Subject Characteristics and Play of Repeated Games

How to cite

Duffy, John. 2020. "Subject Characteristics and Play of Repeated Games." AEA RCT Registry. AEARCTR-0006318. https://doi.org/10.1257/rct.6318-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

John Duffy*

* Primary investigator

Registry ID
AEARCTR-0006318
DOI
https://doi.org/10.1257/rct.6318-1.0
First registered
Last updated
Status
On going
Trial dates
2020-08-17 – 2020-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/6318

Abstract

We study play in a series of indefinitely repeated prisoner’s dilemma games that vary in the probability that the game continues from one period to the next. Subjects play against a known, programmed grim trigger strategy, so that it is possible that they decide how to play each indefinitely repeated game based on expected earnings from their action choices. In a treatment variation, the payoffs from the programmed grim trigger strategy go to another human subject, while in the baseline treatment they do not. We collect measures of subjects’ cognitive abilities and their social value orientation. We also elicit priors as to how frequently across the various games they face that they will choose the cooperative or defect option in the first round of each new repeated game. We are particularly interested in whether these measures correlate with play in the repeated game.

Intervention

We study how subjects play a series of indefinitely repeated prisoner’s dilemma games that vary in the probability that the game continues from one period to the next. Subjects play against a known, programmed grim trigger strategy, so that it is possible that they decide how to play each indefinitely repeated game based on expected earnings from their action choices. We collect measures of subjects’ cognitive abilities and their social value orientation. We also elicit priors as to how frequently across the various games they face that they will choose the cooperative or defect option in the first round of each new repeated game. We are particularly interested in whether these measures correlate with play in the repeated game.

Experimental design

We vary the continuation probability across games. Treatments involve different orderings of the sequence of indefinitely repeated games (continuation probabilities). In another treatment variation, the payoffs from the programmed grim trigger strategy go to another human subject, who does not make decisions.

Design details

Randomization method
Players will be randomly assigned to treatments
Randomization unit
Individual
Clusters
1
Observations
120 subjects (or more, as budget allows).
Arms
Minimum 30 subjects per treatment, 4 treatments
Primary outcomes
Our outcome variables are cognitive measures, social values orientation measures, prior beliefs about how often the subject thinks it is a good idea to begin play of and indefinitely repeated game by choosing to cooperate, and actions choices in all periods of all such repeated games.

Cited by

No citation data available for this registration yet.

Keywords