This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

  1. Home
  2. ASSA Annual Meeting
  3. 2024
  4. Program
  5. Computational and Experimental Economics
« Back to Results

Computational and Experimental Economics

Paper Session

Sunday, Jan. 7, 2024 10:15 AM - 12:15 PM (CST)

Grand Hyatt, Bowie A
Hosted By: Economic Science Association
  • Chairs:
    Yaroslav Rosokha, Purdue University
  • Rosemarie Nagel, Pompeu Fabra University

Toward a Behavioral Taxonomy of 2x2 Games

Rosemarie Nagel
,
Pompeu Fabra University
Amil Camillo
,
Pompeu Fabra University
Fabrizio Germano
,
Pompeu Fabra University

Abstract

This paper presents a comprehensive analysis of the 2x2 ``ordinal'' games to derive a taxonomy of games, in line with behavioral rules and experimental evidence. We introduce a graph-theoretic method for computing similarity classes based on the neighborhood structure of games and the discontinuities of behavior within the games. Using this method, we derive taxonomies based on standard game theory and behavioral rules, such as iterated best responses to uniform random play (level-k) and rules based on social preferences (e.g., the equal split or social maximum). We contrast these taxonomies with actual behavior to derive a behavioral rule and experimental data-based taxonomy of 2x2 games. Our approach should provide a useful tool for experimental design and algorithmic game theory and identify strategically interesting games and classes of games that have received little attention in the economic literature.

Managing Strategic Complexity

Annie Liang
,
Northwestern University
Jeff Ely
,
Northwestern University
Ben Golub
,
Northwestern University
Chaofeng Wu
,
Northwestern University

Abstract

Standard game-theoretic analysis yields highly incomplete descriptions of behavior in complex games. In a large database of chess play, we quantify the significant margin by which machine-learning predictions trained on human play outperform a Zermelo's Theorem benchmark. We then define an interpretable heuristic for the complexity of a position and document two stylized facts using this heuristic: (i) complexity predicts deviations of play from standard game theory; (ii) the frequency of strategically complex positions increases with the ratings of the players. We then develop a tractable model of how players manage complexity. In general, stronger players may be more or less likely to complicate a position (relative to weaker players), and we provide a testable condition that distinguishes between these cases.

Cooperation Under the Shadow of Political Inequality

Yaroslav Rosokha
,
Purdue University
Xinxin Lyu
,
Purdue University
Denis Tverskoi
,
University of Tennessee
Sergey Gavrilets
,
University of Tennessee

Abstract

We study cooperation among individuals and groups facing a dynamic social dilemma in which the benefits of cooperation are divided according to political power obtained in a contest. The main theoretical and experimental results focus on the role of incumbency advantage. Specifically, incumbency advantage in the political contest leads to a rapid breakdown of cooperation in the social dilemma. In addition, we provide simulations based on the individual evolutionary learning model of Arifovic and Ledyard (2012) to shed light on the difference between the behavior of individuals and groups.

(Re)Inventing the Traffic Light: Designing Recommendation Devices for Play of Strategic Games

Mikhail Anufriev
,
University of Technology Sydney
John Duffy
,
University of California-Irvine
Valentyn Panchenko
,
University of New South Wales
Benjamin Young
,
University of Technology Sydney

Abstract

We present the results of a novel experiment investigating individuals' ability to offer incentive-compatible recommendations for strategic games. Subjects designed recommendation devices for five canonical 2x2 games played by Bayesian, expected-utility maximizing robots to achieve Pareto efficiency and fairness. Most subjects succeeded in achieving this objective in Matching Pennies and Battle of the Sexes, but only a minority found the desirable device in Prisoner's Dilemma and the two Chicken games. However, the vast majority of subjects designed an incentive-compatible recommendation device for the Chicken games. Subjects failed to recognize that strategic incentives meant the socially-efficient outcome could never be recommended in Prisoner's Dilemma. Our approach requires participants to use a holistic approach to equilibrium reasoning. Our findings suggest that equilibrium reasoning is most challenging for individuals when strategic incentives conflict with cooperative outcomes.

Discussant(s)
Colin Sullivan
,
Purdue University
Valentyn Panchenko
,
University of New South Wales
Amil Camillo
,
Pompeu Fabra University
Tridib Sharma
,
Mexico Autonomous Institute of Technology
JEL Classifications
  • C9 - Design of Experiments
  • C7 - Game Theory and Bargaining Theory