Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

A very simple question on quantum entanglement

  1. Jul 18, 2011 #1
    Hi there, I've recently read some material on QM and entanglement in particular, and even thou I managed to understand the material I felt like it didn't contain the answer to one fairly simple question...

    When an entangled pair is produced, conservation of energy laws cause the members of the pair to be anti-correlated, member A is always complimentary or opposite of member B, be that spin, position of whatever. According to the entanglement theory, measuring the property of A immediately sets the property of B, which implies a change in B occurs upon measurement of A, but I don't really see it this way - since A and B have been produced anti-correlated, measurement of A doesn't change B in any way, it just indirectly defines B as the opposite of A, nothing really changes in B, it has been the same from the moment of production, the only thing that changes is that initial state of B is now known through the measurement of A.

    In the same logic, I can take a white and black lab mice, put them in identical boxes, shuffle the boxes to the point it is unknown which box holds the while and which box holds the black one, send one to China, the second to the US, opening either of the boxes and inspecting its content will immediately define the content of the other box, no matter the distance in between, but does that mean the two mice are entangled? I don't think so...

    My question is: What exactly suggests any FTL interaction between entangled particles or change if any of them takes place upon measurement?
  2. jcsd
  3. Jul 18, 2011 #2


    User Avatar
    Science Advisor

    First of all, entangled states do not have to be anti-correlated .. they can also be correlated. The correlation (or anti-correlation) between the entangled properties has nothing to do with conservation of energy.

    Second, you made (at least) one other incorrect statement in your description ... namely, the statement
    This is not true. If A & B are in an entangled state, then you cannot say anything about the states of the individual particles, you can only describe the overall entangled state. Only after the entangled state has been measured do the individual particles assume well-defined states. This is the prediction of QM, and it has been verified many times in experimental contexts. Your phrasing of the problem is equivalent to a "local hidden variables" theory, which has been ruled out via a series of really neat experiments. You can learn more about this from the links on Dr. Chinese's website (www.DrChinese.com).

    Anyway, if you read about those experiments, and about Bell's Theorem, you should be able to understand why the case of entanglement is NOT the same as the mouse example you posted.
  4. Jul 18, 2011 #3
    Sorry, it is poor English skills, I used anti-correlated as a synonym of complimentary, but I see now it wasn't appropriate...

    Why can't I say anything about the states of the individual particles? All in all, there is a 50/50 chance of any of them being in a given state with the other being the oposite, just because WE DON'T know doesn't mean the particles are undetermined from the moment of their creation, or does it?

    This is what I fail to understand, why does entanglement theory assumes particle B is in an undetermined state and only gets determined upon measurement of A? Again, just because we don't know the state of B doesn't mean B itself doesn't know its state and decides AND assumes that state the moment A is measured in some FTL interaction. B is NEVER undetermined, it is ALWAYS complimentary to A from the moment it is "born", it is only US as observers that aren't unaware of the states of individual particles until we measure one.

    This goes back to Schrödinger's cat - just because an observer is obstructed from observing doesn't really put the cat in a state of superposition, it is never both dead and alive, it is EITHER dead or alive, no matter of the uncertainty of an observer. Also the cat itself is an observer to begin with, logically Schrödinger's cat example is never in a state of superposition since the cat collapses it, since it knows whether it is alive...

    I guess what I am trying to say the view of superposition is not an objective property, it is subjective to an observer, but in the objective sense there is no superposition, even if we refuse to individually concern ourselves with particles A and B, the particles have their fixed states all along in what we conceive as entanglement and view as a composite state of superposition...

    I don't think A and B are stateless just because we don't know their states, that is all... A and B always have their complimentary states. I fear this is more of a philosophical rather than a scientific matter, I understand the actual experiments being performed, their statistical results and such... The results make perfect sense and all, but I just don't see any evidence of any process taking place in the moment of observation of A that induces a response in B over distance

    Well, considering the question I asked was referring to that same statement, it shouldn't really be conceived as a statement, but a question without a question mark, I mean the answer to my question is what would make this statement incorrect, and that was what I was hoping to get - what makes TRUE the statement that any objective change in B occurs in the moment of disentanglement, in order to differentiate B from the state it assumed in the moment of its production?

    You make it sound like individually, A and B do not physically exist prior to observation, like it is not two entangled particles BUT a sole entanglement, which morphs into two particles upon measurement, which is purely conceptual and subjective view, especially since experiments involve separating A from B we do have two individual particles with their appropriate states all along, those are just unknown prior to measurement.

    All in all, entanglement is presented to be some link, one that cannot transfer information and "disappears" upon any attempt to verify it... which is kind of counter-intuitive, I could just as easily have an imaginary friend, one that can't talk and disappears every time someone tries to look at him...

