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

B Quantum Physics -Double Slit Delayed Observation

  1. Oct 21, 2016 #1
    Thanks in advance to all who reply. Forgive me if this naive, I am not scientifically inclined, but I am VERY curious.

    After re-watching a documentary on quantum physics, I have a "thought experiment" that I would like feedback on. Maybe this experiment (or something like it) has already been done.

    What if the double slit experiment was conducted for 10 days. Each day, a device would RECORD activity at the slits and store the data to a new thumb drive. Another device would RECORD the patterns on the "back wall" and store that data to a DIFFERENT (new) thumb drive. NO HUMAN will "observe" the experiment during these 10 days, or exam what has been recorded on the thumb drives.

    At the end of each day, both thumb drives are labeled (ex "Day 1 - Slits" and "Day 1 - Back Wall"). At the start of each day, a new pair of thumb drives is used.

    On day eleven, the 10 pairs of thumb drives are lined up on a table. Five of the pairs are "randomly" selected, and the "slit thumb drive" of each selected pair is completely destroyed, having never been examined by a human.

    If you then examine the "Back Wall" thumb drive fore each of the 10 days, would there be any difference between those that no longer have a "slit thumb drive" in existence (and can never be "observed" by a human being)?
  2. jcsd
  3. Oct 21, 2016 #2


    User Avatar
    Science Advisor
    Homework Helper
    2017 Award

    Hi Bill,

    Unfortunately, something needs to be done to record the activity at the slit. That something destroys the interference pattern.
  4. Oct 21, 2016 #3


    User Avatar
    Science Advisor
    Gold Member


    The general rule is: If it is possible, in principle, to determine the path (slit) a particle traverses: then there will be no interference. Observation by a person, as far as anyone knows, is not a factor.
  5. Oct 21, 2016 #4
    Thanks BvU,

    I am not sure of your answer. Are you saying that permanently destroying the recorded activity before it is observed by humans makes no difference?

    I am trying to identify what amounts to an "observation" of activity.

    Is it human perception? I have read that if the "cameras" at the slits are turned off, the back wall shows wave interference. When the "cameras" at the slits are turned back on, the back wall shows no wave interference. What if instead of turning the "camera" off completely, you were able to cut a wire just behind the camera's sensor (whatever that is) such that the device is no longer capable of manifesting the information (real time OR recorded)?

    If such an inanimate "sabotaged camera" still destroys the interference pattern, why wouldn't a toaster pointed at the slits do the same thing?

    Thanks again to you and anyone else who replies.
  6. Oct 21, 2016 #5


    Staff: Mentor

    The best quick heuristic is probably this: anything that causes an irreversible change.

    No, because irreversible changes will be occurring long before you perceive anything.

    Won't matter. The sensor inside the camera that detects photons will still undergo an irreversible change, and that's sufficient to "observe" the photons coming through the slits.

    How do you know it wouldn't? If photons coming through the slits and hitting the toaster produce an irreversible change, then I believe it would destroy the interference pattern. Nobody has run the experiment with toasters in the apparatus, so we don't have actual data on the question.
  7. Oct 21, 2016 #6
    Thanks DrChinese,

    Please see my recent post (above) and give me your feedback (thanks!).

    In my original "thought experiment", it would NOT be possible to determine the path because the recorded data had been permanently destroyed (albeit, after the fact).

    What if you change my original experiment and use USELESS / DEFECTIVE / PLACEBO thumb drives on 5 of the 10 days? There would be no possible way of "knowing" / "perceiving" / "observing" slit activity (real time or after the fact).

    What's the difference between a camera that is "turned off" and one that has no means of relaying data ? When is the "camera" not a toaster?
  8. Oct 21, 2016 #7
    Thanks PeterDonis,

    You said : "Won't matter. The sensor inside the camera that detects photons will still undergo an irreversible change, and that's sufficient to "observe" the photons coming through the slits."

    Maybe I am relying too much on a simplistic explanation of the "double slit experiment". My perception was of a "camera" (the device used to monitor the photon just before or after it passes either slit) that is physically "affected" by the photon at all times (while turned on or off). I mean, the photon is there, the camera sensor is there, the lens, the wires, the curious scientists, etc). Turning the "camera" on (I thought) only makes it possible to report what that "physical affect" is. No?
  9. Oct 21, 2016 #8


    Staff: Mentor

    I don't think that's correct. The "camera off" setup has nothing that can be affected by photons in the paths of the photons going through the slits.

