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

Susskind on double slit

  1. Feb 29, 2012 #1
    I'm watching Leonard Susskind's online QM lectures, and in it he talks about the double slit experiment. He says that if the particle leaves a "mark", allowing you tell which slit it went through, (for example a "1" vs a "0" on a computer read out) this destroys the interference pattern. But I'm curious, does the "mark" have to permanent? For example if the particle, traveling through one slit, causes a dial inside an empty room to twitch then return to normal, would that destroy the interference pattern?

    My thinking is that, since information is never destroyed, the twitch of the dial is still 'recorded' in the room, and recoverable, in some sense.

    if you want me to find the exact lecture and post the video, I can.
  2. jcsd
  3. Feb 29, 2012 #2
    That would be most appreciated if you would do that.
  4. Mar 1, 2012 #3
    The part I am talking about is the discussion around 14.00 minutes, and at 25.00 minutes.

    15 min: "The question is whether it left a mark."
    25 min: "So if no mark is left that you or anybody else can ever determine afterwards whether the electron went through the upper route or the lower route, then there is an interference pattern."

    I'm interested in whether it's the mark that's important, or the ability to determine it afterwards, that destroys the interference pattern. If it leaves a mark (changes the apparatus), but then that information is immediately lost, what happens to the interference pattern?
    Last edited: Mar 1, 2012
  5. Mar 1, 2012 #4
    Certainly if the information is lost, but if wave function collapse had occured, you can't reverse that collapse. Whether wave function collapse has occured or not is another story.
  6. Mar 1, 2012 #5
    You can pretty safely assume that any detectable interaction will destroy the interference pattern.

    One might ask why collisions with the wall in which the slits are placed doesn't destroy the interference pattern. It is because the wall is too much of a macroscopic system to be affected, the interaction between the electron and the wall is not detectable.
  7. Mar 1, 2012 #6
    Well that's where it get's complicated though. He talks later about Schrodinger's Cat. That once you start dealing with "wavefunction collapse" you're really including the experimenter in the experiment. Whether they looked at the result or not.

    I mean it's a simple thought experiment: perform the double slit exp. with a detector at one slit, inside an empty room, with only a camera looking at the photographic plate (so we can see whether the interference pattern is destroyed or not).
    Now it seems obvious to me that we would see no interference pattern, even though we (the experimenters) have no information about what slit the particle went through. But then doesn't that solve the Schrodinger's Cat problem? It implies that the cat is either alive or dead, definitely, whether we look or not. It seems to me the situations are identical.
  8. Mar 1, 2012 #7
    Actually, I retract that. The situations are not the same, because in the first case you have information about the system through whether the interference pattern was destroyed or not.
  9. Mar 1, 2012 #8
    Well, there are such things as quantum eraser experiments, where you have a device which records which path a particle goes through the double slit. If that record made by the device is automatically destroyed before the particle hits the screen, we actually find there is an interference pattern, as if the which-way information was never collected! Of course, you can always interpret this by saying the device was in a superposition of states, and it's only when a human comes and observes the device readout that the wave function actually collapses. Still, it's very telling that if a record of the which-way information EXISTS, regardless of whether anyone actually reads the record, then the interference pattern is not there, but if you destroy the information before anyone has seen it, then the interference pattern is there. That's what makes a lot of people lean toward decoherence: they feel that it is the creating of a record that leads to the appearance of collapse, not some mystical property of conscious observation.

    And it has occurred to me that quantum eraser experiments give us a (gruesome and impractical) way in principle to tell whether consciousness-causes-collapse is correct. Just have a human see the record, and then have the human be so utterly destroyed that there is no way even theoretically to find out what they saw. Then if human observation really leads to collapse, we'll see no intereference pattern, whereas if there is no special role for humans then we will see an interference pattern.
  10. Mar 1, 2012 #9
    Measurements are irreversible.

    The which-way info in quantum delayed choice experiments is potential. It doesn't exist in the classical sense. By being potential, it can be erased.

    Also, the apparent lost of interference pattern - it would be hidden fringe and anti-fringes which form the scatter pattern.
  11. Mar 1, 2012 #10

    Ken G

    User Avatar
    Gold Member

    Yes, it is a common misunderstanding of quantum erasure that you will somehow magically induce an interference pattern if you later decide to erase the detection. That isn't true, you never get an interference pattern in the raw data, whether you erase the detection or not. So that is the answer to the original question. But the subtlety of erasure experiments is that if you have a pair of entangled particles, you can do a higher-level analysis, which involves correlating the outcomes of experiments on the two datasets. Thus you have the option to subdivide the original dataset subject to how it correlates with the other dataset, and then if you erase the detection, you can recover an interference pattern, not in the original raw data, but in the correlated subsets of the original data. In other words, quantum erasure says that it is indeterminate why you don't get an interference pattern when you detect which slit-- you can either interpret it as because no interference occurs in the first place, or you can interpret it as interference does occur, but two sets of interference patterns interfere with each other to destroy the interference that does occur! Put differently, it means that detection doesn't necessarily prevent interference, it could also simply act to mask the interference that is always there whether you detect or not. So detection produces not under-interference, but over-interference, if you will!
  12. Mar 1, 2012 #11

    Ken G

    User Avatar
    Gold Member

    The issue is how "big" is the system you have in mind. You are right that if you separate the slits from the photons, or the cat from its environment, then the two situations are the same, and we would say "the cat is alive or dead we just don't know which," or "there is or is not an interference pattern there whether we look or not." But the problem with the cat paradox is you can also choose to look at the larger system, which includes the cat and everything it interacts with, and then ask if the uber-system has within it an alive or dead cat.