    Let's say we measure A and it is spin up, which logically makes B spin down, how do we know B "became" spin down in the moment of measurement of A and wasn't spin down prior to it, especially since we have only one measurement allowed?

    I don't claim anything, nor try to redefine anything, I just hope things get explained to me, I have read quite a lot prior to asking those questions and I still see an inconsistency, a gaps filled by what seems like an assumption (represented by all BOLD text), and I doubt I am that stupid I failed to get it, but who knows...
    Last edited: Jul 18, 2011
  5. Jul 18, 2011 #4


    User Avatar
    Science Advisor

    Again, I suggest you start reading about the EPR paradox and Bell's theorem, to help guide your questions. The link I gave to DrChinese's website is a great place to start.
  6. Jul 18, 2011 #5
    You seem to have quoted my entire post without even reading it, since if you did read it, you'd notice I mentioned that I have already read and understood the material you advice me to read prior to posting, including the articles of DrChinese, perhaps you could be at least helpful enough to provide more specific pointers that contain the answer I am asking for?

    I particularly fail to see a major problem with a "hidden variable theory" which seems to be the issue I am addressing... we can't reach full vacuum, we can't reach absolute zero, there are always tons of factors which are way too much and too small to be accounted for, giving a certain degree of an illusion of randomness, meaning on such a small scale it would be impossible to come up with a certain result even if we know all the hidden variables, such an experiment would be unable to model within 100% accuracy, but this doesn't mean classical physics fails at small scales, it just means unaccountably small factors get too prominent
    Last edited: Jul 18, 2011
  7. Jul 18, 2011 #6


    User Avatar
    Science Advisor

    Well, you may have read some of those links, but your post makes it clear that you have not yet understood them. In fact, I couldn't tell that you had read them at all from your previous post.

    1) Entanglement works as I have already described .. I can't explain it any better than that. Your ideas about the "reality" of the particles in the entangled state fall under the realm of interpretations of QM, which can be instructive, but do not give conclusive answers about questions like this one. My view is, there are no known measurements (there may not even be any theoretically possible ones) we can do to ascertain the "reality" of the particles in an entangled state without destroying the entanglement, so why bother worrying about it?

    2) There is no way to explain the experimental results of Aspect, or the more recent ones of Zeilinger, with a local realistic model where the properties of the particles are determined at creation. I suggest that you read those papers more carefully to understand why this is the case. Very briefly, the significant experimental result is that the coincidence statistics only depend on the *relative* angle between the polarizers used to measure A and B. If the polarization states of the particles were well-defined before the measurement, then this would define a "preferred basis" for the polarization measurements, and the coincidence statistics would show a more complicated dependence on the settings of the detectors at A and B.

    I suggest that you read David Mermin's excellent article, "Is the moon there when nobody is looking?", which was published in Physics Today in 1985. You can find it http://citeseerx.ist.psu.edu/viewdoc/download?doi=". That should help you better understand why QM is incompatible with local hidden variable theories.

    I also suggest reading the original EPR paper (you can find it on Dr C's website), or re-reading it if you have already looked at it. It gives a very clear statement of the issues related to the questions you are asking.
    Last edited by a moderator: Apr 26, 2017
  8. Jul 18, 2011 #7

    Try thinking of entanglement as a 'state' that two particles share that is independant of space and independant of time. It 'exists' in each particle for a particular period of time in particular places but has only two pertinent 'points' in time/space for each particle- the entanglement and the collapse of the entanglement. In a very 'real' sense one particle is 'still' entangled until it has been measured in some way even long after the 'first' particle's state collapses to a measured condition. As long as the same entangled property of both particles is measured, the entanglement will be confirmed. It is our temporal nature to ascribe the collapse to the 'first' measurement over the 'second' collapse -but they are on an equal footing. There is no causal relationship between the two collapses. They are the same collapse occuring at two different time/space locations, that is all.

  9. Jul 18, 2011 #8
    The thing that is troubling me is the scientific consensus is pro-entanglement without actually being able to explain its operational mechanics, it seems like a HUGE assumption to swallow, what is worse - kind of dogmatic - "no one knows why and how, but it happens". The fact it is claimed to be an FTL interaction makes things all that much worse, making it far too "far fetched" to be accepted in a dogmatic manner, without an actually explanation how does this process exactly occur.

    It may be just me, but I have to know how stuff works and cannot settle for the "well, it just works" view - not only the mechanics of a process must be explained but should also be plausible and simple enough to work flawlessly - we all know the more complex the mechanic the more often it breaks - and from my experience the universe is quite stable.

    Mathal - you mean partial collapse? Like for example, two entangled particles can have their Z entanglement broken, but only it and otherwise remain entangled so that those same particles can be measured along Y and having their Y entanglement broken and so on, a sort of "thread by thread" disentanglement?
  10. Jul 18, 2011 #9


    User Avatar
    Science Advisor
    Gold Member

    Welcome to PhysicsForums!

    SpectraCat has given you some good information, perhaps I can add to that a bit. Your mice example happens to fit the same situation as entangled particles. So that is why it seems to make sense to you. Unfortunately, there are a limited number of cases it does work. Actually, there are cases in which it turns out there is no possibility your example will match the predictions of QM. That is what Bell discovered.

    The Mermin example is good, and I have a variation of it on my website. If you don't follow the thinking, I will try to walk you through it. Needless to say, your concept was in fact in vogue from 1935 to 1965 before Bell came along.

    But consider this: if you use measurements at 0, 120 and 240 degrees instead of 0 and 90 degrees, what would you expect with a hidden variable model? Some kind of mixture of properties, correct?
  11. Jul 18, 2011 #10
    Hi Mathal,

    you got me curious

    i can understand the different space part, but not the time part.

    1. one version/understanding/hypothesis is:

    when one of the entangled pair is measured the wave-function collapses, i.e. the other twin also assumes a definite state at the same instant in time

    2. the other could be what you are saying...same collapse happening at different time/space...

    3. there are more hypothesis/explanations

    however right now, I am trying to understand what you mean when you say that both the particles don't collapse at the same time because to me 1 above explains the results of the delayed choice quantum eraser DCQE "completely", then why assume 2?

    also why assume/state there is no casual relationship?

    the way I understand is, and I am open to new logic, that this casual relationship "transcends/is outside" time-space and its instantaneous?
  12. Jul 18, 2011 #11


    User Avatar
    Science Advisor

    That is a mis-representation of the scientific consensus on entanglement .. it is not dogmatic at all in general .. in fact, these are some of the most oft-challenged, and experimentally tested foundational ideas in QM. The reason people accept that the QM description of entangled states is correct is because it is has been shown to be consistent with every experimental result obtained to date.

    Furthermore, you should be precise when saying that there is a claim of "FTL interaction". The "speed" at which the collapse of an entangled state happens has been shown experimentally to be many orders of magnitude faster than the speed of light. Since we tend to think of interactions with measurement apparatus as happening locally, it is hard to understand how a local interaction at one end of an entangled state can "instantaneously" determine a result another spatial region containing the entangled state that is separated from the measurement region by a space-like interval.

    This is why people have proposed that *something* must be transmitted FTL between the two regions of the entangled state. However we can state with certainty that any such "transmission" does not involve photons, or EM fields, or mass, or any "normal" entity that can be used to carry information between spatial locations. For example, even within the theoretical framework of QM, there is no way to use this mechanism for FTL communication, so it seems that entanglement and the collapse of entangled states does not raise a conflict between QM and special relativity.

    Well, if you really want to know how entanglement works (and I can appreciate that sentiment), and you are not satisfied with the answers that have been turned up so far, then perhaps you could consider a career in scientific research studying these questions in more detail. The rest of that statement is your personal prejudice I am afraid... my view is that the universe is how we find it to be. I appreciate a simple mechanism as much as the next person, but sometimes things are NOT simple.
  13. Jul 18, 2011 #12
    Sure, entanglement is an experimentally verified and accepted property of nature, but I find it strange that no one seems to have even the slightest idea how exactly does it work. In this regard, does entanglement have any parallels - any other process, which takes place without any insight of its mechanics? It just doesn't seem to be in accordance with the scientific approach, for if something happens then there must be a mechanic through which it happens, otherwise it is taking it for granted, something more typical in the area of theology...

    It is not even known what is that "something" which travels orders of magnitude over luminal, neither how does that "something" achieve that velocity... on the other hand, hidden variable theories may be able to explain the phenomenon without using mysterious entities, traveling in mysterious ways...

    As for complexity - there are two types of it - the first is complexity of systems that are made of lots of simple components that are easy to understand on their own - like a CPU for example - a modern CPU contains billions of transistors, in numerous circuits, combined into higher and higher order circuits, the system is complex but based on transistors which are not complex at all - this same type of complexity seems to apply to everything else - physics, chemistry, biology and so forth. And then we have the "other" type of complexity, which is not due to "stacking" of numbers of less complex units, but is simply complex, such complexity that is typical to theoretical areas of science that rarely yield in any practical applications.

    I am much more prone to embracing the first type of complexity, because it is plausible and practical, it works without assumptions and even thou complex is also consistent. IMO this type of complexity is a much better candidate for explaining test results, even if it employs millions or even billions of hidden variables, but with the entanglement theory it is the second type of complexity, a matter so complex no one even comes close to suggesting how it actually works, instead people just go with it, simply because experiments seem to produce a certain effect. "It works because it works" just doesn't ring a bell for me, there is nothing for logic and reason to hold on to...
  14. Jul 18, 2011 #13


    User Avatar
    Science Advisor
    Gold Member

    Yes there are parallels. All quantum mechanics falls into the same category. So does general relativity. No one has any idea how any of it "works" other than the mathematical formalism. We simply apply the rules and get the best answer possible for a prediction. This is science.

    And before you go too far, I suggest you take a moment to learn why Bell's Theorem ruins your hypothesis (as it did Einstein's similar 1935 hypothesis). Because I don't think you want to stick your neck out too far before you get to square 2. You are currently at square 1, you have grasped the EPR argument (or at least some of it, it is actually quite sophisticated). Bell is square 2. Aspect is square 3.

    I will be glad to walk you through it if that helps.
  15. Jul 18, 2011 #14
    @afstgl actually people do have suggestions for how entanglement works such as non-local pilot waves, holographic projection... but since these ideas are not experimentally testable these ideas are considered purely interpretational/speculative

    What is certain scientifically is that naive realist models like you suggest have been ruled out.

    (I even have suggested a simple explanation of how it works, check my home page)
  16. Jul 18, 2011 #15
    What is the address of your homepage? I would like to look at it.
  17. Jul 18, 2011 #16
    if you left-click on a username it gives you various links.
  18. Jul 18, 2011 #17
    afstgl, my thinking is much like yours. In the case of two photons with linear polarization along a particular axis if the "A" photon passes through a linear filter aligned with the y-axis (zero angle) then by construction the "B" photon will pass through a linear filter aligned with the y-axis (zero angle). But the problem is there are an infinite number of axes we could measure with respect to (i.e. 90 degrees, 10 degrees, 10.1 degrees, 10.11 degrees, ....). Penrose in his book "The Road To Reality" makes this point clearly. So now each photons needs a notebook with infinitely many entries one for each angle this clearly makes no sense. Maybe it is possible to specify the answer (pass or no pass) for the y-axis and the x-axis and come up with an equation to fill in the infinite number of results for the angles in between. I think this fails but it has been a long day and I am not up for thinking this through right now. You might want to look at Penrose's crack at this.

    On Bell's inequality I think I understand the QM result it is the "classical" or at least no signaling result I am still trying to understand.

    Gisin's paper "Can relativity be considered complete? From Newtonian nonlocality to quantum nonlocality and beyond" talks about non-signaling correlations. Also interesting, also I have not understood it yet.
  19. Jul 18, 2011 #18
    unusualname, I do like your

    h) No wave-function collapse or decoherence mechanism is required
    (The cat is dead or alive, but we have to "look" to know which one)

    That is the way I think about it. What we know is limited and best described by statistics but reality is what it is, is definite.

    On the other hand I like Feynman's the particle takes every path idea. So maybe I am not so definite.
  20. Jul 18, 2011 #19
    My model is consistent with decoherence, but decoherence is just the correct statistical explanation of the evolution of states (rather than an explanation of macroscopic non-superposition), in fact that has to be the case since decoherence "mechanism" has been observed experimentally.

    (btw the most remarkable thing about my model is not simply as an explanation of QM but rather that it claims relativistic physics is due to discrete time evolution)
    Last edited: Jul 18, 2011
  21. Jul 18, 2011 #20
    When you speak of a causal relationship you are in the camp of (1.) the unmeasured particle somehow now having the attribute of the measured particle -without being measured-the key point to consider. If the unmeasured particle is measured for a different quantum property and you accept that it has this property plus the unmeasured matching property of the other particle then you've found a way around Heisenberg's uncertainty principle. I would say that by measuring a different property for the particle nothing can be said about the entangled property of this particle. This would hold true even if you had two particles totally entangled (if that is even possible). It is only the act of measuring the same property of an entangled state for both particles that will result in their being measured as 'having' this entangled property.
    I prefer (2.)addendum- Clearly, my point of view is that an entangled state allows an entangled measurement to occur if it is made on both particles. The entangled state then is a clear potential for both particles, never a clear actuality unless it is measured.

    Another point to consider AKA the cat problem -you measure particle A for the entangled property but in a confined unreadable box. Particle B is 'then' measured and the measurement is observed. Only then is the reading for Particle A observed and found to match. Which collapse came 'first'. In my way of thinking when the measurement is made this particular entangled particles state collapses to it's measured state whether this measurement is observed or not. The same applies for both particles.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook