Live data from Hacker News

Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

antipolygraph.org

151–160 of 196 posts

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#151

The polygraph is pseudoscience. Of course it can be beaten, it doesn't work to begin with. Why the US still puts so much weight on them when they know they don't work is a mystery to me. Just get rid of the damn things and do your homework better.

Reminds me of this scene from The Wire

(NSFW) https://www.youtube.com/watch?v=AJ5aIvjNgao

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#153
post #76
post #74

I know we all know that polygraphs don't work, but: is that true in practice or in principle? I mean, there are papers out there about methods capable of extracting complete words from brainwaves. Does it really seem impossible for a machine with current tech to detect signals of lying with high certainty?

I think a good liar is just exceptionally good at convincing themselves that what they're saying is the truth. I think that even better liars are so narcissistic that they basically believe what they say as it's coming out of their mouth, on account of it being them that is the one saying it. I say all of this as a horrible liar.

There are a lot of different ways to lie, and the method you point out is one of many.

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#154

Polygraphs are a joke for this reason among so many others: Psychopaths, sociopaths and narcissists can fool them because they are able to feel zero (or are incapable of feeling any) biological fear or anxiety (i.e. 'stress response'), while telling complete lies.

Realistically, people who have to take them regularly can also learn to do “better” at them, as you can with any “test.”

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#155
The polygraph was largely invented by William Moulton Marston, the same guy who invented Wonder Woman. It was clearly a farce when he was selling this idea 100 years ago, and it's amazing that it's considered anything short of snake oil today.

There's a great book on the inventor called the Secret History of Wonder Woman: https://www.amazon.com/Secret-History-Wonder-Woman/dp/080417...

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#156
post #23

50 percent chance an honest person barred for life. 50 percent chance a criminal gets let through. If that’s true it’s a wonder the FBI managed to get this far and continues to function.

In a bit skeptical of the 50% number mentioned. I think the newspaper article might be misquoting the source. I know people who have failed lifestyle polygraphs and weren't banned for life, they just had to try again

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#157

Earlier quoted context omitted.

It actually makes sense for law enforcement even knowing the device does nothing because it can be used to give them probable cause to search somewhere they wouldn't otherwise legally be able to search. When the shit hits the fan the company that sold it takes the fall, which is part of what the police were paying for. Perhaps a similar logic is at work for the military, but one would think they'd be more concerned w…

Just like those dogs that indicate the presence of drugs. Except of course when there aren't any. I've had my car searched a couple of times at the Canadian/American border because a dog gave a signal. Given the number of times I've crossed that border the false positive rate must be ridiculously high. Since you don't have much in terms of rights when crossing a border I wondered why they even bother with the pretext…

> Since you don't have much in terms of rights when crossing a border I wondered why they even bother with the pretext.

To find other "bycatch" - cash to confiscate, for example, or the gun that might be forgotten after the last hunting ride, or a pack of medicine without a prescription.

In the end it's a game of probability - assume that out of 1000 cars 1% (10 cars) have something that's not an illegal drug, but still illegal. Now, police are usually banned from warrant-less searches of vehicles for a good reason, so assuming they would follow their own rules, they would possibly only search the one car of the dude with red eyes, and the other potential hits would be missed.

Now, with a "trained dog", they have the legal authority to conduct a warrant-less search - which in turn enables more insidious stuff such as racial profiling the drivers, or simply do a rough check of all cars.

"Drug dogs" are nothing more than an instrument of abuse.

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#158

The polygraph is pseudoscience. Of course it can be beaten, it doesn't work to begin with. Why the US still puts so much weight on them when they know they don't work is a mystery to me. Just get rid of the damn things and do your homework better.

The issue is accuracy. In terms of being able to infer lies from skin response, this is an effect that actually exists for many people. The net scientific conclusion is that can work but poorly, not that is is complete nonsense like ESP or something. Being pseudoscience or not is not quite the same question. People claim acupuncture works for them (California is now forcing insurance to pay for it), but it is clearly…

Astrology, Tarot, and Fortune Telling are all pseudoscience, but a good practitioner is going to be very good at Cold Reading.

http://skepdic.com/coldread.html

It works as long as the "mark" is naive. As James Randi explained many times.

Which is why intelligence communities want to suppress information about polygraphs that would make people less naive.

Makes perfect sense.

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#159
post #74

I know we all know that polygraphs don't work, but: is that true in practice or in principle? I mean, there are papers out there about methods capable of extracting complete words from brainwaves. Does it really seem impossible for a machine with current tech to detect signals of lying with high certainty?

I think at some point you might be able to detect lying with high accuracy given a DNA sample, a high resolution/high fps camera pointed at the person's face, and a vocoder-generated questioning regime.

However, it would be very expensive to develop the training data for such a system, and it might not work on rare ethnicities. Or other abnormal people, like autists and evil geniuses.

Re: Accused spy Alexander Yuk Ching Ma evidently beat the polygraph

#160
post #87

The polygraph is pseudoscience. Of course it can be beaten, it doesn't work to begin with. Why the US still puts so much weight on them when they know they don't work is a mystery to me. Just get rid of the damn things and do your homework better.

Let us never forget this infamous explosives 'detector': https://en.wikipedia.org/wiki/ADE_651 Sold to law enforcement in military around the world for a while.

Someone I knew involved in defense procurement explained the proliferation of such devices. Basically these small companies who make these things will hound politicians saying they've got a device that can save soldiers/police lives but your bureaucrats who handle supply chain sourcing and testing are slowing us down - please help us accelerate this and we can save lives. They'll usually bring up that this is a small company who is competing with the military industrial complex keeping them out of the game. This usually results in one or multiple politicians pressuring the applicable agency, to the point where they'll threaten to withhold funding to consider a trial or small purchase. Fast forward a year and these devices are deployed and thrown away due to the fact they obviously don't work. It's worse in the law enforcement circles where they actually believe the stuff until a court gets involved and things get thrown out and guidance is given to drop the crap.
Post reply on HN