The EPR Paradox

First of all, a short explanation. Although there are other examples of the Einstein-Podolsky-Rosen (EPR) paradox and violations of Bell’s inequalities, we choose this version because we can provide you with actual data to use in formulating your own solution to the paradox. A source of two correlated identical particles of opposite spins sits on the straight line between two identical particle detectors. Each detector can measure the polarization state of the entering particle, and each detector has three polarization switch positions (1, 2, and 3) and two display lamps (green and red). Each time the experimenter pushes the button, the two correlated particles are shot out of the source in opposite directions into the detectors. The data show two patterns:

(1) For runs that have the same switch settings on the two detectors, the same color lights flash on them.

(2) For all runs, without regard for switch settings, the pattern of flashing is completely random.

This experiment gets to the heart of QM and the application of its three rules for events. We can use classical mechanics to explain the first pattern: let the two particles carry the same instructions to be applied at the detectors. For example, this instruction set might work: flash red at switch positions 1 and 3; flash green at switch position 2. But this classical scheme with predetermined instruction sets will not handle the second pattern. Why not? What is the surprising conclusion?

Reproduced here is a small part of a data set for the experiment (from the Mermin reference in the answer). Each entry shows the switch settings and the colors the lights flashed for each run. The switch settings are randomly changed from run to run.

31GR      13RG     31RR       33GG

21RR      31RG     33GG       11GG

22RR     12RG      31RG       13RG

33GG     13GR      31RR         31RG

11GG      22GG     33RR          23GR

23RR      12RG      32RG        31GR

32GR      12GR      31RG         23RG

12GR       22GG     11RR         22RR

12RG       23GR      23GR       12GR

11GG      33RR       12GG       32GR

12GR      23GG        21GR       12GG

22RR      23GG        13GR      31GG

12GG       33RR        33GG      32RG

33RR       23GR      11GG        21GR

11RR       21GG      12RR        22GG

Answer

There seems to be no classical thinking that would reproduce the data set. A predetermined instruction set would be akin to an algorithm for generating random numbers but no such sets of numbers are truly random. One must accept the conclusion that Nature is quantum mechanical and therefore classical physics is only an approximation. The rules of QM agree with the results, but the details are too complicated mathematically to present herein. The references provide the extended discussion.

Even more surprising is the suggestion that locality is violated. That is, information from the first detector passes to the second detector without passing through imaginary spherical surfaces surrounding each, as if more dimensions exist in our world! Someone, someday, will determine a fundamental reason for this behavior of nature.