Learning in the Trust Game

Claude Meidinger and Antoine Terracol

Table of Contents

Abstract

We use experimental data from a repeated trust game to estimate structural learning models that let us differentiate between the learning processes of the two players. We find that the two players cannot be described by the same learning process. Long-run simulations then show that the interaction of the two estimated types of players can lead to distinct outcomes.

Keywords

  • learning
  • beliefs
  • reinforcment
  • simulations