Question

I was reading the Feynman lectures in physics and after thinking about it for a while it seems particularly unreasonable to talk about hidden variables. Let us say that the electron has some internal variables as yet unknown which determine its trajectory given a set of initial conditions just like in classical mechanics. But since these hidden variables are unobserved, coupling it with a classical system should make their effect unchanged. This is what Feynman says, I think, in the last paragraph of Ch1 Vol 3, that if in the double slit experiment, if these inner variables dictate that the electron goes through the upper slit and land at a particular place on the opposite screen, and some other place for the lower screen, then the probability must neccesarily be the sum of two Gaussian like peaks, which does not agree with experiment.

So if I concluded that inner workings of an electron had some additional hidden variables, then it should yield, as they should be independent of the classical apparatus, mutually exclusive probabilities that do not quiet add up the way as observed. But then I do a hidden variables search on the archive and a lot of smart guys still write about it, as late as Feb 2011.

So the argument I have used might be somehow incomplete, can anyone explain how?

EDIT: Sorry for editing this question almost three years later. I tried to locate the exact reference from the Feynman lectures I was referring to and this is the updated source, Sec 7 Ch 1 Vol 3

We make now a few remarks on a suggestion that has sometimes been made to try to avoid the description we have given:

Answer 1

I am afraid that this question of yours, while excellent, could belong to a psychology forum.

Feynman's perspective was already totally sensible; however, since his death, an amazing sequence of ever more straightforward proofs that hidden variables can't exist has been found.

Let me mention the GHZM state, Hardy's "paradox",

Weak measurement and Hardy's paradox.

and various experimentally realized thought experiments disproving various theories of "nonlocal realism" as well, see e.g.

http://arxiv.org/abs/0704.2529

The idea that the probabilistic nature of quantum mechanics is due to the hidden variables was indefensible when Feynman was writing his lectures in the 1960s, and maybe even in the late 1920s when quantum mechanics became settled. But it is much more indefensible today.

I am not sure whether I fully understand your proof but it is possible to describe the experimentally validated proof in similarly simple terms. Quantum mechanics remains unacceptable for many people, for quasi-religious reasons. But as Sidney Coleman said at the end of his excellent lecture,

http://motls.blogspot.com/2010/11/sidney-coleman-quantum-mechanics-in.html
Quantum mechanics in your face

in which he described the conceptual issues of QM very clearly, it's time for those people to seriously consider the hypothesis that quantum mechanics is really how the world works. Well, the lecture is 17 years old by now, too, indicating that the opposition to quantum mechanics will never disappear