To the real world may lead to interesting findings. A memory researcher started testing eyewitness reliability and found it very shaky. Among other things.
Probably this had been known by lawyers for quite a long time, but they just used it to their advantage.
gl,
I’ve actually heard about this for many years. The problem is that all evidence tends to have issues with it. Part of the reason why there are supposed to be multiple independent witnesses and pieces of evidence. And why at least in the US, “innocent until proven guilty” beyond a reasonable doubt is supposed to be how things work. People are fallible, even a lot of evidence from “scientific” approaches tends to have an alarming high rate of false positives/negatives.
Of course, you’d never know this from watching TV or listening to prosecutors….
~Jon
Yeah, my post may have been unclear, this all happened in the nineties originally AFAIK.
To deal with false positives correctly etc in medical and judicial testing, you need correct Bayesian inference, otherwise you might end up with very wrong conclusions or impressions.