A video about the delayed choice eraser that I made

[Strilanc]

This was my attempt at making an explanatory video discussing some of the misconceptions about the experiment.

I'm looking for feedback on how it could have been better. In particular, based on some of the confusions in the comments on the video, it's clear to me that I should have e.g. gone through Kim et al's experiment step by step to better establish what the quantum experiment is and how they relate to the breakdown that I'm giving.

 

[pinball1970]

Very good for a non physicist if that is worth anything. I read about this in Jim Al Kalilis book "Quantum, a guide for the perplexed," but this explains this a lot better.

 

[vanhees71]

It is very important to understand that the coin-die analogue is NOT a true analogue of the quantum situation, because in the latter case the single-photon observables are not determined but maximally uncertain, because the photon pairs are prepared in Bell states.

The message is of course correct: There's indeed nothing violating causality and there are no retrocausal interactions. It's far from incomprehensible, because the very fact that physicists are able to conceptualize such a delayed-choice experiment using quantum theory (Wheeler) and also to realize it in the lab with high accuracy (e.g., Kim et al) shows that these physicists very well understood what's going on. If they'd not they couldn't have made the experiment at all let alone predict precisely the outcome of the experiment.

 

[Strilanc]

It is very important to understand that the coin-die analogue is NOT a true analogue of the quantum situation, because in the latter case the single-photon observables are not determined but maximally uncertain, because the photon pairs are prepared in Bell states.

In the classical analogue, Bob and Charles' initial estimates of the boxes contents are also maximally uncertain.

I understand that entangled states are different from probabilistic states. They are observably distinguishable using e.g. Bell inequality tests. But the DCQE doesn't do anything that forces entanglement. In principle you could prepare a probabilistic distribution of separable photons, insert them right after the BBO crystal, and get the same result.

Here's something related that bugs me. In a Bell inequality test, delaying one of the parties' choices is considered to be a flaw. It introduces the possibility of communication. So loophole free Bell inequality tests ensure the choices are spacelike separated. The DCQE takes this flaw and actually flaunts it. As if it were a strength! It's not the spacelike-separated eraser, it's the delayed eraser. It's so close to being a slam-dunk can't-do-this-classically experiment, but then it goes and defines itself around behaviors and loopholes that make local classical analogues possible.

The message is of course correct: There's indeed nothing violating causality and there are no retrocausal interactions. It's far from incomprehensible, because the very fact that physicists are able to conceptualize such a delayed-choice experiment using quantum theory (Wheeler) and also to realize it in the lab with high accuracy (e.g., Kim et al) shows that these physicists very well understood what's going on. If they'd not they couldn't have made the experiment at all let alone predict precisely the outcome of the experiment.

Thanks.

 

[vanhees71]

The difference is that in the classical analogue the coin and the die are correlated because Alice has sorted them according to values which are then predetermined. In the QT case you have a Bell state, where the two-photon system is completely determined (pure state), but the single-photon observables are maximally indetermined (the reduced states are maximum-entropy mixed states).

Of course, the analogy is correct in the sense that there are no actions at a distance and no violations of causality necessary to explain the correlations, because the correlations are prepared in both cases in the very beginning by preparing the dice and coins by Alice or preparing a Bell-state of entangled photon pairs by parametric down conversion.