Contextuality-by-Default 2.0: Systems with Binary Random Variables

TL;DR

This paper introduces a refined version of the Contextuality-by-Default theory for binary random variables, using multimaximal couplings to better capture contextuality in finite systems, aligning with quantum and behavioral science applications.

Contribution

It proposes a new definition of contextuality based on multimaximal couplings, improving the theoretical framework for analyzing binary systems.

Findings

The new theory aligns with previous in cyclic systems.

Multimaximal couplings provide a more accurate measure of contextuality.

The approach is applicable to quantum and behavioral systems.

Abstract

The paper outlines a new development in the Contextuality-by-Default theory as applied to finite systems of binary random variables. The logic and principles of the original theory remain unchanged, but the definition of contextuality of a system of random variables is now based on multimaximal rather than maximal couplings of the variables that measure the same property in different contexts: a system is considered noncontextual if these multimaximal couplings are compatible with the distributions of the random variables sharing contexts. A multimaximal coupling is one that is a maximal coupling of any subset (equivalently, of any pair) of the random variables being coupled. Arguments are presented for why this modified theory is a superior generalization of the traditional understanding of contextuality in quantum mechanics. The modified theory coincides with the previous version in the important case of cyclic systems, which include the systems whose contextuality was most intensively studied in quantum physics and behavioral sciences.