Effect algebras are not adequate models for quantum mechanics (Q535625)

The author proves the following facts: If an effect algebra admits an order-determining set of states (= probability measures), then it is a sub-effect algebra of \([0,1]^X\) for some set~\(X\) (thus the effect algebra is representable by fuzzy sets). If an effect algebra \(E\) admits at least one state, it admits hidden variables in the sense that there is a state-preserving morphism of \(E\) to some MV-algebra. These results indicate that effect algebras are not appropriate models of quantum mechanical systems. As an alternative, the author suggested sequential effect algebras in [\textit{S. Gudder} and \textit{R. Greechie}, Rep. Math. Phys. 49, No. 1, 87--111 (2002; Zbl 1023.81001)]. These structures, equipped with an additional operation, the \textit{sequential product}, and with a modified definition of a dispersion-free state, admit to overcome the above inadequacies. It is shown that a sequential effect algebra admits an order-determining set of dispersion-free states iff it is a sub-effect algebra of \([0,1]^X\).