Comment by seibelj | Hacker News Reader

Comment by seibelj | Hacker News Reader

I'm a huge fan of true crime, listening to thousands of hours of podcasts, endless TV shows and documentaries, I own at least 30 true crime nonfiction books, etc. But I don't have any formal credentials in criminology.

After all my listening, I believe that if I was law enforcement I would appreciate the lie detector. It doesn't actually detect lies, and a trained person or a sociopath 100% in control of their emotions and heart rate can beat it. And many people have.

However, the power in the lie detector comes from its name - "The Lie Detector" - and the common belief that it actually works.

The interesting result of a lie detector isn't that it actually detects lies. No matter what the result is, if I was an investigator, I would tell the suspect that it said they lied and it is 100% accurate. The reaction to telling the suspect this, and what they do next, could very well result in a confession, or at least meaningful information. It's a psychological trick that is quite useful. The lie detector is such an absurd piece of theater, and most people have never seen one before let alone had one taken, that they cannot prepare for the result. If you interrogate 100 suspects, you can use their reaction to further your opinion of whether they are truly guilty by comparing to your past interrogations.

Now, would I ever agree to take a lie detector test? Hell no! And they are also not admissible in court, as the law understands that they don't actually work.

But in an interrogation process, where cops are lying and deceiving and tactically revealing information, it is a good arrow in the quiver.

Firstly, who has more reason to research lie detectors and realise they don't work, a criminal or an innocent person? Clearly, it's the former, so you're using a "trick" that is weighted against innocent people.

Secondly, most criminal justice systems reward early confession with reduced sentences and plea bargaining. In such systems, presenting a suspect with fabricated evidence—which is what a "positive" lie detection report is—seems just as likely to induce a false confession as a real one.

If we allowed cops to plant evidence and falsify statements, we could probably get a lot more confessions. If we eliminated the "innocent" verdict at trial, or reintroduced trial by ordeal, we could do the same. But we don't do this because these are fundamentally unjust, and so is allowing law enforcement to claim they know people are guilty due to magic.

The media believes it works, so not agreeing to take one can be used in the court of public opinion to ruin a reputation.

For example there's currently a women missing in Hawaii this week called Amanda Eller, her boyfriend was given a lie detector (which he "passed"), the police announced he wasn't a suspect and his test pass. Imagine if he had said "hell no!" as you yourself would.

That's the problem. Police keep using something that doesn't work, but if you refuse they will take the refusal itself as suspicious and the media will find out. It may not hold up in a court of law, but good luck ever getting another job with google returning articles about how you MAY have killed someone.

> However, the power in the lie detector comes from its name - "The Lie Detector" - and the common belief that it actually works.

Ha. It reminds me of this hilarious video clip (https://www.youtube.com/watch?v=DgrO_rAaiq0) from The Wire, S5E01, which Detective William "Bunk" Moreland interrogated DeShawn and made him believe the office copy machine was a lie detector. This plot itself was based on an false urban legend.

> [...] they are also not admissible in court, as the law understands that they don't actually work. But in an interrogation process [...] it is a good arrow in the quiver.

I think the first things come to my mind is: what types of psychological tricks are and aren't allowed in an interrogation process? Is it permissible to use a nonworking "lie detector" as an interrogation technique? What are the potential abuses of this trick, etc? These are some interesting things that can be argued and debated.

I don't have background knowledge any of these subjects I mentioned so I won't start a debate, other interesting may want to do so.

Anyway, interesting perspective. I appreciate your comment, which moved the question from the first layer: "does a polygraph work and whether we should use it", to the next layer: "a polygraph doesn't work, but should we use it anyway?"

> However, the power in the lie detector comes from its name - "The Lie Detector" - and the common belief that it actually works.

In this case no need develop and pruchase a non fuctioning product that may or may not be abused by the legal system, just make up a story: that little box in the corner is a "100% accurate voice lie detector" and be done with it.

"The reaction to telling the suspect this, and what they do next, could very well result in a confession"

... sure, you can extract "confessions" using all manner of lies and pressure tactics like this. A person who is persuaded that the machine is 100% accurate might just decide there's no way out and play along with confessing in the moment because they have no other alternative and it's a very stressful and unusual situation to be in. These are not what I'd call reliable confessions or meaningful information though.

The Forensic Brain Wave Analysis (FBWA)is supposedly nearly foolproof. Although anything that's declared to be infallible makes my BS detector go off.

that puts too much power in the hands of the police... like a drug dog that doesn't smell drugs, but alerts when the handler "has a feeling" and gives the dog a surreptitious clue.

nothing like that belongs in the legal system.

we need facts and evidence; not suspicions, inuendo, "hunches", and witch hunts.

Reply on news.ycombinator.com