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

Predictable unpredictables (very confused)

  1. Aug 18, 2004 #1
    I have never been the best at probability and statistics -- it has always struck me as somewhat odd but, nonetheless, I can usually grasp it to a greater or lesser degree. One of the assumptions that I have made is that probability works because of some underlying tendecies which are present or absent throughout a fairly constant portion of the population (ie, the reason that X% of Americans believe Y is related to certain genetic factors, upbringing, etc.). However, I have been completely befuddled by the fact that scientists now believe that particles can arise, change, move and cease without any cause (I believe it is quarks that do this (?)) --> ie, one such particle shot between two magnetic plates will go straight, while another identical particle will move upwards, without any changes in the conditions, so that the movement of the particles is known to be without cause. From what I have read, it has been largely accepted that there are no hidden variables, no missed causes, involved in the movement, creation and destruction of these particles. However, despite their being uncaused, the arising, movement and cessation of the particles can be represented as probabilities. How can something which is completely uncaused be anything except random. It seems to me that if these particles truly lack any causation or control they should be completely unpredictable; unless some underlying factor controls their motion, each particle should move completely independently of the any other and of anything else since each one, not subject to any causes related to the other particles, acts completely on its own --> it seems then that probability should not be able to make reliable perdictions about the particles. How can something that is uncaused and completely uninfluenced by anything else possibly be said to follow a predicable pattern ? Isn't that a contradiction.
    hoping for some help,
  2. jcsd
  3. Aug 18, 2004 #2
    Well, it looks like you are learning QM ! This is a great mystery of nature, but it looks like god is playing dice ! Indeed, we gave up the hope of making deterministic predictions, in the meaning you are quoting : that is the same causes lead to different effects. :surprise:
    Yes indeed, that's a chock.

    However, what we are still able to predict are the probabilities. The same experience repeated a huge number of times by physicist 1, will lead to the same proportions of A and B events as another performed by physicist 2. We agree :approve: But we both are unable to predict left or right. Just proportions.

    I feel this is not the deepest mystery in QM. Even though the probabilities do not come from statistics, the result is similar. We were already familiar with probabilistic predictions, from statistical mechanics. Particle/wave duality, that is basically the same as Heisenberg undeterminacy, seems to me far more disturbing. It is doubtlesly a matter of taste :

    The most disturbing fact from QM, is the fact that the electron in the Young double-slit experiment takes two paths at the same time, unless you watch it.

    Hey, by the way : Hawking thought that, not only does god play dice, he sometimes throw them somewhere we can't see them. That is in a black-hole, there is information loss. This hypothetical supplementary undeterminacy has recently been disproved. But the QM undeterminacy is now established with no doubt. It is not an hypothesis. It is Nature. It is now as well established as the fact that the Earth rotates around the Sun.
  4. Aug 18, 2004 #3
    thanks, but still confused

    thanks for your response but it still leaves me somewhat confused. Consider two identical situations, that is, two situations with the same causes in operation; the fact that the outcome in those two situations can be different means one of two things (though, ultimately, these two might be collapsable into one/the same option) : (1) the causes are not causes at all, that is, they have no effect on the particle --> I suspect this is wrong, b/c with the removal of the conditions, different outcomes might occur (ie, the causes condition certain outcomes and prevent others -- the removal of those conditions would remove the possibility of those same outcomes (?)), (2) the causes are incomplete (I am not saying there are hidden causes, only that the causes that are present are not enough to determine the outcome) or not fully tied to the event (they condition it --> like a road --> the road conditions the path that one can drive, but not the way that one can drive it (ie, you can weave, go fast, go slow, etc.) --> the road is a condition tied to many effect)) --> the fact that the same causes can produce different effects means that the causes are not tied, or at least are not tied completely to the effect. I suspect that 2 is the correct choice/option. The only thing I can think, that is, the only way that I can imagine the possibility of getting a probability out of this situation, that is, from the fact that the particles follow a conditioned but not a caused/determined path, is that certain features of the quark make it more disposed to one path than to another (just as certain features of the human personality make a certain percent of people more disposed to driving fast and straight down a road), but I fear that this idea is really just a disguised form of asserting the presence of a hidden variable. You could say that the conditions allow a certain number of outcomes and the particle HAS TO pick one and just "picks" randomly, but this lacks the logic that I am so desperately after and also defies our ability to pick probabilities; some normalizing, regularizing force must be in effect in order for the particles to reliably have the same probability for certain courses of action -- it CANNOT be utterly random. So what is it, beyond the conditions, that accounts for this predictable unpredictability ? Does anyone know ? Can anyone point me toward a good layman's book which deals with this ?
    thanks much,

    P.S. : wasn't there just an article in the N.Y. Times in which Hawking admitted that he had been wrong in thinking that black holes ate up information ?
  5. Aug 19, 2004 #4
    You should check out the EPR paxadox. I am not quoting any link, you can google it, even on PF forums I am pretty sure you will find a lot of stuff for your greedy mind. Try to find the original paper (Einstein Podolsky Rosen, "Can quantum-mechanical description of physical reality be considered complete?" Physical Review 41, 777 (15 May 1935). )
    Bohr's answer : N. Bohr: "Can quantum-mechanical description of physical reality be considered complete?" Physical Review 48, 696 (15 Oct 1935). (Niels Bohr's response to EPR)

    Bohr replied in two steps. The first step was basically : "Damn EPR, they issued a fairly well written paper, but there MUST be something wrong in it. It is so beautifully written, I could not figure out a sensible answer yet. But I'll do it soon." The second step (I think) the paper I quoted above.

    PS : yes, he did admit loosing his bet and everything. This is indeed another nice story.
  6. Aug 19, 2004 #5
    But the particle doesn't have to pick one, and often doesn't pick one. It often picks them *all*!!

    "reliably have the same probability"-->is what QM is: a framework for predicting that probability. :wink: If it were random, there wouldn't be any reliability.

    If what you are looking for are interpretations of the explanations, then on the shelf of any bookstore there will be many (just read critically). The fact is, we can't prove dinterpretations. Basically, if we can't measure *what* is happening *between* point A and point B, then we can't say for sure that anything is happening. And if we could measure it, it would change the outcome of the experiment. :yuck:

    It kind of boils down to the particle-wave duality concept: Why do the behaviors change? Why does a particle interfere with itself? Or even, "how"? No one knows, and since we haven't found an experimentally-verifiable explanation, most aren't guessing. We have to leave some questions to ask God when we die. :approve:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook