Why is lie detector technology not considered viable in a USA court of law?

It honestly doesn’t make much sense to me. Here we have this awesome technology that is able to literally tell whether a defendant is lying or not. Why aren’t using it for anything? The only two arguments in support of banning them in courts are that:

  1. They are unreliable.

Sure, they are not 100% perfect. But no technology is. We get false positives all the time in DNA analyses, fingerprints, witness statements, or wrongly attributed evidence. What makes lie-detecting technology different?

  1. They violate a defendant’s right to privacy.

This one is a little more understandable. However, there is one problem with this argument: The same can be said about the entire prosecution process. If refusing to speak under oath isn’t a crime for the goverment, why is being forced to use a lie detect one?

With all of that out of the way: Why exactly is lie detector technology not considered viable in a USA court of law?

They are more than unreliable. They don’t work at all. The results are no better than guesswork.

Polygraphy is pseudoscience at best.

From here: Do Lie Detector Tests Really Work? | Psychology Today
we learn that " The American Polygraph Association is the world’s leading association dedicated to the use of evidence-based scientific methods for credibility assessment. It is an organization whose members are largely polygraph examiners. They estimate the accuracy of the polygraph to be 87%. That is, in 87 out of 100 cases, the polygraph can accurately determine if someone is lying or telling the truth.

That sounds pretty impressive, but it is important to keep in mind that the polygraph is failing 13% of the time. The federal government sought an unbiased evaluation of the polygraph, so they tasked the National Academy of Sciences with a full investigation of the polygraph’s accuracy. In 2003, this large team of notable scientists came to the conclusion that the polygraph was far less accurate than the polygraph examiners had claimed. Some scientists have claimed that the accuracy may be closer to 75%. This lackluster performance is the reason why polygraphs are not used as evidence in criminal trials. They just cannot be trusted."

That is the problem right there: No, you do not have this technology, this technology you don’t have is not awesome, it cannot tell whether somebody is lying or not.
The detail that this “somebody” is for you literally automatically a defendant is a worrying sign in my eyes.

The real question is why polygraphs are being used at all. According to Wikipedia,

US law enforcement] and federal government agencies such as the FBI, NSA, CIA, and many police departments such use polygraph examinations to interrogate suspects and screen new employees.


I think that, if not used for evil, such tech could be incredibly useful for helping to bring justice.

Also, what is wrong with assuming that lie detecting technically would be used primarily on defendants?

It’s not a lie detector. It’s a stress detector. If a question stresses you out, you can have the exact same reaction as someone who is lying. It’s very easy to have a false positive.

With a bit of training, you can easily defeat one.

Here are some statistics from various tests that have been done to assess their accuracy:

  • Six prior reviews of field studies:
    • average accuracy ranged from 64 to 98 percent.
  • Ten individual field studies:
    • correct guilty detections ranged from 70.6 to 98.6 percent and averaged 86.3 percent;
    • correct innocent detections ranged from 12.5 to 94.1 percent and averaged 76 percent;
    • false positive rate (innocent persons found deceptive) ranged from O to 75 percent and averaged 19.1 percent; and
    • false negative rate (guilty persons found nondeceptive) ranged from O to 29.4 percent and averaged 10.2 percent.
  • Fourteen individual analog studies:
    • correct guilty detections ranged from 35.4 to 100 percent and averaged 63.7 percent;
    • correct innocent detections ranged from 32 to 91 percent and averaged 57.9 percent;
    • false positives ranged from 2 to 50.7 percent and averaged 14.1 percent; and
    • false negatives ranged from O to 28.7 percent and averaged 10.4 percent.

(Statistics are from the Office of Technology Assessment from an evaluation done in the 1980s)

Those numbers are abysmal. Sometimes the machines are accurate. Sometimes they aren’t. And when they aren’t, they are way off.

So, answer this. If you are innocent, would you be willing to face a lie detector that might have up to a 75 percent chance of wrongly convicting you? More than likely it’s going to convict you. Do you want that to be admissible in court?

Or do you want a guilty person to have as little as a 35 percent chance of being caught? Do you think that is a good enough rate to be admissible in court? Your guilty criminal might have as much as a 65 percent chance of beating the machine.

Something that is this ridiculously unreliable is most definitely not “incredibly useful” for helping to bring justice.

Granted that lie detectors are junk science and don’t work. I am nevertheless sceptical of surveys that assess their accuracy rating. How was the accuracy determined?

For example:

Just for a start it isn’t a ‘guilt’ detector, it’s a ‘lie’ detector. Somebody might lie without being guilty.

Second, it’s ambiguous. There are at least two possible meanings.

  1. out of every 1000 ‘guilty’ people, the machine detected 863, and missed 137.
  2. out of 1000 people identified as ‘guilty’, 863 were really guilty, and 137 were wrongly accused.

There’s a big difference between the two.

Third, and most important, how did they determine the actual list of errors objectively? If they list 100 lies the machine registered as true, how do they know they are lies in the first place? If they list 100 true statements the machine registered as lies, how can they be sure that they are true? And can they be sure they caught ALL the mistakes? Maybe there were another 20 errors that they didn’t spot.

I also have this awesome deck of cards that will foretell a person’s entire future life. Why doesn’t the government use it in schools to steer people to the correct colleges and careers?

What’s wrong with that is that defendants don’t have to answer any questions , whether they are hooked up to a polygraph or not. And also because any witness can be lying , so if they were reliable it really should be the witnesses getting hooked up, not the defendant.

I would have thought of witnesses, for instance. There are usually more witnesses than defendants in any normal case. So if you think first and foremost of the accused when you think about finding out the truth, and then your aim is

then I sense some lust for condemning the evil perpetrator. That does not pass my “beware of vigilantism” smell test, I’m sorry to say.

I could see working a polygraph into a TV game show. Hook it to the contestant or to the Host when he/she says, “Door number one, door number two, or door number three?” [the Monty Hall dilemma].

A court of law, nope nope nope.

I actually think this gets at a really interesting question: Why has the polygraph industry been so unsuccessful at pushing its stuff into courtrooms when other junk science has managed to get in?

They’ve been suspected of lying about the effectiveness but there’s no way to be sure?

How do you feel about the principle of presumption of innocence, OP?

How do you feel about Blackstone’s statement that:

Fourthly, all presumptive evidence of felony should be admitted cautiously, for the law holds that it is better that ten guilty persons escape than that one innocent suffer.

How do you think a jury should deal with a (if we take the results quoted above at face value) 1/4 chance that a polygraph result is inaccurate?

Presuming you have a answer to the previous question, do you think the average juror could actually adhere to the framework you suggest.

A polygraph is great for determining if you’re lying… about having an orgasm.

Why would anyone even think that an orgasm detector would be a useful tool in the justice system?

I was going to say, if we don’t allow lie detectors then why do we still allow drug-dog indications as legal justifications for searches?

I doubt that the FBI etc. are under any illusion about their accuracy (although I’d be less confident about regular police). But there may be some validity in their use as an investigative tool that relates to the exact reason that they cannot be used as evidence. Many people believe they work, and are intimidated by a machine. They may have no value in exposing liars who continue to lie, but they may cause someone to reveal something in the belief that their lies will be exposed.

Caveat: polygraphs are not effective detectors of lies, but they can be an effective interrogation tool. If the suspect believes polygraphs detect lies, and the suspect may be punished or penalized for lying, then the suspect is more likely to tell the truth.

Some jobs, especially government jobs, require applicants to submit to a lie detector test. I don’t have evidence for this, but I believe the main purpose is to prime employees to believe that polygraphs are effective, so that they are more likely to tell the truth if threatened with one.