

> Except for the last two, all of them seemed open to interpretation. For instance, I absolutely had misrepresented my past drug use, but only the number of times, not the drugs or the fact of doing them.

If only there was somebody that could've helped you determine what the answer should be. Maybe Kevin?

---

I mean yeah, if you go into a polygraph expecting a fight I don't think it's fairly surprising you don't get the job. Who wants to hire somebody combative to your employees? Sure, it's unfair but you're taking it out on Kevin who didn't come up with it.

Maybe you misinterpreted the intent of the article - isn't the purpose of it to point out absurdities and fundamental flaws in the uniquely American polygraph obsession?

Who wants to work for or with someone who pretends they can read your mind?

UK police increasingly using polygraph tests - https://news.ycombinator.com/item?id=39161771 - Jan 2024 (5 comments)

Spanish police plans to extend use of its lie-detector while efficacy is unclear - https://news.ycombinator.com/item?id=24905780 - Oct 2020 (38 comments)

Accused spy Alexander Yuk Ching Ma evidently beat the polygraph - https://news.ycombinator.com/item?id=24197310 - Aug 2020 (186 comments)

We tested Europe’s new lie detector for travelers and triggered a false positive - https://news.ycombinator.com/item?id=21358288 - Oct 2019 (114 comments)

Why Lie Detector Tests Can’t Be Trusted - https://news.ycombinator.com/item?id=20556201 - July 2019 (190 comments)

Attempts to Censor AntiPolygraph.org - https://news.ycombinator.com/item?id=20311040 - June 2019 (45 comments)

The Lie Behind the Lie Detector - https://news.ycombinator.com/item?id=18431683 - Nov 2018 (96 comments)

An AI Lie Detector Is Going to Start Questioning Travelers in the EU - https://news.ycombinator.com/item?id=18351733 - Nov 2018 (202 comments)

Personal Statement of a CIA Analyst - https://news.ycombinator.com/item?id=18155548 - Oct 2018 (104 comments)

The Lie Generator: Inside the Black Mirror World of Polygraph Job Screenings - https://news.ycombinator.com/item?id=18120270 - Oct 2018 (95 comments)

NCCA Polygraph Countermeasure Course Files Leaked - https://news.ycombinator.com/item?id=17277049 - June 2018 (4 comments)

Do Polygraphs Actually Work? - https://news.ycombinator.com/item?id=12951926 - Nov 2016 (4 comments)

An Ex-Cop's War on Lie Detectors - https://news.ycombinator.com/item?id=10002889 - Aug 2015 (64 comments)

How to Beat a Polygraph Test - https://news.ycombinator.com/item?id=9481385 - May 2015 (92 comments)

Man accused of teaching people to beat lie detector tests faces prison - https://news.ycombinator.com/item?id=6308878 - Sept 2013 (152 comments)

The Lie Behind the Lie Detector [pdf] - https://news.ycombinator.com/item?id=6307479 - Aug 2013 (22 comments)

Seeing threats, feds target instructors of polygraph-beating methods - https://news.ycombinator.com/item?id=6229185 - Aug 2013 (4 comments)

All lies? Scientists threatened with legal action over lie detector article - https://news.ycombinator.com/item?id=457996 - Jan 2009 (1 comment)

My NSA polygraph experiences - https://news.ycombinator.com/item?id=428489 - Jan 2009 (46 comments)

also https://news.ycombinator.com/threads?id=giles_corey

Keep your propaganda to yourself... the polygraph is a scientific tool... it's been proven the body goes through physiological changes when humans lie

How so?

A wonderfully written piece.

It’s not a lie if you believe it. - George Costanza

I watch a lot of true crime programs and the polygraph is mostly used as justification to harass innocent people. The police consider the refusal to take it as a tacit admittance of guilt, despite the fact it can't be used in court. It's all shit.

Polygraphs are b.s. pseudoscience, unreliable, and subjective.

Polygraph "examiners" are like drug dogs who always fail to find drugs because, since they do not find the drugs, they receive no praise and they do not receive their "toy."

I never understood why people believed this BS. The process of administering an exam requires direct interpretation from another human. They are just as fallible as I.

More so. Becoming a polygraph examiner is possible with problems in your background that would exclude you from Public Trust. One guy in Florida was drugging and fondling people taking the test.

One of my college professors was adamantly against Polygraphs, he made the case that over time it causes sociopaths to collect at whatever work address requires them because they can pass effortlessly and normal people on average get stung by false positives.

I used to think this, but learned it filters for three modes:

1. The average person who hesitantly wishes that the examiner will supportively pursue establishing the true state of his/her personality and the World at large.

2. Someone entirely lacking empathy, unable to value the examiner as someone seeking truth.

3. Those who effortlessly pass, every time.

I think the reason three-letter-agencies maintain such expensive theater is to discover and protect #3.

Not a single comment in here connecting the fake bullshit detection technology of the previous generation to the present bullshit generation technology (“AI”)? I’m disappointed, y’all.

my experience with the polygraph was it detects your ability to submit. it's less a delusion than a ritual. as a filter, it produces a completely polarized bimodal distribution of banal followers vs. the basest human malevolence and contempt, while disqualifying relfection or doubt. if the purpose of a system is what it does the polygraph is perfect. it finds expendable souls.

That’s about as well founded of a statement as the idea that polygraphs are lie detectors.

You can add dissociative identity disorder, a.k.a. multiple personality disorder, to the list of All-American delusions.

The author is uncooperative throughout the whole test, fails it, then claims the polygraph is quackery.

the only lack of cooperation communicated in this article is the account on drug use. that said, the claims on quackery are unrelated to the author's specific set of behaviors during the test. i reckon that i, knowing the complete lack of scientific backing, would also be uncooperative to some degree.

Writing mostly for the sake of writing.

That’s not to say it wasn’t interesting; but it could have been 1/3 the length and provided the same information value. It’s as if the author wanted us to experience our own suffering, wondering when “the rest of the story” would finally be delivered.

Reply on news.ycombinator.com