Macro state of a measurement device and correlation

[entropy1]

I have a question that seems to reflect my main concern with QM. Here it is:

Consider a series of polarisation-entangled photon pairs that are sent in opposite direction to two measuring devices (e.g. at opposite ends of the universe). The measurement consists of detection of a photon after passing a polarisation filter. The results are later compared. Suppose the measuring devices can't communicate the results in any way prior to comparison.

The devices are macro-objects. The measurement results of either devices are read from the dial (computer screen) by Alice and Bob respectively, thereby experiencing reading that macro-result. However, the devices don't "know" each others results prior to comparison, while they do yield a result prior to comparison. It is certain that the results are correlated (the photons passing the filters at a specified relative angle).

I imagine that the measurement results (on both sides) are unambiguous, because the devices are macro-objects and the readouts are practically conclusive. However, how can there be constructed a correlation before the necessary information is available (from the other side)?

After all, theoretically, the correlation can only be (theoretically) established when the results are compared (and the results are no longer spacelike separated).

Is there an interpretation for this?

 

[Paul Colby]

After all, theoretically, the correlation can only be (theoretically) established when the results are compared (and the results are no longer spacelike separated).

I think the word "theoretically" is poorly used here. The correlation exists as dictated by theory, however, experimental verification only occurs when notes from the two detectors are compared. In this aspect the problem is not different than a classical case of correlation.

 

[entropy1]

The correlation exists as dictated by theory, however, experimental verification only occurs when notes from the two detectors are compared. In this aspect the problem is not different than a classical case of correlation.

Correct. However, the outcomes of the respective measurements must become correlated at some point. That point can't be when the notes are compared, for the measurements are conclusive before that, right?

 

[Paul Colby]

Yes, for a two particle entangled system, measurement of one state by passing the polarizer, determines the state of the other particle. The order in time of the measurements is completely inconsequential.

 

[entropy1]

Yes, for a two particle entangled system, measurement of one state by passing the polarizer, determines the state of the other particle. The order in time of the measurements is completely inconsequential.

So, are you saying that when the notes are compared at Alice's, by sending Bob's notes to her, the correlation becomes established there, just as well as the other way round?

If so, then I still don't see how Alice's apparatus could have 'guessed' the results Bob got, nor the other way round.

 

[Paul Colby]

If so, then I still don't see how Alice's apparatus could have 'guessed' the results Bob got, nor the other way round.

That's because, like so many here, your mind insists that the detection device and particle need to "somehow know" the result. I prefer to take the quantum rules at face value. Assigning state vectors, the born rule, the measurement axiom are all taken together as a fundamental aspect of nature. When one particle passes a measurement the STATE of the other is determined. If this state happens to be an eigenstate of the distant measurement device, then one gets perfect correlation. The only spooky bit is that QM is the way of the world.

 

[entropy1]

That's because, like so many here, your mind insists that the detection device and particle need to "somehow know" the result.

At least, both devices know their own result (while separated)!

I prefer to take the quantum rules at face value. Assigning state vectors, the born rule, the measurement axiom are all taken together as a fundamental aspect of nature.

I respect that.

When one particle passes a measurement the STATE of the other is determined. If this state happens to be an eigenstate of the distant measurement device, then one gets perfect correlation. The only spooky bit is that QM is the way of the world.

Is there, however, since the order of measurement is irrelevant, an element of randomness in play? The state of Bob may be determined by Alice, but also vice-versa! So, there isn't really (unambiguous) determination. However, the readouts of the devices are still conclusive...

What I (think I) mean is, that it is the link between the quantum- and macroworld that remains obscure, right?

 

[Nugatory]

What I (think I) mean is, that it is the link between the quantum- and macroworld that remains obscure, right?

We have a quantum system prepared in the spin singlet state, and it's easy to get from there to the prediction that there is a 50% chance that A is up and B is down, a 50% chance that B is up and A is down, and a 0% chance that both are up or both are down. The (anti)correlation between the spins is a bit of a red herring because there's a 100% probability that it will appear, no matter which of the two 50% chances come through for us.

But what is it about a measurement of the spin at either A or B that turns a system in which either of two outcomes are equally likely into a system in which one of those two 50% possibilities has been realized? Without an answer to that question, there's no answer to the question you posed to start this thread.

 

[entropy1]

But what is it about a measurement of the spin at either A or B that turns a system in which either of two outcomes are equally likely into a system in which one of those two 50% possibilities has been realized? Without an answer to that question, there's no answer to the question you posed to start this thread.

So yes, what would that be? Maybe decoherence? Which mechanism makes this odd behaviour insightful?