Strategy vs. direct-response method: evidence from a large online experiment on simple social dilemmas

Abstract

This paper examines behavioral differences in sequential games that arise when choices are elicited via the direct-response method, where players observe the choices made by those who acted earlier and respond with a single choice, or strategy method, where they make their choice(s) in response to all possible choices by those who acted before regardless of whether such actions were taken. We conducted a 2  ×  2 between-subject large-scale online experiment with over 8000 participants on Amazon MTurk, manipulating both the elicitation method and participants’ ex-ante beliefs about player 1’s choices via an information-provision treatment. In neutrally framed binary-choice games, a sequential Prisoners’ Dilemma and a mini-Ultimatum Game, we document that the strategy method does not appear to alter player 2’s preferences, i.e., their tendency to reward cooperation or reject unfair offers. However, it reduces the tendency to reward defection and reject fair offers, which we interpret as a reduction in mistakes.