Matt Jones
University of Colorado, USA
EPR with people
A number of recent experiments in psychology have attempted to demonstrate contextuality in human behavior. Whereas analogous experiments in physics carry theoretical assurance of no-signaling from properties like spacelike separation, these human studies have no such guarantees, and in fact signaling is highly like to be present. This situation has been a motivation for the development of the contextuality-by-default approach of Dzhafarov and colleagues, which seeks to define contextuality in the presence of signaling. The present experiment takes a different approach, creating a situation with two human subjects in which signaling theoretically should not occur. Two subjects are placed in a closed room and asked to converse until a timer goes off. They then proceed into separate sealed rooms (connected to the first), each to answer one of two randomly selected binary questions on a computer. Each question takes the form of a 2x2 coordination game, a structure that is paradoxical for rational game theory but that humans routinely succeed at. Once data are collected, they will be subjected to the classical CHSH criterion for contextuality.
