There's an experiment. "Quantum eraser". This is "me asking advice", I don't understand it to explain it.

It involves, producing two entangled photons, and doing the double-slit experiment on one of them with a different polarisation-changing filter over each slit. Repeat lots of times and see if you get an interference pattern, or actually not, because the polarisation-changing filters make the photon not destructively-interfere with itself (because the two states "at this point coming from slot A" and "at this point coming from slot B" are no longer exactly the same).

The mysterious bit is, if you put a linear polarisation filter in front of the *other* photon, this ruins the polarisation and the interference pattern goes away. Which looks like a specific physical effect of waveform collapse. People go to lots of effort to make sure that the same effect applies if you make the path between the other entangled photon and the "linear polarising filter or not" really long, so you make that choice *after* the other photon hits the screen, and yet, still seems to affect it.

This seems really mysterious. In fact, it sounds so mysterious it's actually impossible.

But what I was missing was, every diagram has a "coincidence counter" which only counts photons if one from each path both arrive (at the same time, if the paths are the same length, or at corresponding times otherwise). This seems like a standard precaution, to ensure you're only counting the actual photos, and not stray cosmic rays or whatever.

And yet, normal two-slit experiments don't (seem to?) need to use one.

And specifically, the linear polarising filter *throws away* half the photons, which means that at the screen you DON'T get an interference pattern. Whereas if you only look at the half of the photons which correspond to ones which passed the linear polarising filter, then you DO. (If you look at the OTHER half of the photons, you'll see an opposite interference pattern, which adds up to a smooth non-banded pattern of photons if you overlay the two halves).

What actually happens does (as always) correspond to "things only interfere if they're smeared out over multiple potential possible values (in this case two different paths through the slits), if you've already interacted with them, then not". And I don't quite follow what *does* happen because I've not tried to follow the equations. But the whole "mysterious effect travels back in time causing waveform collapse" seems to just not exist, except in how people choose to interpret the experiment.

So, I'm confused, many physicists seem to agree this is important, but I don't quite see how.

And "you get exactly the same experimental results but only look at half of them according to the result of the other entangled particle" seems a really important concept but all explanations seem to leave it out and say "you get a different result" instead. Do I understand that right??

It involves, producing two entangled photons, and doing the double-slit experiment on one of them with a different polarisation-changing filter over each slit. Repeat lots of times and see if you get an interference pattern, or actually not, because the polarisation-changing filters make the photon not destructively-interfere with itself (because the two states "at this point coming from slot A" and "at this point coming from slot B" are no longer exactly the same).

The mysterious bit is, if you put a linear polarisation filter in front of the *other* photon, this ruins the polarisation and the interference pattern goes away. Which looks like a specific physical effect of waveform collapse. People go to lots of effort to make sure that the same effect applies if you make the path between the other entangled photon and the "linear polarising filter or not" really long, so you make that choice *after* the other photon hits the screen, and yet, still seems to affect it.

This seems really mysterious. In fact, it sounds so mysterious it's actually impossible.

But what I was missing was, every diagram has a "coincidence counter" which only counts photons if one from each path both arrive (at the same time, if the paths are the same length, or at corresponding times otherwise). This seems like a standard precaution, to ensure you're only counting the actual photos, and not stray cosmic rays or whatever.

And yet, normal two-slit experiments don't (seem to?) need to use one.

And specifically, the linear polarising filter *throws away* half the photons, which means that at the screen you DON'T get an interference pattern. Whereas if you only look at the half of the photons which correspond to ones which passed the linear polarising filter, then you DO. (If you look at the OTHER half of the photons, you'll see an opposite interference pattern, which adds up to a smooth non-banded pattern of photons if you overlay the two halves).

What actually happens does (as always) correspond to "things only interfere if they're smeared out over multiple potential possible values (in this case two different paths through the slits), if you've already interacted with them, then not". And I don't quite follow what *does* happen because I've not tried to follow the equations. But the whole "mysterious effect travels back in time causing waveform collapse" seems to just not exist, except in how people choose to interpret the experiment.

So, I'm confused, many physicists seem to agree this is important, but I don't quite see how.

And "you get exactly the same experimental results but only look at half of them according to the result of the other entangled particle" seems a really important concept but all explanations seem to leave it out and say "you get a different result" instead. Do I understand that right??

Tags: