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

Delayed Choice Quantum Erasers

  1. Jan 25, 2009 #1
    I'm a very thickheaded layperson so be warned, I'll probably require a lot of patience.

    Anyway, I'm having a very difficult time understanding the delayed choice quantum eraser experiments. It seems to me that it is implying that either the idler photon is going back in time and changing the signal photon or the signal photon is somehow determining how the idler photon will behave. Yeah you don't see interference fringes in the overall pattern but doesn't that simply mean that the interference pattern is still there, just hidden among all the other subsets?

    Also does the copenhagen or any other nondeterministic/agnostic interpretations have an explanation that prevents the need for backwards time travel?
  2. jcsd
  3. Jan 25, 2009 #2


    User Avatar
    Science Advisor

    I speculated about this on another recent thread:
    Likewise, if the signal photon is detected first, in the standard Copenhagen picture I imagine this would collapse the 2-particle wavefunction in a way that changes the probability the idler will end up at different detectors--if the signal photon is detected at a position that's near a peak of the D1/D0 interference pattern and a trough of the the D2/D0 interference pattern, that should increase the probability that the idler will go to D1 and decrease the probability it'll go to D2, for example.
  4. Jan 26, 2009 #3
    Jesse you saying nothing in complicated way - means you know not answer. Why bluffing eh?
  5. Jan 26, 2009 #4
    Yes the fringes are "hidden" in the detections at D1 and D2.
    They only become visible if you consider the coincidence counter information to match either of those sets against detections at D0.

    I don't see how you got that the idler photon is going back in time.
    The photon will either arrive at D1/D2 (where the which-path information is erased) or it will arrive at D3/D4 (where the which-path information has been preserved).
    If that which-path information is preserved, you can't very well see an interference pattern just like the regular quantum eraser experiment without a delayed choice.
    Specifically, that which-path information is attained by collapsing the wave function (Copenhagen interpretation).
    The wave function needs to be in tact for it to interfere with its self and cause the pretty fringe lines.

  6. Jan 26, 2009 #5


    User Avatar
    Science Advisor

    Jarwulf's question was how you could explain the interference patterns in a way that doesn't require backwards in time influences. I gave an answer to that--if you measure the signal photons at D0 first, their probability distribution will be a non-interference pattern, but this causes a wavefunction collapse which alters the probability that the idlers go to different detectors in such a way as to create a D0/D1 interference pattern and a D0/D2 interference pattern. No backwards-in-time influences needed here, just the first measurement of the signal photons influencing the probabilities of the subsequent detection of the idlers at different locations.
  7. Jan 28, 2009 #6
    Okay if I'm reading correctly its detection of the signal photon/its location which influences which detectors the idlers will end up with.


    but isn't the signal photon measured before the idlers reach the first beamsplitter?

    Before the idlers for D1 and D2 reach the second beamsplitter the only difference between them and the D3+D4 idlers is the path they take after BSa or BSb. How would a photon 'know' what detector to go to preserve or remove its which-path information based on what location its signal counterpart ended up at?
  8. Jan 28, 2009 #7


    User Avatar
    Science Advisor

    In quantum physics you can't really imagine that the photons are taking any definite path until they are measured--think of the normal double-slit experiment, where if you look at the pattern on the screen created when you cover up the second slit (so you know all the photons went through the first slit) and then you look at the pattern created if you cover up the first slit (so you know they went through the second), the sum of these two patterns is not what you get if you leave both slits open simultaneously (and don't measure which slit they went through). In QM you assign the particle a wavefunction which involves a superposition of different possible positions (including the positions corresponding to all the different possible paths), and only when you measure the particle do you assume it will be found at a definite location, with the probability of it being at a given position being equal to the square of the "amplitude" the wavefunction assigns to that position. For an entangled system you have a two-particle wavefunction which assigns amplitudes to all combinations of positions for the particles, and then if you measure one of the particles (like the signal photon) this causes a sudden jump in the wavefunction which alters the amplitudes of finding the other particle at various positions.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook