PDA

View Full Version : Entanglement



Icarius1
2009-Jul-01, 03:12 AM
After listening to the recent Astronomycast on entanglement and waveforms I had a question about the supposition that knowing whether one particle is spin up or spin down instantly changes the the other particle to the opposite.

I am just wondering why people think information has to be moving between the particles or some faster than light communication has to be occurring when from my perspective if one particle is observed as being spin up, then the other was always spin down, our observation of one only made it clear which was which.

If we only know the state of a particle by observing it, I would have thought that that doesn't mean the particle -doesn't- have a state, just that we don't know what it is. Each particle is in a certain state already, just unobserved.

Why is it 'creepy' (to quote) that knowing one particle's state means the other particle is the other state?

robross
2009-Jul-01, 05:03 AM
After listening to the recent Astronomycast on entanglement and waveforms I had a question about the supposition that knowing whether one particle is spin up or spin down instantly changes the the other particle to the opposite.

I am just wondering why people think information has to be moving between the particles or some faster than light communication has to be occurring when from my perspective if one particle is observed as being spin up, then the other was always spin down, our observation of one only made it clear which was which.

If we only know the state of a particle by observing it, I would have thought that that doesn't mean the particle -doesn't- have a state, just that we don't know what it is. Each particle is in a certain state already, just unobserved.

Why is it 'creepy' (to quote) that knowing one particle's state means the other particle is the other state?

The quantum-mechanical explanation is that until you observe the first particle's spin, it does not have a definite spin, merely a probability. The act of observing the particle causes the wave function to collapse and this results in a definite spin for the particle. Then, once having observed the spin of this particle, the spin of the other particle is instantly known to be the opposite, even if that particle is on the other side of the universe and cannot possibly have any opportunity to be notified of what spin it should have.

I do not claim to understand this, I'm merely describing what "they" say happens.

Rob

aastrotech
2009-Jul-01, 08:49 AM
I think that the idea includes the idea that by applying a charge to a particle you can change its spin. Even if you don't know or observe its spin you know that its spin has changed. So in the particle pair even if you change the spin of one of the pair at random every whichaway when you observe it you can know that the other particle has the opposite spin.

I think what they are saying, I don't know what to call it a theory or speculation, is that there is some kind of subspace connection between the two particles. I guess that if you imagine a wave on water as having a conical shape but the cone is stretched into a long string and in addition to the wave going up and down like a water wave the string can whip around (in subspace) and get entangled with the string of its particle mate. Even if the particles are across the universe from each other they still share this connection. Ordinary light and matter has to travel on the surface of the water their speed governed by the dynamic of the water. The information that these particle pairs exchange isn't governed by the dynamic of the water because they are in some sense part of the same thing like the two ends of a wormhole.

I think it goes somthing like that.

papageno
2009-Jul-01, 12:21 PM
I am just wondering why people think information has to be moving between the particles or some faster than light communication has to be occurring when from my perspective if one particle is observed as being spin up, then the other was always spin down, our observation of one only made it clear which was which.


What you have in mind is a so-called hidden variable.

Experiments have been done to distinguish hidden-variable interpretations of quantum mechanics (spin is determined but hidden before measurement) from the canonical interpretation (spin is not determined before measurement): the hidden-variable fails, the canonical does not.

Look up delayed-choice experiments: it is possible to set up experiments where one measures the spin of one particle, but delays the choice of the outcome of this measurement. The measurement of the spin of the other particle gives outcomes that depend on the outcome of the first particle, even if the choice for the first measurement is delayed after the measurement of the second.
It does not matter when you look at the results of the measurement of the first particle: the measurement of the second particle gives alway a correlated outcome.

If the spin were a hidden variable as you picture, then the outcome of the second measurement should not correlate to the delayed outcome of the first.

dwnielsen
2009-Jul-01, 01:57 PM
And, in case you are in search of the common name (you may already know it): "Bell's theorem".

Ken G
2009-Jul-01, 04:53 PM
Yes, when people first look at Bell's theorem, it is natural to think "this is too complicated, isn't there an easier way of seeing what is mysterious about entanglement?" But the answer is no, nothing short of the complexity of Bell's theorem will successfully uncover the problem with hidden variables that pagageno alluded to, or the "it's one or the other we just don't know which" situation from the OP.

Icarius1
2009-Jul-05, 03:02 PM
Thanks I just read the wiki on Bell's theorum, it became somewhat clearer why the hidden variable theory is not favoured.

Cheers

papageno
2009-Jul-06, 08:28 AM
Thanks I just read the wiki on Bell's theorum, it became somewhat clearer why the hidden variable theory is not favoured.


Based on experimental results, the current trend is in favour of non-local effects, whether it is the traditional interpretation or a (non-local) hidden-variable interpretation.

In any case, it still hard to wrap you head around this problem.

Ken G
2009-Jul-06, 03:29 PM
Based on experimental results, the current trend is in favour of non-local effects, whether it is the traditional interpretation or a (non-local) hidden-variable interpretation.

In any case, it still hard to wrap you head around this problem.Personally, I think it helps a lot to excise the word "effect" from the thinking about entanglement. I've seen an awful lot written about entanglement that is misleading or downright incorrect, because it over-interprets the meaning of that nebulous word. How many times have we seen words to the effect that "you can alter one particle and it instantly alters the other?" That really isn't what entanglement does, it is our information about the particles that are intstantly altered, but that makes sense because that information is in our heads. The key point is that for physicists not privy to the information we have, they will not perceive any "alteration" of the other particle whatsoever, and the limited information they have will serve perfectly well to make correct predictions (statistically) about that particle.

Put differently, in my opinion there is way too much emphasis on "nonlocal versus local", rather than the real issue, "statistical versus deterministic", or "information versus reality". The most incorrect things said about entanglement generally involve very deterministic language, and when one uses that, the nonlocality takes on a kind of mystical air that seems to violate relativity. It's true that purely local descriptions don't work either, but note the purely local descriptions are always expressed deterministically, confusing us as to the real source of the "weird" behavior. When the entire effect is expressed correctly, entirely in terms of correlations between unknowns, then it is much more subtle and not so obviously "weird". Some weirdness persists, but it is similar to the weirdness of the probabilistic aspects of quantum mechanics that are already there. It is kind of a higher-order weirdness above the common weirdness of QM, but it's not the weirdness of doing one thing to one particle and have that change the other particle, as in a "cause and effect" relationship.

In short, I'm saying that the core weirdness of entanglement is the same as the core weirdness of all quantum mechanics-- the complex relationship between our information about reality and reality itself. All entanglement does is explore the non-local elements of that weirdness, but it was already pretty weird locally! That's why hidden variables are always lurking in the background-- they were always an effort to get past our information and into an "actual reality", but their failure shows that we so far have never been able to cross that divide.

Jetlack
2009-Jul-06, 07:36 PM
"That's why hidden variables are always lurking in the background-- they were always an effort to get past our information and into an "actual reality", but their failure shows that we so far have never been able to cross that divide."

Thats a polite way of putting it :-)