The role of payoff parameters for cooperation in the one-shot prisoner’s dilemma
How to cite
Gaechter, Simon; Lee, Kyeongtae; Sefton, Martin; Weber, Till O.. 2022. "The role of payoff parameters for cooperation in the one-shot prisoner’s dilemma." AEA RCT Registry. AEARCTR-0009784. https://doi.org/10.1257/rct.9784-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
This study investigates the effect of variation in material payoffs on cooperation in experimental one-shot Prisoner’s Dilemma (PD) games. Following Mengel (2018) we vary three payoff indices. Indices of risk and temptation capture the unilateral incentives to defect against defectors and co-operators respectively, while an index of efficiency captures the gains from cooperation. In a previous analysis, we found that risk has no effect on cooperation, temptation reduces cooperation, and efficiency increases cooperation. Here we provide a confirmatory analysis of these results using a between-subjects design that varies payoff parameters in an orthogonal way and where participants will play one one-shot game.
Experimental design
Pairs of participants play one one-shot Prisoner’s Dilemma game. We have a between-subjects design with 8 prisoner's dilemma games. We vary the PD’s payoff matrix across the eight games as shown below:
G1: T=600; R=500; P=200; S=90
G2: T=600; R=500; P=200; S=20
G3: T=800; R=500; P=200; S=90
G4: T=800; R=500; P=200; S=20
G5: T=600; R=500; P=400; S=180
G6: T=600; R=500; P=400; S=40
G7: T=800; R=500; P=400; S=180
G8: T=800; R=500; P=400; S=40
Notes: S=Temptation payoff; R=Reward payoff, P=Punishment payoff, S=Sucker payoff. Payoffs in experimental currency.
Participants will be randomly matched into pairs and to one of the eight games.
Participants will be asked to complete two tasks: (1) Choose whether to cooperate or defect in the Prisoner’s Dilemma game, and (2) indicate the likelihood that the other person will choose to cooperate. The choices will be neutrally labeled as A and B. The order of tasks and choices will be randomised.
After taking part in the prisoner’s dilemma game, participants will be asked to complete a socio-economic background questionnaire.
The experiment will be run on Amazon Mechanical Turk with 200 participants in each of the 8 games, that is, with a total of 1,600 participants. The experiment is programmed with the software LIONESS Lab (https://lioness-lab.org/).
Design details
Cited by
No citation data available for this registration yet.