    I can't say anything more specific unless you can give a specific reference for where you are getting your description of the setup.
  10. Oct 21, 2016 #9


    User Avatar

    Staff: Mentor

    Which one? There are some very bad and misleading ones out there, and from the rest of your question I suspect that you've been victimized by one of these.

    Of particular importance: the presence or absence of a human being looking at the results of measurements at the slits is completely irrelevant. Any interaction with anything macroscopic (in this context, "macroscopic" means anything more than a few atoms, anything big enough to be thought of as a classical object, including any sensor of any type whatsoever and the memory cells in a thumb drive) counts as an "observation" or "measurement" and prevents interference at the back screen.

    Destroying the record of the measurement at the slit will no more affect the pattern on the back screen than tearing up the paper receipt for a credit card charge will cancel the transaction.

    (This is basically what @BvU said in fewer words above)
  11. Oct 21, 2016 #10


    User Avatar
    Science Advisor
    Gold Member

    Turning off the camera recording does nothing. The interference will still disappear if the information could have been used to determine the path. Admittedly, you can modify the example as much as you like and ask where the exact point is in which interference disappears. But the rule still applies: "If it is possible, in principle, to determine the path (slit) a particle traverses: then there will be no interference."

    An example may assist: if you place 2 polarizers, one each, behind the 2 slits, and there is no camera to otherwise record the path taken: If they are parallel, there WILL be interference. If they are perpendicular, there will be NO interference. See the bold above. The reason is that it would be possible to observe the polarization of photons as they hit the screen and determine the path when the polarizers are perpendicular. You don't actually need to take that step though, to eliminate the interference.
  12. Oct 21, 2016 #11
    Thanks Nugatory, BvU, DrChinese, and PeterDonis ,

    I understand what you are saying now.

    In this thread, you all seem to be saying that the wave collapses at the point where the photon has an INTERACTION (impact) with anything else.

    It seems like all of the "Discovery Channel" type presentations to lay people (like me) say that the wave collapses at the point of human OBSERVATION. I never gave thought to how a photon or electron could be observed without it also being interacted with.

    Apparently, mine is a widespread misconception. I just found this on Wikipedia (not necessarily a source of truth, but often an interesting read):

    https://en.wikipedia.org/wiki/Observer_effect_(physics) :

    "In quantum mechanics, there is a common misconception (which has acquired a life of its own, giving rise to endless speculations) that it is the mind of a conscious observer that causes the observer effect in quantum processes. It is rooted in a basic misunderstanding of the meaning of the quantum wave function ψ and the quantum measurement process.

    According to standard quantum mechanics, however, it is a matter of complete indifference whether the experimenters stay around to watch their experiment, or leave the room and delegate observing to an inanimate apparatus, instead, which amplifies the microscopic events to macroscopic measurements and records them by a time-irreversible process. The measured state is not interfering with the states excluded by the measurement. As Richard Feynman put it: "Nature does not know what you are looking at, and she behaves the way she is going to behave whether you bother to take down the data or not."

    Thanks again for your time.
  13. Oct 21, 2016 #12


    Staff: Mentor

    Not any interaction. The photons going through the slits are interacting with the slits, but that doesn't take away the interference pattern. (I don't say "collapse" because that is interpretation-dependent, and we should stick to things that are true in any interpretation of QM.) That's why I specified an irreversible change.

    These just illustrate why you should not try to learn actual science from pop science sources.

  14. Oct 22, 2016 #13


    User Avatar
    Science Advisor
    Gold Member

    PeterDonis already said this, but I will state a different way. It is not the interaction, but the combined context that controls this. Specifically, please reconsider the example I provided above where you place 2 polarizers, one each, behind the 2 slits.

    Please note: If they are parallel, there WILL be interference. If they are perpendicular, there will be NO interference. CLEARLY, there is an interaction in both cases. The only variable is the RELATIVE orientation of the polarizers. It should be clear that this variable does NOT fit the idea you mention of an INTERACTION. But it is part of the overall context, which is what you must look at.
  15. Oct 26, 2016 #14
    Interesting, but say you rotate one of the polarizers from perpendicular to parallel, does the interference pattern gradually show up?

    Also is this experiment something I could do with my green laser, camera lens polarizer filters, and a cd diffraction grating?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted