[thrance]

You want to read up on Bell's Inequalities, which were experimentally confirmed[1] by Alain Aspect who won the physics nobel prize for it.

In short, you can never properly describe what happens by assuming the particles states are already determined after being entangled.

[1] https://en.wikipedia.org/wiki/Aspect%27s_experiment?wprov=sf...

[newpavlov]

There is a very important and often forgotten caveat to the Bell's inequalities: it covers local hidden states. If we are to assume that measurement and particle creation apparatuses are entangled with each other (after all, they had plenty of opportunities since the big bang), then we have a global hidden state and the quantum measurment randomness becomes a simple artifact of removing global state from the picture. This interpretation of QM is usually called "superdeterminism" and, personally, I like it much more than the black vodoo measurement magic with collapsing wave functions or creation of whole new worlds on each tiny measurement. This video can be a good introduction to this topic: https://youtube.com/watch?v=dEaecUuEqfc (don't mind the clickbaity title, the video itself is good)

[rcxdude]

superdeterminism is deeply weird, just in a different way that most of the other quantum mechanics interpretations are deeply weird. The main thing is that it implies that this global quantum state is just-so arranged that no matter how you make your decisions about what to measure, it's always correlated with the underlying quantum state.

[newpavlov]

I would say it differently. Suprdeterminism opposes to the deeply ingrained assumption that we can design experiments in a way which removes influence of a measurement apparatus (including experementators themselves) on the measured process.

[rcxdude]

It's more than that, though. Just 'having an influence' on the measured process doesn't explain the bell inequality. Super-determinism basically requires that there is some common state from the big-bang which means that if I were to decide to e.g. seed the random number generator I'm using in an experiment with a description of what I had for breakfast that morning, the particles in that experiment (which could in principle come from far enough away they had no way of causally interacting with me or said breakfast) somehow 'know' that I had made that decision, what I had for breakfast, and the details of the random number generator and act accordingly. Absent some mechanism by which this might occur, it requires an incredibly complex kind of setup to the universe to create that result, one that has so many free variables it could explain almost any universe with any physics.

[newpavlov]

It's not a some sort of particle conspiracy. The idea is not so different from the Laplace's demon. We have an initial state of the Universe at the moment of Big Bang (a PRNG seed, if you will) and a set of differential equations (QM is not different in this regard). Theoretically, it allows the demon to predict everything in the Universe. The wave nature of QM equations introduces a certain quirk to it, but, effectively, with your example the breakfast was already "preordained" at the moment of the Universe creation.

Surprisingly, this idea makes many physicists very umcomfortable and they start to object to SD using philosophical arguments about "free will".

[rcxdude]

It should make anyone uncomfortable (a trait it shares with all other known interpretations of QM). It implies a degree of correlelation across many different levels of abstraction which basically nothing else in physics does. As the name implies it's not just an abstract sense of determinism but one which tips the scales of everything at every level towards a specific outcome.

[mrguyorama]

And since that precludes it from ever being testable, falsifiable, or making predictions different from a Bell's Inequality based theory, it just isn't physics

[newpavlov]

It's no different from any other interpretation.

[simonh]

Just to try and summarise the issue. It turns out measurements of one of the entangled particles (particle A) are correlated to the settings of the measurement apparatus. For example the axis on which the polarisation of a photon is measured affects the measurement you get.

That setting is not known when the particles become entangled, and so in principle cannot affect the state of particle B. However since the setting does in fact correlate with the measured state of particle A, it also correlates with the state of particle B.