    I would say the real message of quantum mechanics, which resolves that paradox, is that this kind of question simply does not make sense. If you want to know about the state of the cat, or even define the term "cat", you must separate the cat from its surroundings, performing what would be thought of as a projection onto the cat subspace (or more correctly, the subspace of all observations that you would treat as directly informative about the state of the cat). But when you do a projection, it's like looking at the shadow of the system-- you are not seeing the whole system any more. So if the shadow is a cat that is alive or dead and we just don't know which, the question remains, what is the actual state of the entire system?

    According to quantum formalism, the entire system is indeterminate about the state of the cat, because there is no cat there, there is just an entire system. So something about observing pieces of a system disrupts the "unitarity" of the full system, and that is what is happening when you observe the cat or the presence of a camera next to a slit-- since you are external to the system, you are disrupting the unitarity of that environment, and that is what allows you to make definite statements about whether or not detections will occur and wavefunctions will collapse.

    Now when it gets really interesting is when you demand that you yourself be considered part of the system. Then the projection onto your subspace is what we call doing physics. This brings in the paradox of the "many worlds" interpretation-- if we claim that the physics we do is restricted to a subspace of that which physics is meant to be informative, then we must say we are ourselves just part of the full system, and we are not in fact observing entire systems only pieces of the "many worlds." Of course, others point to the paradox of even defining what physics is if it applies only to a subspace of that which physics is supposed to inform us about! It really gets down to whether we think physics is a tool used by humans, or humans are a tool used by physics. That is a philosophical question, so is the place where the cat paradox becomes an issue for philosophy rather than physics.
  13. Mar 1, 2012 #12
    Ken G, you might be interested to know that von Neumann gave a rigorous mathematical proof in his famous "Bible" of quantum mechanics that you get exactly the same experimental results regardless of how you answer the question "At what stage does the wave function collapse?".
  14. Mar 1, 2012 #13

    Ken G

    User Avatar
    Gold Member

    That is interesting, but doesn't surprise me. My own view is that the different interpretations of quantum mechanics are all equivalent, insofar as they are restricted to being intepretations of the observations that gave us quantum mechanics (and not assertions about how observations would come out that we now find technologically impossible). Also, they all differ mainly in how they answer, when or if collapse occurs, so I'm not surprised that this issue could be viewed as independent of the actual predictions of quantum mechanics.

    I would also point out, however, that the reliability of von Neumann's proofs came into question when his "no hidden variables" proof was shown to rely on certain hidden assumptions that have subsequently come into question. von Neumann is not likely to have made any errors in logic, but one can always circumvent a proof by simply rejecting one of its assumptions.
  15. Mar 1, 2012 #14
    There was nothing wrong with his hidden variables proof, it was just that his definition of what a hidden variables theory is was overly restrictive. But I can assure you, the proof I'm talking about doesn't contain any philosophical assumptions, it's just a straight up mathematical demonstration that the the predictions of quantum mechanics are independent of the point at which you cut the von Neumann chain, meaning regardless of where you draw the line between observer and observed.

    However, the philosophical conclusion he draws from this result is quite controversial. He argues that regardless of where you cut the chain, there are particles on either side of the cut, so there's no principled way to make a cut in the middle of the chain. Thus he proposes to cut the chain at the top, between the human mind and the human brain, because he believed the mind is nonphysical and thus you don't have physical particles on both sides of the cut. Hence consciousness-causes-collapse was born. (He was also open to placing the cut between God and the Universe, but he didn't flesh out this possibility.)
    Last edited: Mar 1, 2012
  16. Mar 1, 2012 #15
    Reference to the Ghirardi thought experiment I provided you with?
  17. Mar 1, 2012 #16

    Ken G

    User Avatar
    Gold Member

    I am pretty much in agreement with von Neumann, in that I feel the placing of the "cut" is very much up to the physicist and relates to how our minds work, which distinguishes it from approaches that see the "cut" as a physical process that we should study just like any other. But I see three possible approaches that are all consistent with the general view that the cut is not a target of physics study:

    1) place it at the mind as he does, so consciousness "causes" collapse. This is the philosophical stance that physics is the tool of the mind, so anything that physics does is subjugated to the workings of the mind.
    2) don't place it anywhere, call it an illusion formed by the workings of the mind. This is similar to #1 because it is still relegated to the action of the mind, but it claims that the mind cannot affect reality so apparent effects are illusions of the mind and the reality is something different. This gives rise to the "many worlds" approach.
    3) say that the placement of the cut is fundamentally indeterminate, like the placement of when the Doppler shifting of light occurs. This is actually my preferred approach, all we can say is that a cut occurs, and reality itself is undecided on the issue of when it occurs. The cut is a holistic property of how mind, physics, and reality interrelate. This is probably the most consistent with Bohr's general approach.

    But I agree that the main conclusion we should draw is that physics itself does not care which of these approaches, or any other, we adopt in regard to the "cut". It's more an issue of personal philosophy about how best to interpret quantum mechanics. That would be refuted if studies that treat the cut as something physical give rise to new physics about the nature of the cut. Alternatively, the above three approaches would be supported if progress on new theories is advanced by adopting the perspective that understanding the mind is a necessary component of understanding physics.
    Last edited: Mar 1, 2012
  18. Mar 1, 2012 #17

    Ken G

    User Avatar
    Gold Member

    I'd like to hear more about that. Does it relate to the double slit?
  19. Mar 1, 2012 #18
    No - but nonetheless an interesting experiment.

    PM me your email and I'll send you the PDF.
  20. Mar 1, 2012 #19
    Here is a description of the thought experiment I gave in another thread:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook