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

Explain me Einstein-Podolski-Rosen effect

  1. Nov 14, 2004 #1
    Can someone explain me Einstein-Podolski-Rosen effect, its nature and why is it happening?
  2. jcsd
  3. Nov 14, 2004 #2
    Okay, the Einstein-Podolsky-Rosen paradox. :biggrin:

    This is Bohm's version of it that most people mean when they talk about it.

    Two particles are emitted at the same time in the decay of a single particle and each of these two particles have properties X and Y that are correlated with the properties X and Y of the other particle (this is called "entanglement"). Correlation means that if you measure both particles for property X or measure both particles for property Y then if a particle gives you a value of 0 for that property, the other particle will always give you a value of 1 for that property.

    (You can also measure one particle for property X and the other for property Y but this gives you random values and doesn't tell you anything useful here. Or so most people think. It is, however, part of the solution to the paradox and I'll get back to this later.)

    All this correlation here really means is that you can know what a particle will say for a property even before you measure it if you have already measured the other particle for the same property.

    That's all quite simple and is just like an experiment you could set with a few playing cards. :smile:

    What is different about the usual interpretation of standard quantum theory to playing cards is that the value of property X or the value of property Y for both particles are not decided until you measure one particle and it decides if its value will be 0 or 1 for that property (a decision called "wave-function collapse").

    As usual, the matching opposite value is then found in the other particle if you then measure it for the same property. But this is actually very strange. How does this other particle know what the measured particle decided when this measured particle itself didn't decide until you measured it? :bugeye:

    What's even stranger is that the particles can be so far apart that the information about what the measured particle decided must pass to the other particle faster than the speed of light for them to match up! This violates Einstein's Theory of Special Relativity that nothing can travel faster than light.

    What's stranger still is that these faster-than-light effects cannot be used for any technology or even directly observed.

    All very strange. :uhh:

    Einstein didn't like standard quantum theory at all and called an effect like this "spooky action at a distance".

    Other versions of quantum theory were proposed to try and avoid the strange effects like these of the usual interpretation of standard quantum theory. It was then found by Bell that you could test standard quantum theory against these other versions. And experiments were done and it was discovered that standard quantum theory was correct.

    So now lots of physicists say that Einstein was wrong about standard quantum theory and that faster-than-light effects do exist but they are strange in that we can't use them for any technology or observe them directly.

    And that's usually the end of the story of the Einstein-Podolsky-Rosen paradox. :smile:

    However... :wink:

    A few people who have been paying attention to developments in the field of the interpretation of quantum theory know that things have changed. I'm one of those people. :biggrin:

    Yes, Einstein was wrong about standard quantum theory but it now seems he was right that faster-than-light effects do not exist.

    It appears that most physicists make a mistake when they think that the measured particle deciding what property X or Y will be immediately forces the other particle to have the matching opposite value. It doesn't. What matching 0 or 1 values the properties X or Y will have in both particles were actually decided the moment the original particle decayed into two particles but this needs to be described in the correct way which is much harder to understand than the standard wave-function collapse approach.

    The solution involves taking standard quantum theory to its logical conclusion and what you then have is a precise version of something arrived at by Bohr in a rather less precise way with his "principle of complementarity". Bohr said that a quantum particle could be descibed either as a particle or a wave and you used whichever description made most sense in any experiment but never both use descriptions together. Both descriptions were "complementary".

    The solution to the Einstein-Podolsky-Rosen effect has been to realize that all sorts of properties of particles have their own modern version of complementarity that come not from experiment but from standard quantum theory itself.

    It turns out that you can just as easily say that measuring property X of one particle instantly makes the other particle generate a random value of its Y property as you can say it makes the other particle generate a matching opposite X value, and that values of both X and Y for both particles can exist in the appropriate description from the moment the original particle split into the two particles you later measure. This shows that it's a mistake to think that measuring the X or Y property of the first particle forces the other particle to do anything at all.

    This is the what I meant by part of the solution of the paradox was that when you measured the X property of one particle you could also measure the Y property of the other particle but most people didn't think it did anything useful other than produce random values. You can use "complementary" descriptions that have matching 0 and 1 values for the X and Y properties of both particles right from when the start when the original particle split up into the two particles.

    It's difficult to put all that in a way that is easy to understand. It certainly took me a while to understand the solution myself. :bugeye:

    All this stuff just means is that all the values of everything are decided at the start and no information is going anywhere. The correct way to describe it is somewhat complicated but that's all it says. :smile:

    This post ended up taking a lot longer than I intended too. :yuck:

    I'll finish this post with a short version: :wink:

    The short version is that the Einstein-Podolsky-Rosen effect by which one particle seems to affect the properties of another by communicating at faster-than-light speeds appears to just be caused by the way physicists have been describing the two particles and by realizing that there are other descriptions available and that these other descriptions have the properties decided from the start shows that the measured particle then doesn't affect the other particle in any way. This means that there are no faster-than-light effects.

