Cognitive Ability, Rationality and Cooperation in Repeated Games

Last registered on October 01, 2025

View Trial History

Pre-Trial

Trial Information

General Information

Title

Cognitive Ability, Rationality and Cooperation in Repeated Games

RCT ID

AEARCTR-0016845

Initial registration date

September 23, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

October 01, 2025, 7:15 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

United Kingdom of Great Britain and Northern Ireland

Region

Primary Investigator

Name

Daniel Sgroi

Affiliation

University of Warwick

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Ali Moghaddasi Kelishomi

PI Affiliation

Loughborough University

Contact Investigator

PI Name

Andis Sofianos

PI Affiliation

Durham University

Contact Investigator

Additional Trial Information

Status

In development

Start date

2025-10-02

End date

2026-01-31

Keywords

Behavior

Additional Keywords

cooperation, rationality

JEL code(s)

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

This study aims to understand the extent to which people’s quality of decision making and their ability to engage in strategic interaction are related to cooperation in repeated games. Existing research indicates that the quality of decision-making, alongside cognitive ability, significantly influences crucial real-world outcomes, including wealth accumulation. While there is a growing literature on the influence of cognitive ability on strategic behaviour and cooperation in repeated games (including work by collaborators in this project) there is a notable research gap regarding the impact of decision-making quality on cooperation. We use adherence to GARP (the generalized axiom of revealed preference) as our benchmark of decision-making quality.

External Link(s)

Registration Citation

Citation

Moghaddasi Kelishomi, Ali, Daniel Sgroi and Andis Sofianos. 2025. "Cognitive Ability, Rationality and Cooperation in Repeated Games." AEA RCT Registry. October 01. https://doi.org/10.1257/rct.16845-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

The primary outcome of interest is the cooperation rate in an induced infinitely repeated Prisoner’s Dilemma (PD) game.

Intervention (Hidden)

Intervention Start Date

2025-10-02

Intervention End Date

2026-01-31

Primary Outcomes

Primary Outcomes (end points)

The primary outcome is the cooperation rate compared across different groups

Primary Outcomes (explanation)

Individuals are grouped based on their decision-making quality through a choice task suggested by Choi et al (AER, 2014). Specifically, decision-making quality is measured using the consistency of choices with the generalized axiom of revealed preference (GARP) model using the Afriat’s efficiency index following Polisson et al (AER, 2020). We then look at the rate at which individuals in each group select the cooperative outcome in a repeated prisoner’s dilemma, and compare this rate across groups.

Secondary Outcomes

Secondary Outcomes (end points)

We are also interested in describing the dynamics associated with the cooperation rate though there is no specific hypothesis attached to this.

Secondary Outcomes (explanation)

Cooperation dynamics follow the rate of change across the cooperation rate described above.

Experimental Design

Our experimental design comprises a two-part experiment conducted within the same session in a physical laboratory. In the first part, we use a task designed to measure adherence to the generalized axiom of revealed preference (GARP) which we use to group individuals into different experimental arms each of which corresponds to a different level of adherence to GARP. We then randomly match individuals within each arm to play a repeated prisoner's dilemma.

Experimental Design Details

Our hypothesis is that the level of cooperation will differ between groups indicating that adherence to GARP matters when determining the likelihood of cooperative play in the repeated prisoner's dilemma. We will test this vs the null that adherence to GARP is not relevant.

Randomization Method

O-Tree randomly matches players to play the repeated prisoner’s dilemma within each group.

Randomization Unit

Randomized at the individual level.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

256 individuals.

Sample size: planned number of observations

256 individuals (each will play the stage game up to 92 times).

Sample size (or number of clusters) by treatment arms

256 individuals split into 4 groups using F-GARP scores, so 64 per arm.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Pairwise power based on means and sds taken from a pilot: alpha = 0.0500 power = 0.9000 delta = -25.0000 mean1 = 87.5000 mean2 = 62.5000 sd1 = 35.3553 sd2 = 51.7549 This gives us a suggested N of 55 per arm. Our target of 64 gives us a safety buffer of 15%.

Supporting Documents and Materials

IRB