Communication is key! Communication and partnerships in the prisoner’s dilemma game

How to cite

Serdarevic, Nina; Tjøtta, Sigve. 2018. "Communication is key! Communication and partnerships in the prisoner’s dilemma game." AEA RCT Registry. AEARCTR-0002735. https://doi.org/10.1257/rct.2735-4.0

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

Nina Serdarevic* , Sigve Tjøtta

* Primary investigator

Registry ID
AEARCTR-0002735
DOI
https://doi.org/10.1257/rct.2735-4.0
First registered
Last updated
Status
In development
Trial dates
2018-02-14 – 2018-02-15
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/2735

Abstract

Communication and language is ubiquitous in the evolution of cooperation. However, the power of language is frequently expressed in terms of solving complex problems that arise among people who need to coordinate their behavior to solve them (Wilson and Harris, 2017; Bickerton, 2009; 2014). If communication is to be regarded as a key ingredient for the achievement of a common good, should we not expect its merits to translate to the least complex dilemmas that often characterize everyday situations? Numerous problems are simple to solve coordination wise – one often needs to persuade only one other person to cooperate and not necessarily many. But, even in these seemingly simple situations the other person needs to choose to behave appropriately. The present paper examines if and how language is used to foster cooperation between two people in a repeated prisoner’s dilemma game with mutual partner choice

Intervention

The present paper examines if and how language is used to foster cooperation between two people in a repeated prisoner’s dilemma game with mutual partner choice. We abstract from the coercive nature of the standard prisoner’s dilemma game and rather focus on mutual consent as a way to form partnerships. Subjects are given the option to mutually decide if they want to enter a partnership with another subject or not. If both subjects choose one another, they proceed to a production stage. If one or both subjects abstain from actively choosing one another, each is left with no partner and no earnings in the current round.

Experimental design

First, we elicit cooperative types in a one-shot prisoner’s dilemma game. We also measure subjects’ personality traits with the Ten Item Personality Inventory (hereafter TIPI, Gosling, 2003). Second, subjects play a 10 period-repeated prisoner’s dilemma game within pairs of two subjects with either communication or no communication. Third, we re-elicit cooperative types to examine to which degree communication affects the willingness to change one’s cooperative strategy. The experiment will be programmed using z-Tree (Fischbacher, 2007).

Design details

Randomization method
Within session randomization using an urn.
Randomization unit
Individuals are randomized into either treatment or control within each session. Clustering is expected to be on the group level (partnership).
Clusters
108 clusters (partnerships) in total, 54 clusters in each treatment.
Observations
108 clusters in total
Arms
216 individuals in total, 108 individuals in each treatment
Primary outcomes
Contributions, Earnings, frequency of exclusion, matching, language and content used in the chat, behavioral types
Secondary outcomes
Contributions given partner's contributions, correlation between personality traits and cooperative types, frequency of one-sided partner choice

Cited by

No citation data available for this registration yet.

Keywords