# Testing whether entanglement is a matter of information or non-local?

1. Jul 2, 2014

### QuestionMarks

Depending on who one asks and their interpretation of QM, entanglement seems to be either:

a) No problem at all. It's just a matter of information. If you knew one entangled electron was spin up, then the other must have been spin down by inference of the prepared state of the system.
b) Potentially "spooky action at a distance." One electron might well be able to tell the other to "flip" its spin instantaneously. Something worth figuring out, but not something we like because that would be non-local.

And I've seen some pretty entrenched belief systems based on either way. But couldn't we test this? I would assume we have already?

If not, why not some similar scenario as follows...
We have a set of points, notated by the following letters, where the distance between each adjacent point is the same as those of any other two adjacent points:

A---B---C---D---E---F---G

At (D), two entangled electrons are released, electron 1 moving to (A) and electron 2 moving to (G). Both paths are the same in characteristics. At point (C) the spin of electron 1 is measured as "up," thereby we "know" that of 2 is "down" at (E). At point (B) (or even we could say immediately at the measurement on point C), electron 1 is hit and its spin changed from "up" to "down". As both particles reach their endpoints, electron 2 is measured at (G).

If electron 2 is shown as "up," then would we not know that electron 1 somehow told electron 2 to flip sometime after (C-E)? Would this not support then that entanglement is not merely inference of information from the system, but rather an actual effect of one particle upon the other (even if it is non-local)? Surely something like this experiment has already been done?

thanks all.

2. Jul 2, 2014

### DrChinese

Once electron 1 is observed to be spin up at (C), it is no longer entangled with electron 2. So what happens next to 1 has no effect on 2. You can be sure that once electron 2 is down, it will stay down (until some interaction or observation places it into a different state)! That is fairly fundamental.

So the expected null result in this case does not place us any closer to answering your question.

3. Jul 2, 2014

### DrChinese

By the way, your question does not follow the norms of how entanglement is usually explained. The "information only" view is generally discredited if you are rejecting non-locality. You usually either reject "realism" or reject "locality" (or both). This is because of Bell's Theorem.

4. Jul 2, 2014

### QuestionMarks

Thanks DrChinese. Hah, good to know when one's brain feels it is missing something obvious, it nearly always is. So my proposition is bunk, but is there any experimental evidence that leans us closer to either understanding? Any key papers?

And I get Bell's theorem's sentiments, though I guess I should acknowledge that I'm not necessarily rejecting non-locality. I grant that's traditional, but I'm willing to consider alternatives. Perhaps consider me biased towards realism to keep the dialogue interesting.

5. Jul 2, 2014

### DrChinese

The state of the art has not proven or disproven direct non-local action. There is plenty of proof for quantum non-locality but there are some interpretations in which the action is not direct (example: time symmetric versions which respect c but do not respect causality as we know it).

It is possible to entangle particles that have never interacted, for example. Such particles need not have ever existed in the other's past time cone. Also: tests for non-locality have indicated that if non-local effects have a finite limit on the speed of action, it must be at least 10,000 c.

6. Jul 2, 2014

### QuestionMarks

All interesing ideas. Can any particular experiment names or search terms be put to these so I can do more reading?

Particularly though, how can you have a nonlocal effect that is indirect or still respects c as you say? And is "nonlocal" still an apt word for this? My thought would be that any event occuring to one particle and causally-correlating with an event on/to another could be described as information transfer. Ive certainly read a handful of experiments where QM causality gets a little wonky, but I never felt any such adbicated those instances of our responsibility in considering nonlocal mechanisms.

Last edited: Jul 2, 2014
7. Jul 2, 2014

### DrChinese

1. Here are a few:

http://arxiv.org/abs/quant-ph/0201134
Decision to entangle 2 particles is made AFTER the particles have already been detected. Effect precedes the cause.

http://arxiv.org/abs/1209.4191
Entanglement Between Photons that have Never Coexisted; The observed quantum correlations manifest the non-locality of quantum mechanics in spacetime.

http://arxiv.org/abs/0808.3316
Testing spooky action at a distance; ...if such a privileged reference frame exists and is such that the Earth's speed in this frame is less than 10^-3 that of the speed of light, then the speed of this spooky influence would have to exceed that of light by at least 4 orders of magnitude.

2. An indirect non-local effect is one in which the causal arrow move back and forth, but otherwise respects c. The space-time diagram of the experimental context does not have anything propagating at a speed other than c or -c. You don't have to agree anything is proven to be time-reversed, but that is also an acceptable interpretation.

8. Jul 3, 2014

### morrobay

In the context of spin 1/2 particles in a basic Bell inequality : n[x+,z+] + n[y-,z-] ≥ n[x+,y+]
Realism assumes that variables "Determine precisely the results of individual measurements". That is, if electron 1 is spin up on z axis then if electron 2 were measured it would be spin down on z axis , from conservation laws.
While once electron 1 is observed to be up it is no longer entangled with electron 2 and electron 2 will stay down.
Then when the above inequality is violated does this suggest there can be non-local influences between electrons that are not entangled ?

Last edited: Jul 3, 2014
9. Jul 3, 2014

### DrChinese

The inequality is not going to be violated if the x, y and z are all mutually orthogonal axes for measurement.

I would say: There is no suggestion that there is a non-local influence between electrons that are not entangled.

10. Jul 3, 2014

### QuestionMarks

This is all great reading resources DrChinese. Super thanks! I'll dig into these.

On the backwards c, I'm glad you mentioned that. That was actually an interpretation (in my own mind)of some results I read a while back, but when I looked around to see if this was a valid idea, it was very poo-pooed upon. Granted I understand one's interpretation of QM may create strong arguments to the contrary. Makes me wish, however, most QM conversations could be held in reference to the knowledge that a general interpretation is not yet consensus.

Anyways Thanks!

11. Jul 3, 2014

### trendal

Just kind of an aside here...but how can you say with absolute certainty that two particles have never interacted? If we go all the way back to the beginning of this universe, all particles would have interaction with each other because they all would have been in the same space.

12. Jul 3, 2014

### Staff: Mentor

Actually its not dependant at all on interpretation. Its simply the vector space nature of pure states.

If you have two systems that can be in states |a> and |b> then system 1 can be in state |a> and system 2 in state |b> - their combined state is the pure state |a>|b>. Similarly system 1 can be in state |b> and system 2 in state |a>. Again their combined state is |b>|a>. But pure states form a vector space (that's the basis of the superposition principle) so a linear combination is also an allowable state eg 1/root 2 |a>|b> + 1/root 2 |b>|a>. By definition such states are called entangled.

In fact it is now known that a few reasonable assumptions and entanglement more or less implies QM:
http://arxiv.org/abs/0911.0695

The situation is this. The three axioms mentioned in the paper lead to either probability theory or quantum mechanics. Quantum mechanics is singled out if you want continuous transformations between pure states (physical continuity requires this ie if you can apply a transformation for 1 second you can apply it for 1/2 second) OR you can allow entanglement. At a deep fundamental level entanglement seems to be built in to the very foundations of modelling physical systems.

Thanks
Bill

13. Jul 3, 2014

### Staff: Mentor

You cant.

But what you can say is when you observe one part of an entangled system it's now entangled with what you observed it with rather then what it was originally entangled with.

Thanks
Bill

14. Jul 3, 2014

### morrobay

It my understanding that QM predicts that two particles will be spin up or spin down: 1/2(sin(Θ/2))2

x y z...........x y z
- - -...........+ + +
- - +...........+ + -
- + +...........+ - -
+ + +..........- - -
+ + -...........- + +
+ - -...........- + +
- + -...........+ - +
+ - +...........- + -

So if expectation values for the inequality are from table above and as shown spins in the
inequality are opposites then can it be violated at 900
n[x+,z-] + n[y-z+] ≥ n[x-,y+]

Last edited: Jul 4, 2014
15. Jul 3, 2014

### QuestionMarks

I'll check out that paper eventually, but I feel I'm likely to still disagree. Wherein we still currently have no fundamental consensus regarding the basic reality QM implies, I would find it overwhelmingly difficult to believe that entanglement enacts the same roles in all interpretations. Sure it can be given the same mathematical formulations, if you mean just that, but the reality (if there is one) is different. For instance, the MWI can have some sentiment of different entangled systems as different universes/histories. Meanwhile, Bohm's interpretation could quite literally count as spooky action if your requirement for that is simply a nonlocal hidden variable. QM may not depend on interpretation for the Math to be functional, but the ontology does.

Last edited: Jul 3, 2014
16. Jul 3, 2014

### Imafungi

Has any entanglement experiment proven that this following analogy is not akin to what is going on in the experiments (that is to say, has any entangle experiment proven non locality, in which this following analogy cannot be used to describe the most likely potential of what occurred)?

There is a red ball and a blue ball. You and I are in another room blindfolded. A random person walks in the room with the red ball and blue ball and puts each in a bag. You and I walk in the room and each grab a bag. I go to the moon. You stay there. We know without looking that the results will be 1 red ball and 1 blue ball. I look in my bag on the moon, and its the blue ball. I instantly know that your ball is the red ball.

If this analogy can be used to describe every entanglement experiment, the analogy would be related to the 'spooky entanglement interpretation'; I got my bag and you get your bag. I go to the moon. Before I or your look in our bags, I have a red/blue ball in my bag, and you have a red/blue ball in your bag. When the bags were in the same room before they were separated, they were entangled. When I looked in my bag the red/blue ball turned into a blue ball, which then faster than the speed of light, notified your red/blue ball to turn into a red ball.

