Re:Man accused of teaching people to beat lie detector tests...
by mistercow on Sunday September 01, 2013 @02:17AM
(#6309324)
>So how can you know how to cheat them?
I don't understand the basis of your question. To say "this test does not correlate with the phenomenon it purports to measure" is not the same as to say "this test is completely random".
Counting how many times a person says the word "baby" during an answer is a piss-poor way to tell if they're lying, in large part because that factor won't correlate strongly with deception. It would also be extremely easy to fool, by simply controlling how many times you say "baby".
Re:Man accused of teaching people to beat lie detector tests...
by Bulkington on Sunday September 01, 2013 @09:31AM
(#6310273)
Didn't mean to say polygraphs are random, but that the US jusice system is so out whack that once someone is under suspicion, conviction (for something) has little to do with guilt or innocence on initial grounds.
So how can you know how to cheat them?
But if they are any good, then surely the technicians recognize false negatives. "Hah! You're trying to cheat the polygraph! That's a crime!"
It ain't Kafka anymore. It's USA. I can't believe I used to worry about redneck country cops and a few days taking it in the pokey. Might as well just take your chances with a RND function if you're detained now. Of course, if you're a young black male the algo is weighed heavily against you. Money buys a lot of decimal places.