In-Group vs. Out-Group Communication

Last registered on December 26, 2025
View Trial History

Pre-Trial

Trial Information

General Information

Title
In-Group vs. Out-Group Communication
RCT ID
AEARCTR-0017442
Initial registration date
December 10, 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
December 26, 2025, 1:52 AM EST

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

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Jeanne Hagenbach
Affiliation
CNRS - Sciences Po
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Charlotte Saucet
PI Affiliation
University Paris 1 - Panthéon Sorbonne
Contact Investigator
PI Name
Chloe Tergiman
PI Affiliation
Pennsylvania State University
Contact Investigator

Additional Trial Information

Status
In development
Start date
2025-12-11
End date
2026-12-11
Keywords
Behavior
Additional Keywords
Strategic Communication, Hard Evidence, Social Groups
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
We examine situations of strategic communication between two individuals who belong or do not belong to the same group. Our objective is to study how the level of information transmitted from Senders to Receivers depends on whether communication takes place in-group or across different groups. We are particularly interested in examining how a given message is interpreted when coming from an in-group or an out-group member.

We consider a specific communication game with clear, well-known, theoretical predictions: the Sender (she) communicates hard information about a binary state, aiming to convince the Receiver (he) that the state is high. The Receiver, on the other hand, seeks to accurately guess the true state. A key feature of our setup is that we allow for the possibility that the Sender may not be informed of the state in the first place, a case in which the only message she can send is an empty one (Dye, 1985).

The probability that the Sender is uninformed is common knowledge and, in theory, shapes how Receivers interpret empty messages: they may result from strategic intent to hide unfavorable news or from mechanical constraints. We investigate how Receivers weight these two possible interpretations of an empty message, and whether these weights depend on group affiliations. In essence, we ask if Receivers are more inclined to give the benefit of the doubt to in-group Senders than to out-group Senders.
External Link(s)

Registration Citation

Citation
Hagenbach, Jeanne, Charlotte Saucet and Chloe Tergiman. 2025. "In-Group vs. Out-Group Communication ." AEA RCT Registry. December 26. https://doi.org/10.1257/rct.17442-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
The experiment, run in the lab, focuses on strategic information disclosure between a Sender and a Receiver. We vary two dimensions of the communication interaction :
(1) whether Senders and Receivers belong or not to the same group. The groups are built within the lab before the communication starts.
(2) the probability with which the Sender is initially informed of the payoff-relevant state. This probability When this probability equals one, we are in the standard disclosure game. When the probability is lower than one, we are in the framework proposed by Dye in 1985.
The design is 2 by 2 and the four treatments are implemented between subjects.
Intervention Start Date
2025-12-11
Intervention End Date
2026-12-11

Primary Outcomes

Primary Outcomes (end points)
The key outcome variables are the messages sent by the Senders and the guess made by the Receivers about the payoff-relevant state. We are particularly interested in the reaction of the Receivers when they receive no information from Senders.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The experiment has two main parts: first, participants engage in a collective word game designed to create group identities; second, they play a series of Sender-Receiver communication games.

- In Part 1, participants are randomly assigned in two teams of four people. They play four rounds of a word game with their team members. Precisely, each team member receives two letters per round, and teams must collectively form the longest possible word through intra-team discussion in a chat. Payments for Part 1 ensure that subjects try to find valid words, and also words that are longer than the ones found by the other team.

- Part 2 consists of six rounds of a Sender-Receiver communication game in which players, assigned permanently as either Senders or Receivers, are matched in pairs according to treatment: either with partners from the same team (InGroup) or from the other team (OutGroup).

Each round of communication game goes as follows: (1) a card color is randomly drawn, red or green; (2) the Sender is informed or uninformed of the color according to a probability that depends on the treatment; (3) the Sender sends either a precise color message or an empty message to the Receiver; (4) the Receiver observes this message and reports the probability that the card is green.

In the communication game, payoffs are standard: Senders earn payoffs increasing in the Receiver’s belief that the card is green. Receivers are paid according to a binarized scoring rule that rewards truthful reporting of their beliefs about the probability that the card is green.

More details about the design are given in the attached pre-analysis plan.

Experimental Design Details
Not available
Randomization Method
There are four treatments implemented between subjects. Every experimental session will implement one of the four treatments. We will randomly decide the order in which we run the treatments.
Randomization Unit
Randomization happens at the experimental sessions level.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
We will recruit as many subjects we need to make sure we have 100 cases in each treatment in which the Receiver sees an empty message from the Sender.
Sample size: planned number of observations
In each treatment, we plan to have 100 observations for which the Receiver has seen no information (an empty message) from the Sender.
Sample size (or number of clusters) by treatment arms
In each treatment, we plan to have 100 observations for which the Receiver has seen no information from the Sender.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Paris School of Economics
IRB Approval Date
2025-04-03
IRB Approval Number
2025-020
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information