Quantum physics, as Alice might have observed, just keeps getting curiouser and curiouser. The latest example of quantum strangeness to cross my path is the delayed choice quantum eraser experiment.
I don’t pretend to understand the technical details, so I’ll rely on a pretty straightforward Wikipedia article. It tells us that the delayed choice quantum eraser is a variation on the famous double-slit experiment, in which a subatomic object’s status as a particle or wave depends on whether or not it is observed as it passes through the slit(s). When unobserved, it behaves like a wave. When observed, it behaves like a particle.
In the delayed choice quantum eraser setup, the subatomic object again has the option of passing through one slit (as a particle) or both slits (as a wave), depending on how and when it is observed. But there is a fascinating twist. Quoth Wiki:
[The experimenter] arranges to detect which one of the slits the photon passes through, but also [to] construct the experiment in such a way that this information can be "erased" after the fact. It turns out that if one observes which slit the photon passes through, the "no interference" or particle behavior will result, which is what quantum mechanics predicts, but if the quantum information is "erased" regarding which slit the photon passed through, a wavelike interference pattern can be observed.
However, Kim, et al. have shown that it is possible to delay the choice to erase the quantum information until after the photon has actually hit the target. But, again, if the information is "erased," an interference pattern can be recovered in a certain subset of the photons which reach the detector, even if the information is erased after the photons have hit the detector. [Emphasis in original]
An earlier version of the Wikipedia article on Answers.com provides a useful description of the experimental procedure:
First, generate a photon and pass it through a double slit apparatus. After the photon goes through slit A or B, a special crystal (one at each slit) … convert[s] the photon into two identical entangled photons … One of these photons continues to the target detector, while the other entangled photon is deflected by a prism to bounce off a mirror some distance away. Now, if the second photon (coming from slit A or slit B) is observed, it is known which slit the original photon went through, so the photon behaves like a particle. If the second photon’s paths from slit A and B are combined, the which-way path is not observed, and the first photon behaves like a wave. The experimenter can choose to observe or not observe the which-way information by erasing (or detecting) information about the second photon’s path.
The results from Kim, et al. have shown that, in fact, observing the second photon’s path will determine the particle or wavelike behavior of the first photon at the detector, even if the second photon is not observed until after the first photon arrives at the detector. In other words, the delayed choice to observe or not observe the second photon will change the outcome of an event in the past.
The Answers.com piece also includes some interesting discussion of what this may mean:
How can this be? It would seem that the "choice" to observe or erase the which-path information can change the position where the photon is recorded on the detector, even after it should have already been recorded.
One explanation of this paradox would be that this is a kind of time travel. In other words, the delayed "choice" to "erase" or "observe" the which-path information of the original photon can change the outcome of an event in the past. Another explanation would be that in fact both outcomes occur. The universe itself exists in a superposition of states in which either the original photon goes through slit A or slit B and in which the which-path information is either "observed" or "erased". This is described in detail in the Everett many-worlds interpretation of quantum mechanics.
The time travel explanation doesn’t seem very satisfying to me. On the other hand, the notion that "the universe itself exists in a superposition of states" is interesting. What might make this idea more palatable is if we stop thinking of "the universe" as existing in multiple states and instead think in terms of our thought processes. That is, we might say that our idea of how the photon should behave is what exists in an indeterminate state, and that we have to choose which behavior it should manifest only when we actually commit ourselves to observing it. Until the observation is made, nothing has actually happened – various possible scenarios are in play, but none has been in actualized.
In this view, it is the contents of our own consciousness that "exist in a superposition of states." The contents of our consciousness include the photons themselves and everything else in "the universe," all of which consist ultimately of mental events.
Of course, this Idealist interpretation raises many questions and problems of its own, but it does seem as if quantum physics is pushing us toward this position. Attempts to explain the phenomena by convoluted Realist schemes increasingly resemble the epicycles added to planetary orbits by pre-Copernican astronomers trying to rescue the idea of an Earth-centered universe. Maybe it’s time to take the Galileoesque leap – a quantum leap, we might say.