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

Acceptance of Many Minds

  1. Nov 25, 2007 #1
    I am a layperson interested generally in quantum mechanics, with a reasonable grasp of the formalism (as reasonable as any layperson, at least). I do not deal with physicists generally, and so it is hard to gage how accepted various models and interpretations are. In particular, I was for a long time told and later taught (both by teachers and textbooks) that quantum mechanics was fundamentally probabilistic, a state of affairs which seemed to me to be extremely objectionable; I was therefore rather surprised when I learned of an apparently simpler theoretical model which was deterministic and purportedly consistent with experimental results. Totally ignoring the existence of such a model (or more frequently mentioning it as a curious aside) seems like a very strange thing for educators to do, and the public generally appears to have developed an incredibly inaccurate conception of MWI. My questions generally are, how accepted is many minds/many worlds as an interpretation in the scientific community, and how many physicists believe respectively that collapse (as a fundamental process) does occur, that collapse might occur, and that collapse probably doesn't occur? When wavefunction collapse is invokved in the literature, does it generally refer to an approximation to decoherence or to a fundamental principle?

    Among physicists who believe collapse occurs as a fundamental process, what motivates this belief? Is it accepted that if many worlds were in fact to exist, decoherence would cause their interaction to be non-observable? Is it accepted that if there were no collapse there would be many worlds? Is there some feature of collapse besides computational simplicity which balances out its additional complexity? Is the apparent "weighting" of possible observers considered objectionable?

    I do not mean to frame any of these questions as accusations, I am merely curious as to why collapse models are presented so much more frequently to laypeople and students. You could certainly argue it doesn't really matter, but it has a significant effect on the way some concepts are presented. I also do not understand why the absence of discriminating evidence would support explicit collapse instead of refuting it as the more complex model (although again I do not mean to suggest there is no good reason, just that I do not know it; I do not even know if this is the case outside of my perhaps not entirely representative experience).

    Thanks.
     
  2. jcsd
  3. Nov 25, 2007 #2
    I doubt that many physicists really "believe collapse occurs as a fundamental process", since such a process would be too mysterious to ignore. I'd say the mainstream belief is that one should "shut up and calculate", or in other words, that collapse is simply a feature of the most efficient mathematical trick to correctly predict data (and this is the part that will always be true, whilst history shows that interpretations regularly have to be changed). My colleague calls it a tool that the monkey uses, implying we may not even be capable of correctly interpreting why it works. You should read some Feynman.

    If many-worlds exist, there is debate over whether they would be observable. (Obviously, "decoherence" is generally expected to explain our immediate experience, but there is a question as to whether some weak interaction could be found between worlds - most likely involving gravity since that has not yet been unified with quantum mechanics). What do you mean by "weighting of possible observers"?
     
  4. Nov 25, 2007 #3
    I have heard it occasionally objected that in MW the only a priori reasonable assignment of probabilities to outcomes is based on count, whereas it is empirically the case that certain single outcomes are more "probable" than others. I am unsure how mathematically serious this objection is (my impression is not terribly) or how seriously it is taken.

    Thank you for the answer. Philosophically speaking then, do you think most physicists would find classical determinism plausible, or at least not obviously false? It seems at face value as though the only random element of QM is the collapse of the wavefunction (in most flavors).
     
    Last edited: Nov 25, 2007
  5. Nov 25, 2007 #4
    MW is nice because it implies that the laws of physics are deterministic (and time reversible) despite what individuals observe. But an undisputed fact remains: mainstream QM (basically including MW) says it is impossible (even in principle) for an individual to gather enough data to predict (with certainty) the outcome of some experiments; I guess that whatever you mean by "classical determinism" must be obviously false.
     
  6. Nov 26, 2007 #5

    Fra

    User Avatar

    Regardless of interpretation (MWI or not) I think the induction of a probability must be a serious issue.

    The idea that there is always and exact probability distribution wether we know it or not, is I think almost semi-classical to it's philosophy, because it implies the deterministic evolution of probabilities. It makes no sense.

    A more realistic scenario is that we need to assign/estimate a probability based on incomplete information. And as it seems the part of our history that is retained and encoded in our own state seems to me, to be the only valid basis.

    In this view, concept of objective probabilities is not fundamental, they are rather emergent. And considering the non-unitary possibilities that arise here, it's either a misnomer to call it probability, or we need to update the probability theory itself.

    Decoherence descriptions attempts to resolve this by incorporating the environment too, but I guess like Cesiumfrog said, this is not a real solution. It is somehow the "subjective view" of god. The of course it's a piece to cake to restore unitarity, but that's no solution. I think we need the inside view, no imaginary outside view.

    The outside view will probably be excellent for accelerator experiment when we control and monitor the environment. But that's a special case only as I see it.

    /Fredrik
     
  7. Nov 26, 2007 #6
    David Deutsch and the CQC recently put out a press release claiming that he has shown mathematically that the "bush-like branching structure created by the universe splitting into parallel versions of itself " exactly reproduces the probabilities predicted by the Born rule- demonstrating that parallel universes do weakly interact after decoherence and that interaction affects the possible outcomes of measurements-

    it will be very intersting when the paper is published
     
  8. Nov 26, 2007 #7
    That doesn't sound like anything I would have meant to say.

    I don't think there is any consensus on it yet, but after reading a review article I get the impression that probability must have philosophical problems in any context.
     
  9. Nov 27, 2007 #8

    Fra

    User Avatar

    Ok, sorry. It was what I thought you possibly meant.

    What I meant to say was that the "big" view requires more information capacity than the observer owns. That's what I mean with "gods view", with no emphasis on god except beeing an observer with big information capacity. I've seen people call it the "birds view", which is then used with a reductionist philosophy to explain the relevant/constrained view.

    This is what I mean with this beeing no complete solution. I mean relative to the collision center in an accelerator MAYBE we could "be the bird" relatively speaking, but in the general case, say cosmology I doubt the same idea makes sense.

    /Fredrik
     
    Last edited: Nov 27, 2007
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Acceptance of Many Minds
  1. Mind experiment (Replies: 5)

  2. Mind experiment (Replies: 8)

Loading...