The Power of Reasoning: Experimental Evidence

Abstract

This paper presents an experimental investigation of how a systematic variation in the cognitive demands on subjects affects the optimal play. The innovation of this paper is the choice of a game, which we call the Game of Position. This is a two-player zero-sum game characterized by a dominant-strategy solution that involves iterative steps of reasoning. The equilibrium play is independent of mutual beliefs of players; hence inability of a subject to play the dominant-strategy unambiguously implies the failure of human reasoning prowess. We alter the two parameters of the game to vary the cognitive constraints, as represented by these steps of reasoning, on players. Our main substantive conclusion is that the frequency of the dominant-strategy play sharply increases as we limit the cognitive demands on players.

Keywords: Non-cooperative game theory, cognition, laboratory experiment

JEL Classification: C72, D83, C91