Albert Einstein never liked some of the counterintuitive predictions of quantum theory, arguing instead that there was a further, hidden layer to reality it failed to describe. But since the 1980s, Einstein's objections have been largely ruled out.
Now Karl Hess and Walter Philipp of the University of Illinois at UrbanaChampaign provide evidence that Einstein may have been right to be sceptical  there may indeed be another set of rules underlying quantum theory^{1}.
Quantum theory describes the behaviour of atoms and subatomic particles and says that their energy is 'quantized': it can be altered only in discrete jumps. Although Einstein himself made seminal contributions to quantum theory, he famously disagreed with the Danish physicist Niels Bohr about how it should be interpreted.
In 1935 Einstein, together with the physicists Boris Podolsky and Nathan Rosen, concocted a 'thought experiment' in which quantum theory seemed to permit 'spooky' action at a distance, whereby a measurement made on one particle instantaneously determines the properties of another particle, no matter how great the distance between the two particles.
Uncomfortable with this bizarre outcome, Einstein suspected that a still more fundamental theory underlies quantum mechanics (just as quantum mechanics underlies the older 'classical mechanics' of Isaac Newton). He invoked 'hidden variables'  quantities that do away with things like quantum uncertainty, but which cannot be measured directly. Bohr disagreed, arguing instead that we simply have to resign ourselves to the fact that quantum theory is counterintuitive.
There the argument floundered until the 1960s when Irish physicist John Bell showed that hidden variables could have observable consequences. He demonstrated that the outcome of the EinsteinPodolskyRosen (EPR) experiment differs if hidden variables do or don't exist.
When it became possible to perform EPR experiments for real in the 1980s, the results showed that, provided Bell was right, there were no hidden variables. They seemed to show that Einstein was wrong and Bohr was right.
Hess and Philipp find that EPR experiments don't necessarily rule out hidden variables at all, and there may, indeed, be another layer to reality. They argue that Bell overlooked a large class of possible hidden variables whose behaviour is consistent with the existing experimental findings.
They find that if hidden variables have properties that change over time, yet are related to each other, the predictions change. For example, the hands of a clock in London and a clock in New York circulate periodically and do not directly influence one another, but nevertheless the different times shown by each are correlated with one another.
If hidden variables are 'timecorrelated' in such a manner, Bell's theory breaks down. The researchers show that in such a case, the results of EPR experiments can be explained without needing to invoke the spooky action at a distance that Einstein considered so unlikely. This does not mean that hidden variables exist, just that they cannot be completely ruled out.
The conclusions of Hess and Philipp have been questioned by some other experts in the field (see, for example, refs ^{2}& ^{3}), who think that the duo have not unequivocally escaped the strictures of Bell's theory."

References
 Hess, K. & Philipp, W.Bell's theorem and the problem of decidability between the views of Einstein and Bohr. Proceedings of the National Academy of Sciences USA, advanced online publication, DOI:10.1073/pnas.251525098 (2001).
 Gill, R. D., Weihs, G., Zeilinger, A. & Zukowski, M.Comment on "Exclusion of time in the theorem of Bell" by K. Hess and W. Philipp. Preprint, (2002).
 Mermin, N. D.Shedding (red and green) light on "time related hidden parameters". Preprint, (2002).