A comparison of naïve and sophisticated subject behavior with game theoretic predictions
AUTOR(ES)
McCabe, Kevin A.
FONTE
The National Academy of Sciences
RESUMO
We use an extensive form two-person game as the basis for two experiments designed to compare the behavior of two groups of subjects with each other and with the subgame perfect theoretical prediction in an anonymous interaction protocol. The two subject groups are undergraduates and advanced graduate students, the latter having studied economics and game theory. There is no difference in their choice behavior, and both groups depart substantially from game theoretic predictions. We also compare a subsample of the same graduate students with a typical undergraduate sample in an asset trading environment in which inexperienced undergraduates invariably produce substantial departures from the rational expectations prediction. In this way, we examine how robust are the results across two distinct anonymous interactive environments. In the constant sum trading game, the graduate students closely track the predictions of rational theory. Our interpretation is that the graduate student subjects' departure from subgame perfection to achieve cooperative outcomes in the two-person bargaining game is a consequence of a deliberate strategy and is not the result of error or inadequate learning.
ACESSO AO ARTIGO
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=16316Documentos Relacionados
- Negotiation Strategies in International Organisations: A Game-Theoretic Viewpoint
- Crime and Punishment in Classroom: a Game-Theoretic Approach for Student Cheating
- A Comparison of Library of Congress Subject Headings and Medical Subject Headings*
- Understanding the subject's behavior in the interaction with a decision support system under time pressure and missing information
- Heteroduplexes of phiX174 and G4 DNAs: orientation to genetic map and comparison with predictions from nucleotide sequences.