Misperceptions and intergroup cooperation

How to cite

Gately, Pierce; Jiang-Wang, Ángela. 2026. "Misperceptions and intergroup cooperation." AEA RCT Registry. AEARCTR-0018155. https://doi.org/10.1257/rct.18155-1.0

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

Pierce Gately* , Ángela Jiang-Wang

* Primary investigator

Registry ID
AEARCTR-0018155
DOI
https://doi.org/10.1257/rct.18155-1.0
First registered
Last updated
Status
In development
Trial dates
2026-06-09 – 2027-09-30
Countries
Private
JEL codes
C92, D91
Registry record
http://www.socialscienceregistry.org/trials/18155

Abstract

How groups perceive each other can impede or facilitate intergroup cooperation, which is necessary to solve large scale collective action problems. If groups harbour negative feelings towards each other, it can foster an 'us' vs 'them' mentality which undermines societal progress by curtailing cooperation beyond group lines. A canonical finding in experimental literature is that most people are conditional cooperators, willing to cooperate if they believe others will do likewise. For such individuals’ beliefs about the cooperativeness of others are central to determining their own behaviour. In intergroup contexts perceived polarisation between groups can be self-reinforcing if individuals expect intergroup prejudicial attitudes to translate into discriminatory actions. We propose a series of studies to unpack the behavioural channels through which perceptions of polarisation between groups impact intergroup cooperation.

Experimental design

We are planning to conduct two studies:

Study 1. We will elicit participants' feelings toward President Donald Trump in order to classify them as Trump Lovers or Trump Haters following a method similar to that of Dimant (2024). Study 1 is composed of three treatments. In all treatments, in-group and out-group membership will be defined using the Trump Hate/Love paradigm. That is, all participants will play with a Trump Lover in one group and a Trump Hater in the other group in the S-PGG.

Anonymous treatment: No information will be given to participants regarding the identity of their partner in each group in the S-PGG.

Trump treatment: We will reveal to participants whether they will play with a Trump Lover or a Trump Hater in each group in the S-PGG. All participants will be randomly matched with a Trump Lover in one group and a Trump Hater in the other group.

Charity treatment: Same as Trump treatment; however, a fixed portion of the earnings generated from the group accounts will go toward a charity (Feeding America).

Study 2. If we detect misperceptions in Study 1, we will aim to correct them through information provision treatments in a follow-up study. We will have a control group where no information is provided, and information provision treatments where we will reveal the true personal norms, attitudes, and/or feelings thermometer ratings elicited in Study 1. We will aim to correct misperceptions in one domain per treatment. The final number of treatments will depend on whether we detect misperceptions in Study 1. Participants will then play another S-PGG mirroring the design of the Trump treatment in Study 1.

More details about the experimental design are disclosed in the attached materials.

Design details

Randomization method
Randomization will be done using the "randomizer" tool in Qualtrics
Randomization unit
Randomization will be conducted at the individual level.
Clusters
As each subject makes multiple decisions, we will cluster at the level of the individual.
Observations
We plan to invite a minimum of 2160 participants to take part in Study 1. Participants who take part in the Trump treatment will also be invited to take part in Study 2 (which will be launched approximately one week after Study 1).
Arms
We will randomize participants into treatments on Qualtrics. We aim to recruit approximately 72-74% of participants into our Trump treatment, and 13-14% of participants to each of our anonymous and charity treatments in Study 1. In Study 2 participants will be randomized across control and treatment arms in equal numbers (the exact number of treatment arms depends on the misperceptions we find in Study 1).
Primary outcomes
We have three kinds of primary outcomes in our study: 1. Differences in cooperation toward in-group and out-group members (out-group discrimination) 2. Cooperation between in-group and out-group members (intergroup cooperation) 3. (Mis)perceptions across three domains (affective, normative, and attitudinal) and two reference networks (in-group and out-group)
Secondary outcomes
Differences in attitudes and beliefs toward in-group and out-group members

Cited by

No citation data available for this registration yet.

Keywords