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

Oberver Collapsing Wave funcions vs. Wheeler-Feynman Absorber

  1. Mar 13, 2008 #1
    I’m new to physics forums, I have a question regarding the collapsing wave functions and the concept of superposition of states. I am a geologist (with a rather limited education in quantum mechanics) however I have recently become very interested in the concept of collapsing wave functions. The standard interpretation, as I understand it, is that small particles like electrons or photons when emitted exist in a superposition of states that spread out and increase (the number of possible states) as a wave until an observer collapses their position and poof! it becomes a particle again. This interpretation is of course demonstrated by observations of the classic two-slit experiment where and electron's probability wave spreads out through both slits and interferes with itself until the electron hits the detector and is observed to be a particle once again. Personally, this seems to raise some puzzling philosophical questions like "what qualifies as an observer?” In the book Schrödinger’s Kittens by John Gribbin a rather interesting thought experiment is outlined:

    A single electron is contained (hidden) in a box in such a way that its probability of position spreads out uniformly until it encompasses the entire volume of the box. There is a gate that slides down to divide the box in half so that there is an equal probability that the electron exists in either half of the now divided box. Now, each half of the box has a gate which when opened allows the electron to move into a separate respective room containing a kitten. In each room is a detector which when stimulated by a passing electron releases a poisonous gas to kill the kitten. The two kittens, each of whom with their lives now entangled with the probability that the electron exists in one or the other half of the initial box, would seem to hang in a state of superposition (dead or alive) until an observer comes along and opens a room to see which kitten has expired, thus collapsing the wave function.

    Or are the kittens intelligent enough to be wave-collapsing observers?

    One solution put forth by Gribbin's book comes from the Wheeler-Feynman absorber theory (originally meant to explain resistance in radio transmission) seems to say that the uncertainty of the electrons position is merely an illusion. The electron sent out a sort of pilot-wave (retarded wave) and received a confirmation wave (advanced wave) backward through time. Thus, the electron "knew" which detector it was going to intercept before the experiment began.

    Essentially electrons communicating with electrons through time using photons. The implication of this seems to be that a photon cannot be emitted unless it has received a conformation wave from the electron that will absorb it in the future. Strange indeed! But it makes a lot of intuitive sense to me. Does this theory continue to hold up in current interpretations?

    Sorry to be so long winded but it felt like there were a lot of pieces to set up such a question.
    Last edited: Mar 13, 2008
  2. jcsd
  3. Mar 13, 2008 #2
  4. Mar 13, 2008 #3
    I understand that probability cannot be applied to an event that has already occurred, however that does not get at the question of a supposed superposition of states that exists until a measurement is made. I am really wondering if advanced waves are commonly accepted as a viable culprit for this logical discrepancy.
  5. Mar 13, 2008 #4

    Ken G

    User Avatar
    Gold Member

    Not for complex systems like cats, no. If you want my take on this, I'd say that we should generally stick with Occam's razor and not introduce unnecessary "philosophical" baggage that does not in any way change our predictions-- the goal of science is to make predictions and surround the results of experiments with the minimal conceptual scaffolding required to make sense of the result. As such, we should not impose the way we like to think about things onto the simple outcomes of the experiment, unless doing so is helpful in making useful predictions.

    That may sound like a reasonable stance, but I argue it is not the stance being taken by Gribbin (and perhaps not Wheeler-Feynman either, though I don't know about the possible benefits of that approach, and although Wheeler is given to more fanciful interpretations, Feynman had not one unscientific bone in his whole body). Let's start with the concept of "measurement". What I would say that process is, is the identification of potential outcomes of the measurement, coupled with a physical means (not something ephemeral like "consciousness") for decohering the interference between the amplitudes associated with the different outcomes. So the destruction of coherences (associated with "collapse" of the wavefunction) is not a mysterious byproduct of measurement, it is the express purpose of measurement. The destruction of these coherences forces the quantum system to behave classically, which is the familiar outcome for all the machinery of science-- and that familiar outcome was our goal of doing the measurement in the first place.

    Thus, thinking that coupling cats to quantum systems will put the cat into a quantum state is very backward-thinking-- the real point is that it will put the quantum system into a classical state! Here a "classical state" means "in a state which is either one way or the other, with no surviving coherences between the amplitudes of differing outcomes". This is also called a "mixed state", but I prefer "classical state", because that is what classical thinking is based on. The key point here is, the resulting classical state is engineered on purpose by us, that's what is meant by measurement, so we can hardly blame "the universe" for working that way. It can by chance happen naturally too, which is why a cat can achieve the same "classicalization" of the quantum system as can a proper measurement. (The way it happens naturally involves the coupling of the quantum state to so many "noise modes" that untangling those random correlations would be impossible even in principle. This is the mysterious step, but seems inescapable when transitioning from the quantum realm to the classical world we are familiar with.)

    When one uses this philosophical framework, there is no need to think of the kittens as being "entangled" in the quantum sense, instead they are simply correlated, like you cannot be at work and at home at the same time. The reason this is the "minimal" interpretation is that there is no experiment you can do on two kittens that will ever find a difference between what Gribbin calls "entanglement" and what is simple correlation. In short, you can always imagine the kittens are "either alive or dead I just don't know which yet", there is never any reason to imagine a combination or superposition. Frankly, I think a lesson was missed when that kind of thinking is used.
    Last edited: Mar 13, 2008
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook