Implications of the Delayed Choice Quantum Eraser

[DrChinese]

To be quite clear about this, Bartlemann's socks are perfectly correlated with classical correlation. They were made as a pair and sent on their way. Alice and Bob's photons were made as a pair too, and were sent on their way. The difference is, the socks were sent as a mixed state |red>|blue> OR |blue>|red> but the photons are created (because of spatial symmetry) as a superposition of |H>|V> AND |V>|H>. |H> and |V> span the same state space as |a> and |a+90>, where a is an arbitrary angle, so the EPR correlation is maintained for just as long as the superposition does not collapse or otherwise decohere..

Almost everything here is incorrect as it pertains to the Delay Choice Entanglement Swapping (DCES, which is completely analogous to the DCQE in theoretical points). Bertlmann's socks IS classical (and has no bearing here anyway). But the entangled photon pair in DCES is not, and not only were they NOT made as a pair - they need never have existed at the same time. The EPR correlation does not even exist at the time the photons are measured and the related "collapse" occurs. That is done later.

I think my point in all this is: you are selling a naive version of these experiments. But they are complicated and touch on the outer edges of quantum weirdness. You deny that any collapse has occurred when measurement has taken place, yet in a different context you could say that is absolutely has because the outcome of an observation on entangled twins can be predicted with certainty. That does not imply an ongoing superposition at all!

What I am saying mimics QM itself: know the full context, you know the probabilities. The context need not be local or in a particular causal sequence. Why it works out that way is the mystery.

 

[DrChinese]

I think that seems to be a problem in some cases. I think people start with the premise "This can't be true." So there must be an explanation that fits what they believe to be true. So they "demystify" these things but I don't see how that's possible.
...
These are the obvious questions on it's face that can't be easily explained away because in the end it really depends on which interpretation of QM you choose to accept.
...
I think there's less need to try and "demystify" these things with MWI.

Yes, the determination of which thing "can't be true" drives one's interpretation.

I don't really have an issue with the many worlds of MWI if it answers some good questions. I just don't see how you get very far with that with the Delayed Choice experiments, because the choice of context is delayed past the point where I think there should have been a splitting. And some of the split off worlds later become impossible worlds (because of perfect correlations, which won't support imperfect permutations). But that's just me. :)

 

[Derek Potter]

Almost everything here is incorrect as it pertains to the Delay Choice Entanglement Swapping (DCES, which is completely analogous to the DCQE in theoretical points).

Did you not notice the term "EPR" towards the end of the paragraph?

You deny that any collapse has occurred when measurement has taken place, yet in a different context you could say that is absolutely has because the outcome of an observation on entangled twins can be predicted with certainty.
That does not imply an ongoing superposition at all!

Are you serious? There is perfect correlation in each component state. Nothing else is needed except normal measurement theory.

 

[Kansas_Cowboy]

No, probability is given by the square of the magnitude of something called probability amplitude. Probability amplitude is a complex number. Probability amplitudes are added together like vectors (Pythagorus!) where two contributions arrive together. There is no change in the state that gives rise to the probability amplitude The result can be bigger than either or less than both. Hence interference.

I thought the same thing at one time too, but I was wrong.

But isn't adding the probability amplitudes together a representation of this interaction built into the equation? I'm not implying anything like bits of one wave colliding with the bits of other waves as you mentioned previously. I'm making no assumption as to what this interaction actually entails, merely asserting that it must exist.

Also, I'm a tad confused. If I understand you correctly, you're saying that D1 and D2 results don't show interference? That only the signal photons associated with D1 and D2 produce interference patterns on D0? What then are the results of D1 and D2?

 

[Derek Potter]

Yes, the determination of which thing "can't be true" drives one's interpretation.

I don't really have an issue with the many worlds of MWI if it answers some good questions. I just don't see how you get very far with that with the Delayed Choice experiments, because the choice of context is delayed past the point where I think there should have been a splitting. And some of the split off worlds later become impossible worlds (because of perfect correlations, which won't support imperfect permutations). But that's just me. :)

In MWI the choice of context does not *cause* anything. With adequate fine-graining all the worlds exist "from the beginning". We have done to death the issue of "splitting". You can define MWI as "the silly popularist picture of parallel universes that split whenever someone makes an observation" or as "entangled states of observer and system, interpreted as relative states and augmented by decoherence-based measurement theory". I prefer the definition that will be lost on the hoi polloi but not for that reason. (I am not a snob, some of my best friends are common).

 

[Derek Potter]

But isn't adding the probability amplitudes together a representation of this interaction built into the equation? I'm not implying anything like bits of one wave colliding with the bits of other waves as you mentioned previously. I'm making no assumption as to what this interaction actually entails, merely asserting that it must exist.

The two states don't affect each other. The detector interacts with the sum. That's no different from a TV and a radio blaring in the same room. They don't affect each other but your ear has to deal with the total racket.

Also, I'm a tad confused. If I understand you correctly, you're saying that D1 and D2 results don't show interference? That only the signal photons associated with D1 and D2 produce interference patterns on D0? What then are the results of D1 and D2?

Yes, at D1 and D2, you just get a broad lobe, no interference pattern.

 

[DrChinese]

Are you serious? There is perfect correlation in each component state. Nothing else is needed except normal measurement theory.

Let's see. I measure a photon of entangled pair A at 87.3 degrees. My friend, far away, measures a photon of entangled pair B at 87.3 degrees.

How are those still in a superposition? My photon and my friend's are already measured and the values are known! And how is there any correlation now between these at this odd choice of angle - or any angle we decide upon - if the decision to retroactively entangle these 2 distant particles via swapping is made at a later time? There was no "perfect correlation" until later.

 

[quantumfunction]

Yes, the determination of which thing "can't be true" drives one's interpretation.

I don't really have an issue with the many worlds of MWI if it answers some good questions. I just don't see how you get very far with that with the Delayed Choice experiments, because the choice of context is delayed past the point where I think there should have been a splitting. And some of the split off worlds later become impossible worlds (because of perfect correlations, which won't support imperfect permutations). But that's just me. :)

I think you make a some good points and I think if you support MWI, you have to take into account that what information can be known about the system in the environment seems to mean something.

You ask an obvious question because if you look at the entanglement swapping experiment, you have to ask why wasn't there a split when Bob and Alice's particles were measured? MWI can't answer that question at least in the context of a physical interaction. There's no reason why Alice and Bob's measurements should be dependent on Victor's choice in the future.

It's like the particles are delaying their choice on whether they will be entangled/not entangled until Victor decides what he's going to do even after their particles have been measured.

It looks like whether the information about the state of the particle can be known by the environment is even more important than measurement. So the state of the particle seems to be in limbo until Victor makes a choice even after a measurement has already occurred.

I personally like the idea of the wave function as a non physical reality. So you can define physical apart from real.

So the superposition of states only become physical when information about the system can be known in the environment of the observable state of an isolated system. This begs the question, what's physical? Does an observable state only appear real to observers in a local environment?

Let's just assume the wave function is a non physical reality. What if the "universe" is simply a physical isolated system and it's wave function evolves all possible states of this isolated system. So what we call physical reality is just a projection of states of this isolated system. So we could be holograms, phantasms or whatever you want to call it because this physical isolated system could never physically occupy all of space that's expanding faster than light.

Another question that would need to be asked is if this physical isolated system has a finite configuration or arrangement it can be in. Tegmark and others seems to think it's finite so we can be in some infinite loop of existence as the same and similar states keep occurring ad infinitum.

Just a thought

 

[Gaz]

Well I don't know much but I know this is wrong

As you pointed out, this is incorrect. As a matter of fact, you can determine which slit a photon goes through and it still reach the screen. One way to do that is to place a V polarizer at one slit and an H polarizer at the other. The interference pattern will disappear. On the other hand, if both polarizers are V, there will be an interference pattern.

Using V and H polarizer's can be explained by both QM and classical physics as there would be no interference from v polarized and h polarized em wave so Is no evidence for either.

The which path information is what is supposed to collapse the probability wave. If a photon registers at D3 or D4 then you know which path it came from and the wave collapses. I just can't get my head around QM.

 

[DrChinese]

Using V and H polarizer's can be explained by both QM and classical physics as there would be no interference from v polarized and h polarized em wave so Is no evidence for either.

True enough. (Although I don't know how you think my quoted text had something misstated.)

My point was that you can learn which path information from a quantum particle without otherwise altering it. The relative orientation of the polarizers should not matter to the outcome if a particle only goes through one slit or the other. Obviously, the relative orientation DOES matter.

 

[Gaz]

Sorry Like I said I don't know much about it =)

 

[Derek Potter]

Let's see. I measure a photon of entangled pair A at 87.3 degrees. My friend, far away, measures a photon of entangled pair B at 87.3 degrees.
How are those still in a superposition? My photon and my friend's are already measured and the values are known!

Sorry but in MWI you don't know the value because there is no single value to know. That's why it's called "Many".

The superposition is very simple. The initial two photon state was .707|H>_you|V>_friend +.707|V>_you|H>_friend (anticorrelation). H and V are the possible outcomes in the 87.3 degree basis. After interaction the joint detector state is .707|1>_D1|0>_D2 + .707|0>_D1|1>_D2. In a two-photon experiment that is the end of the matter - ordinary measurement theory tells us that the superposition looks like a mixture.

 

[Derek Potter]

Yes, it does. Only when the entangled photon is detected by D1 or D2 then you get interference patterns at D0 from the idler photon. When D3 or D4 gets a hit AFTER the photon is detected at D0, then you do not get interference at D0. Crazy, huh?

The detection at D1 or D2 is also AFTER the photon is detected at D0

I think that is why Derek Potter likes to refer to the photons as "seeing the path they are going to take" because it is a way to rationalize what occurs. There is no physical evidence that the photons know where they are going to hit, only the logical implication that is contradicted by normal reality.

Huh? I didn't say anything like that, I said "If you look at the ray paths, D1 and D2 can only see one slit, D3 and D4 see both." It's simple optics, D3 sees a reflection of slit A in BSA, D1 sees slit A through BSA but reflected by MA and BSC. It also sees slit B through BSB and reflected by MB but through BSC. Nothing whatsoever about photons knowing where they're going to hit.

 

[Kansas_Cowboy]

The two states don't affect each other. The detector interacts with the sum. That's no different from a TV and a radio blaring in the same room. They don't affect each other but your ear has to deal with the total racket.

Yes, at D1 and D2, you just get a broad lobe, no interference pattern.

I still feel like that is a misrepresentation. At the end of the day, a single photon hits the detector. The detector interacts with a single photon. It is the photon's position upon reaching the detector that is determined by the sum of probabilities. Thus, the probabilities are acting upon the photon's final position, not upon the detector. Can we agree when I phrase it like this?

And on the second point, after some thought, it makes sense. Not only would the timing likely differ slightly for the idler photons to reach D1 or D2 depending on the slit, but from the diagram at least, it appears that they are angled such that they ultimately reach the detectors from the same point of the beam splitters regardless of slit.

 

[berkeman]

Thread closed temporarily for Moderation...

Edit: the thread will remain closed and a problematic discussion has been removed.