Cheap talk and honesty - Experiment 2

How to cite

Fries, Tilman; Parra, Daniel. 2026. "Cheap talk and honesty - Experiment 2." AEA RCT Registry. AEARCTR-0018299. https://doi.org/10.1257/rct.18299-1.1

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

Tilman Fries* , Daniel Parra

* Primary investigator

Registry ID
AEARCTR-0018299
DOI
https://doi.org/10.1257/rct.18299-1.1
First registered
Last updated
Status
In development
Trial dates
2026-04-12 – 2026-12-31
Countries
Private
JEL codes
C91, D83
Registry record
http://www.socialscienceregistry.org/trials/18299

Abstract

This is a follow-up to our initial preregistration with identifier AEARCTR-0015630. We propose an experiment on cheap talk. A receiver guesses a secret number, and receives advice from a sender with misaligned incentives. In our initial experiment we had three treatment conditions that varied whether the sender can provide informed advice and how cognitively demanding the receiver finds it to respond to the advice in a sophisticated way. We observed that receivers reacted heterogeneously to informed advice. In this follow-up, we investigate mechanisms behind the observed heterogeneous effects.

Intervention

Basic setup: A receiver guesses a secret number after receiving advice from a sender with misaligned incentives.

Expert-Rep: Replication of original Expert treatment using new receivers while recycling sender messages from the original experiment.

Belief: A treatment where instead of guessing a secret number, the receiver guesses the probability that the sender decided to send the true secret number.

FixBeliefs: A treatment where, before making a guess, the receiver observes the percentage probability that the sender decided to send the true secret number.

NoUncertainty: A treatment that takes uncertainty out of the equation by informing receivers exactly how many senders decided to send the true secret number.

Experimental design

Senders and receivers are told that the computer randomly generated a secret number j by drawing once from {1,2,3,4,5,6,7}. The sender then chooses a number X in {1,2,3,4,5,6,7} to send to the receiver. The receiver observes the message and guesses a value for j. The sender is incentivized to induce a high guess, while the receiver has incentives to be accurate. The treatments vary the context in which the receiver makes the guess.

Design details

Randomization method
Randomization into treatments by random assignment of participants to sessions, further randomization (matching receivers to past senders) implemented by the experimental software (otree).
Randomization unit
Treatments are randomized on the session level.
Clusters
We will have 305 receivers playing for 24 rounds. Each receiver is one cluster.
Observations
305 participants
Arms
We will collect 70 receivers in Expert-Rep, Belief, and NoUncertainty and 95 receivers in FixedBelief.
Primary outcomes
Receiver's expected root mean squared guessing error (E-RMSE) Naive guess indicator.
Secondary outcomes
See the attached pdf.

Cited by

No citation data available for this registration yet.

Keywords