I honestly dont know much about this subject, but I like to remain skeptical, in the sense of believing in locality, cause and effect, logic, and the ability to explain physical interactions.

17. Jul 3, 2014

### Staff: Mentor

Yes. Google for "Bell's Theorem" and also check out http://www.drchinese.com/Bells_Theorem.htm - DrChinese is a regular contributor here.

18. Jul 4, 2014

### Imafungi

Im sorry but I dont think that in anyway proved my analogy wrong, which is a 'local realism' view of the spooky action entanglement interpretation.

In entanglement experiments. When the first particle is measured, and the second particle is not measured. When the first particle is measured, at that instant does the second particle alert a system of its state? Or do the experimenters have to measure the second particle by themselves after? Or is it in a superposition in some apparatus either just vibrating in place or traveling around in a circle, and then the first particle is measured, and that second particle just instantly finds the nearest detector to exclaim to the experimenters that its entangled pair had been measured?

19. Jul 4, 2014

### Water nosfim

You dont just open , you paint the ball , and statistic you get anather color far and fast

20. Jul 4, 2014

### Staff: Mentor

Well since there is no consensus on what reality even means that's hardly surprising.

In physics generally the math is assumed to describe reality without getting bogged down in exactly what it is in the first place. If you want to go beyond that you are really getting into philosophy rather than science.

Why do you need an ontology beyond a simple interpretation of probability, which you need to make sense of probability to begin with? If you adopt the frequentest view of probability you get the so called statistical interpretation championed by Ballentine. If you adopt a Baysian view then you get something like Copenhagen. Strictly speaking its an ontology, but you would have a lot of trouble convincing those with a background in applied math like me its really more than the formalism implies.

Thanks
Bill

Last edited: Jul 4, 2014
21. Jul 4, 2014

### QuestionMarks

That hints of the "shut up and calculate" culture often found in modern physics, which is a bit of a presumed philosophy itself unfortunately. And though we do not need the ontology per say, why discourage the pursuit? It only becomes philosophy (which is no evil, just a different forum) when we stop looking for empirical evidence. Presuming no experiment can be performed which would give us a better interpretational view of reality would be an unnecessary assumption (and I allow myself optimism at least). So until then, and especially while we are particularly focusing on what empirical support is out there, I see no concern in my thought venture.

Last edited: Jul 4, 2014
22. Jul 4, 2014

### DrChinese

The inequality is not violated at 90 degrees, because:

QM_Prediction(n[x+,z-])=25%
QM_Prediction(n[y-,z+])=25%
QM_Prediction(n[x-,y+])=25%

So that:

25% + 25% >= 25%

By the way, according to QM, the x, y and z spin components of an electron do not commute. A measurement of x requires that both y and z are completely indeterminate (and vice versa). So any x outcome (which will have a 50-50 likelihood) will be accompanied by a random outcome of either y or z. 50% x 50% = 25%.

23. Jul 4, 2014

### DrChinese

First, no one actually understands the underlying mechanism by which entanglement works. That is why there are multiple interpretations. From 1935 until Bell's Theorem, your local realistic explanation was considered one possible interpretation (the EPR interpretation for lack of a better name).

However, Bell showed that interpretation to be flawed. The EPR interpretation has since been proven incorrect many times over in experiment. Until you read and understand Bell, you won't make any progress towards understanding WHY the EPR interpretation is wrong. Better yet, read the EPR paper first and then Bell.

BTW: In QM, the ordering of the measurements on Alice and Bob (particles 1 and 2) do not affect the outcome in any discernible fashion.

24. Jul 4, 2014

### DrChinese

Well, it is incorrect as I explained. That should give you pause. You cannot maintain all of the elements you have proposed (causality and locality).

It is very common to speculate. That is the lifeblood of new ideas, most of which are completely wrong. It is not common to give credit to speculation for which there is no practical implication because there is no new useful prediction to test.

So the reason people exclude local realistic interpretations/theories is that they run afoul of Bell.

25. Jul 4, 2014

### Staff: Mentor

Or maybe you misinterpret taking the math at face value.

Its what led to perhaps the greatest revelation of modern physics:
http://arxiv.org/pdf/hep-th/9704139.pdf

'The most important lesson that we have learned in this century it is that the secret of nature is symmetry. Starting with relativity, proceeding through the development of quantum mechanics and culminating, in the standard model symmetry principles have assumed a central position in the fundamental theories of nature. Local gauge symmetries provide the basis of the standard model and of Einstein’s theory of gravitation.'

It was the math of QM that revealed this - ontological speculations have basically led no-where.

Thanks
Bill

Last edited: Jul 4, 2014