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

Thought experiment

  1. Aug 4, 2015 #1
    I've been reading some cursory stuff about quantum mechanics, and I want to understand it better - so I have a little thought experiment that I'm curious about.

    Let's say there are two rooms that are connected by a thin wall with no window. A person is told to stand in one of the rooms for a short amount of time. In the other room, there's a bucket that's perched very precariously on a high shelf. Based on some particular tiny-particle activity that gets measured by a computer, there's a small chance that the bucket will get prodded and fall off the shelf while the person is standing in the other room. The observer is not told about the bucket or the possibility of hearing a clattering noise. No one else is nearby, and the bucket is automatically reset between tests.

    If the bucket falls, then the observer hears it, but if the bucket doesn't fall, then the observer is not aware that there was a bucket in the first place.

    Will the bucket ever stay on the shelf?
     
  2. jcsd
  3. Aug 4, 2015 #2

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    What happens has nothing to do if the observer is there or not - its entirely got to do with the small chance the bucket has of falling. Assuming in any small time interval the probability of the bucket falling is the same it will follow a Poisson distribution:
    https://en.wikipedia.org/wiki/Poisson_distribution

    Specifically it will be 1 - Pr(no occurences in the time interval) = 1 - e^-λ

    Why you think this has anythiny to do with QM though has me beat - its a simple problem in statistical modelling.

    Thanks
    Bill
     
    Last edited: Aug 4, 2015
  4. Aug 4, 2015 #3
    As far as I understand the setup as described above, the answer depends completely on how the "prodding rules" are set up, i.e.
    So, if the "tiny-particle activity" is a one time event only, then the bucket will stay in place most of the time. But if the "tiny-particle activity" are multiple events which each can prod the bucket, then the result will depend on e.g. the probability of prodding, the number of events, and the time spent doing each experiment. Depending on this the bucket will occasionally fall off the shelf.

    So, as far as I understand the setup, the answer to
    is yes.

    Furthermore, personally I can't see a clear connection to quantum mechanics in this post yet...

    EDIT: And bhobba beat me to it... :smile:
     
  5. Aug 4, 2015 #4
    My guess is he read some of: "the tree doesnt make a sound if nobody hears it fall" , "moon's not there if nobody sees it" , "cat's both alive and dead"
     
  6. Aug 4, 2015 #5

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    Maybe. To the OP QM is a theory about observations that appear here in an objective common-sense classical world. In that common-sense world the tree makes a sound objectively regardless of an observer.

    Thanks
    Bill
     
  7. Aug 5, 2015 #6
    Sorry if the description wasn't clear. Here's a simpler version of what I was trying to say:

    Take a case where a certain observation causes a wave function to collapse. Now modify that setup so that if a certain result occurs, then it is reported to an observer - but otherwise, nothing is reported, and the outcome is left uncertain.

    Since many potential outcomes are possible until the point when one is observed, is the system "forced" to resolve in one of the manners which would cause it to be reported?
     
  8. Aug 5, 2015 #7

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    Well - first collapse is not part of QM - only some interpretations have it - some don't.

    You, like many people, misunderstand what observation means in QM. It means something that occurs here in an objective common-sense world. If a concious observer is present or not makes no difference. Its a common mistake because popularisations and beginner texts do not make clear the meaning in QM. Its an unfortunate choice of words, but we are stuck with it for historical reasons.

    Thanks
    Bill
     
  9. Aug 5, 2015 #8
    Let's say the "tiny-particle activity" is implemented using a Stern–Gerlach setup (please see link 1, link 2 for details), which prods the bucket depending on where the particles hit the screen (e.g. "spin up" yields nothing and "spin down" yields a prodding of the bucket, and anything else yields nothing). Then, the only thing that can make the bucket fall off the shelf is the presence of the magnet in the experiment, regardless if the screen is present or not.

    By the way, here is a nice clip describing the Stern-Gerlach experiement.
     
  10. Aug 5, 2015 #9
    That's interesting, DennisN!

    bhobba, let's say that one possible behavior of the particle causes a real-world common-sense action, and the other possible behavior of the particle does not. What happens?
     
  11. Aug 5, 2015 #10
    When a microscopic event (like an atomic decay) is connected to a macroscopic event (knocking a bucket off a perch) leading to different macrostates, this is an observation, and collapses the microscopic event.
    For some reason, a macrostate is not able to be observed to be in superposition.
     
  12. Aug 5, 2015 #11

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    You need to be more specific.

    Thanks
    Bill
     
  13. Aug 6, 2015 #12
    That reminds me a bit. Not only the observer pick up objects, but the objects pick up the observers. The objects are meaningless without observers. The various interactions -- visible or invisible, known or unknown -- between the observers and objects, build both of them.

    Yeah, that might not be some sort of quantum physics, but it may relate to some kind of physics, something more deep, subtle, and mysterious.
     
  14. Aug 6, 2015 #13

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    There are quite a few interpretations that are objective and exist independent of any observation.

    QM does not say that only observations exist - it says observations exist and is silent on anything else.

    Thanks
    Bill
     
    Last edited: Aug 6, 2015
  15. Aug 6, 2015 #14
    Okay! Let's...try. As we get more specific, my understanding of the topic will get thinner.

    An electron is fired through a hole and comes out with either an up or down spin. For up-spin, the particle is detected by a computer, which then prods the bucket. For down-spin, the particle is completely unnoticed by the computer, so the bucket is not prodded.

    One of the two possible outcomes results in a real-world action - does this force an up-spin to occur every time?
     
  16. Aug 6, 2015 #15

    Nugatory

    User Avatar

    Staff: Mentor

    That would be something along the lines of passing the particle through a Stern-Gerlach device, and if the particle is deflected upwards it will hit the detector attached to the computer and if it is deflected downwards it will fly out the window never to be seen again?

    The computer itself is a macroscopic object, so there's no possibility of the system ending up in a state that is a superposition of "bucket tipped" and "bucket not tipped". Thus, whether someone is watching or not, the bucket will either be tipped with the particle spin-up, or not tipped with the particle spin-down.
     
  17. Aug 6, 2015 #16

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    The observation of up or down spin is independent of what the observation then causes. Bucket or no bucket has no influence on what's detected.

    Thanks
    Bill
     
  18. Aug 13, 2015 #17
    Both leaving the bucket alone, and pushing it over are real world actions that anyone might know. By leaving the bucket hidden from the observer, you try to make the bucket a secret. A secret that gets revealed by a real world action. The action of leaving it alone was also "real" as you consider it. It was just a secret.

    If the guy could see the bucket, if there was no wall, he would look and say, "hey, there's a bucket balanced on an ejection mechanism, hooked up to a detector, that appears to be set to emit electrons thru a Stern-Gerlich magnet ... hmmm, the ejector fires on spin up".

    Making the setup secret is not what makes something unknown, in the quantum sense (as I understand it).
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook