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

Questions on observation

  1. Aug 30, 2007 #1
    Due to people who think they know what they are talking about i have heard different anwsers to the same question.

    By observing a wavefunction of an electron in an experiment such as young's double slit. Does the act of physically being observed with a photon collapse the wavefunction or does the knowledge of it being there force it to collapse?

    Also, a question on quantum entanglement. How do we know that the entangled particles haven't already "choosen" a spin direction? Why do we believe that after observation that they will "choose" opposite spins?
  2. jcsd
  3. Aug 30, 2007 #2


    User Avatar
    Science Advisor

    The point is to measure the spin along VARIOUS directions, not only along z-direction.
  4. Aug 30, 2007 #3


    User Avatar
    Science Advisor

    Nobody knows with certainty.
  5. Aug 30, 2007 #4
    I think you may have misunderstood my question? I asked in entanglement why do we think that the entangled particles choose their spins after they are observed?
  6. Aug 30, 2007 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    See, I think this is where most people have a problem in understanding this aspect of QM, because you need to incorporate the superposition principle into it. This is why both parts of your original question actually come into play.

    If you buy superposition principle (and there are many experimental observations that can only be explained by such principle), then you have to accept that the spins of the entangled pair consist of the superposition of all the allowed directions before each one is measured. This is because they came out of a quantum state that allows for such superposition of states. This is no different than the superposition of paths that a quantum particle can take when it passes through a double-slit.

    Thus, in an entanglement, what makes it different than the classical situation (where the direction of angular momentum is definite, but we just don't know it yet till we measure it) is the superposition effect.

  7. Aug 30, 2007 #6
    Morga -- this is at the heart of one of the open questions in QM -- regarding measurement. I won't expound my personal views -- except to warn you not to take everything that people say at face value. At least ask for references which have been published in respected, peer-reviewed journals. The issue is so foundational that it borders on metaphysics, i.e. what is reality, so it invites a lot of cranks and crackpot theories. It is also never an issue in practice: if you give a real experimental set up, the answer is always unambiguously calculateable.
  8. Aug 30, 2007 #7


    User Avatar

    Like has been pointed out already this is debatable, and might be given different descriptions depending on the fundamental view of things.

    But FWIW, to add a little bit of my personal view, I regard the wavefunction as a representation of a relation between the observer and what is observed. This means that logically speaking the wavefunction is not an objective property of what is observed, it is rather as much a function of the observer - a relation - or the description of the information of the object, relative to the observers full complexity, where the observers knowledge is obviously a part of defining the observer.

    I picture that the way an observer "informs himself" about whatever he is observing is by means of physical interactions. So, physical interactions and communications are in my view at least various views of the same type of things. But if think of it as communication, alot of the confusing things is not as confusing anymore.

    It may argued that the experimental results from the above mentioned stuff stays the same wether there is a human observers or not, but the observer concept doesn't restrict itself to human observers, moreoever the human operating the measurement apparatous are hopefully not in direct interaction with the electron anyway. There obviously got to also IMO be a relation between the electron, and it's nearest surrounding - EM fields, slits and whatever else is there - included. In that case the physical interaction and information transfer seems to pretty much be the same thing.

    If we are taking the viewpoint that the wavefunction represents information, relative to the observer, the answer seems natural. We update our bets(wavefunctions) upon arrival of new information, no earlier and no later.

    Take the silly poker game analogy. When does a player A his strategy, by updating this calculations based on new evidence? When the player B for sure have finally decided what card to play, or when sight of the laid down/played card from B actually reaches A?

    The answer is obvious, what is not so obvious though is what this has to do with physics and QM. Here I think alot of people differ in opinon. So what's right? Personally I don't care what's right, I want to know what's useful! So IMO, whatever theory is "useful" and makes me survive and outperform threats and competitors is good. Like life in general, and that seems to be pretty successful if we think that it all came out of a terrible mess.