    And it also shows that quantum theory is very, very difficult to understand. :bugeye:

    Anyway, it's then no surprise why faster-than-light effects cannot be used in technology or directly observed. They don't exist. :approve:

    Don't anyone worry if they didn't understand all that, the short version at the end has all you need to know. :smile:
    Last edited: Nov 14, 2004
  4. Nov 17, 2004 #3
    If I interpret these words according to their 'plain' sense, then I am forced to conclude that Bell's inequality is satisfied!

    By assuming that a measurement performed at the site of particle #1 can have no bearing whatsoever on the result of a measurement performed at the site of particle #2 – and vice versa – it follows that Bell's inequality must be satisfied (... unless an implicit assumption (e.g. "contrafactual definiteness") which goes into Bell's derivation has been deemed invalid).

    But then, you allude to something apparently altogether unknown to me:
    Can you elaborate on this (... and/or offer a link or a reference)?

    ... Yet, you do give somewhat of a hint regarding your line of "attack" on the problem:
    It appears to me that I am in the class of just such people. ... What more can you tell me? (A link or a reference is also welcomed.)

    More specifically, you wrote:
    Please, inform me: what is this "correct way"? (Again, a link or a reference is also welcomed.)
  5. Nov 17, 2004 #4


    User Avatar
    Gold Member

    I see your 'development' gets at the heart of the one thing that's always troubled me about the theory as it stood until recently:

    "...the value of property X or the value of property Y for both particles are not decided until you measure one particle..."

    It always starts with the claim that the two particles do not have these properties until you measure them. How do we know this? (Or I should say, how did they know this before this new development).
  6. Nov 17, 2004 #5
    The interpretation of quantum theory that I'm describing is the "consistent histories" of Robert Griffiths and Roland Omnes and the "decoherent histories" of Murray Gell-Mann and Jim Hartle. The former is more about interpreting experiment and the latter is more about decoherence and cosmology, but they amount to the same thing.

    It's essentially standard quantum theory and a "many-histories" approach (which is just a more sensible way to say "many-worlds" that are not necessarily real). Bell's inequality is not satisfied, as it shouldn't be.

    One of the most important things this interpretation does is show what can and cannot be said in quantum theory. This actually makes it very hard to understand at first and I've seen more than a few physicists say that it's classical and ignores Bell's inequality, that it removes free-will or that it's just nonsense.

    Nope, it's just very hard to understand at first.

    Is a photon a particle? Is it a wave? No, it's a quantum particle. It's a "something" about which particle and wave descriptions can be used seperately but not together. These descriptions are complementary. This is Bohr's "complementarity".

    Consistent/decoherent histories is filled with the modern equivalent of that because that is what happens when someone wants to talk about the properties of particles in quantum theory. It ends up as complementary descriptions of states and superposition states, X spins and Z spins, position and momentum, and so on. Every possible history of what is being followed make up the incompatible -- but complementary -- frameworks.

    A framework is just a branching tree of the possible histories. You could have a framework for the possible histories of the X spins of a particle and another framework for the Z spins of the particle but both can't be used together.

    At it's most extreme, the precise position at all times of all the particles in every possible history of the universe could make up one framework, and the precise momentum at all times of all the particles in every possible history of the universe could make up another framework.

    In both cases it's the exact same particles but two equally valid descriptions that can be used seperately but not together as no particle has a precise position and a precise momentum at the same time. The frameworks are both "true" in a sense but are also incompatible.

    Best place to start if you want to know more is here:

    http://quantum.phys.cmu.edu/histories.html [Broken]

    I've spent a while studying it and I think I'm actually getting to understand it now. It's kind of mind-bending. :biggrin:
    Last edited by a moderator: May 1, 2017
  7. Nov 17, 2004 #6
    Yeah, we had things like that with properties being decided when a particle interacts with something and even the stranger consequence that properties are decided because the particle doesn't interact with something!

    Now we can say a particle has properties that were decided before measurement and it has other properties were decided before measurement but the particle is a "something" that sort of has -- but don't sort of have -- both properties.

    In some respects, this isn't an improvement. :biggrin:
    Last edited: Nov 17, 2004
  8. Nov 19, 2004 #7

    Thank you for the information and your prompt reply. Severe time constraints prevent me from moving quickly on the matter. However, a preliminary sweep of the topic has brought upon me some creeping suspicions that this interpretation will not find favor in my eyes.

    Whether or not this will truly be so remains, of course, to be seen. Perhaps, I will be pleasantly surprised. :smile:
  9. Nov 19, 2004 #8
    No problem. I'm making slow progress at understanding this interpretation myself but I'm still just a little bit suspicious of it. :smile:

    However, Gell-Mann had discussions with Feynman on it and they apparently ended up in good agreement. And when Gell-Mann and Feynman agree on something, I'm not going to dismiss it without being absolutely sure I understand it, so I continue to work away.

    What could have saved me time would be knowing that the frameworks -- and the histories that make each framework up -- are different ways of looking at aspects of something that will not be pinned down to any one overall picture.

    Quantum theory remains "fuzzy" but frameworks are glimpses of what may emerge from the "fuzziness".
    Last edited: Nov 19, 2004
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook