AN INTRODUCTION TO THE EPR–CHAMELEON EXPERIMENT

Abstract

On September 27 (2001), as a side activity to the Japan-Italy Joint workshop on: Quantum open systems and quantum measurement, the first; public demonstration of the dynamical EPR-chameleon experiment was performed at Waseda University in order to give an experimental answer to a long standing question in the foundations of quantum theory: do there exist classical macroscopic systems which, by local independent choices, produce sequences of data which reproduce the singlet correlations, hence violating Bell's inequality? The EPR-chameleon experiment gives an affirmative answer to this question by concretely producing an example of such systems in the form of three personal computers which realize a local deterministic dynamical evolution whose mathematical structure is very simple and transparent. In the experiment performed on September 27 the local dynamics used was not a reversible one because the interaction with the degrees of freedom of the apparatus was integrated out giving rise to an effective Markovian dynamics which, although mapping probability measures into probability measures, did not preserve the +/- 1-values of the spin (or polarization) observables. This feature was criticized by some of the partecipants and the following two questions arose: i) is it possible to prove that the Markovian evolution, used in the experiment, is indeed the reduced evolution of a bona fide reversible evolution? ii) if the answer to question (i) is affirmative, is it possible to reproduce the EPR correlations by simply considering empirical averages of +/- 1-values, as one does in usual EPR type experiments? An affirmative answer to these questions was given in the paper [AcImRe01] and it is briefly reviewed in what follows.