    Like the poker player. He doesn't give a damn about what cards the other player REALLY has, as long as he wins! The winner is always "right" o:)

    Last edited: Aug 30, 2007
  9. Aug 30, 2007 #8


    User Avatar
    Science Advisor

    It is very easy to prove that the act of measurement changes spin, even without entanglement. See e.g. Eq. (41) in
    http://xxx.lanl.gov/abs/quant-ph/0609163 (to appear in Found. Phys.)
    For a situation with entanglement, see Eqs. (43)-(46).
  10. Aug 30, 2007 #9
    an observer is a type of causal system that is very sharply defined- meaning that specific states and a specific history are probabilistically required for it's current state- the superposition principle states that elements of an observers local environment that aren't causally connected to the observer's structure are in superposition until the observer makes a measurement- that is until the possible states of that element can have a noticeable effect on the structure of the observer/environment in some way-

    the observer will then see a world in which the history and current structure provide a causal basis for the observer's existence- this is where we get weird ideas like Quantum Immortality- because an observer is a quantum system and it can only observe local states that allow it's existence- if it dies the observer cannot observe the history which it is dead- it can only observe histories that it survives in- and since every possible outcome happens an observer will always continue to exist only in futures where it survives- just like an observer continues to only exist in worlds where there is air and water and laws of physics that consistently provide a causal basis for it's existence
    Last edited: Aug 30, 2007
  11. Aug 30, 2007 #10


    User Avatar
    Science Advisor

    This is based on the violation of Bell inequalities which we'd expect to be satisfied if the correlations were just explained by the particles being created in correlated spin states. See my post with the analogy to scratch lotto cards on this thread.
  12. Aug 30, 2007 #11
    Something like Rovelli's "Relational QM"?

  13. Aug 31, 2007 #12


    User Avatar

    Yes, relational QM is close to my thinking. There seems to be a number of people working of relational quantum models, and while they seems to develop it differently, there seems to be at least at the outset some common base, I probably share most of this base too, even though I have not read Rovelli's work in detail to say that I agree all the way.

    The reason why I have turned into this is because I find this to be a far more realistic
    interpretation of reality. It does get more complex, and I've come to conclude that the consistent resolution I can find is an evolutionary model based on the relational interpretation. Fortunately I think that this mess just created, also hides a big potential. This relational information mess, does seems to predict gravity all on it's own, and gravity can thus be said to be a kind of statistical phenomenon of self organisation. But nonone has yet worked this out. But IMO, the connection is intuitively cleare enough to motivate me.

    If I am not totally mistaken about Rovelli, he shares at least a similar thinking. My approach is currently to for a second try to free my mind from any standard procedures and instead try to think myself from as much scratch as I am able to, given the limited size of my brain. I have then started to work on the fundamentals, and along the way I can vaguely associate a little bit with elements from string theory and elements from LQG approaches. The string stuff knocks me off the chair by their IMO overly wild assumptions usually on the first page, so while I do recognize elements of it, I do not in general like the fundamental approach. The LQG approach I can read long with a smile on my face, but after a while I am not sure what their logic is. So I decided that there is no substitute for doing your own thinking.

    If strings are the game, I rest assured that I will find them - no need to assume them from square one. But I think it's healthy for the world that people have different ideas. It would even be stupid for everyone to go in the same direction. It's better to spread the risks.

  14. Sep 1, 2007 #13
    Concerning entangled particles pre-choosing spins among different axises: I'm basing this off a book published in 2004, so tell me if it's old and wrong or if I'm misinterpretting it.

    Bohm theorized that particles can have definite positions and locations (so I assume spins as well?), but both can't be measured simultaneously. Also, in order to conform with experimental results, we have to assume nonlocality for forces. So as long as you don't care much for this locality, the entangled particles can agree on the various spins beforehand (I think, at least).
  15. Sep 1, 2007 #14
    We do not believe, we measure...
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Questions on observation