Narratives, Belief Movements, and Economic Fluctuations

How to cite

Fries, Tilman; Qiaofeng. 2025. "Narratives, Belief Movements, and Economic Fluctuations." AEA RCT Registry. AEARCTR-0015607. https://doi.org/10.1257/rct.15607-1.0

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

Tilman Fries* , Qiaofeng

* Primary investigator

Registry ID
AEARCTR-0015607
DOI
https://doi.org/10.1257/rct.15607-1.0
First registered
Last updated
Status
In development
Trial dates
2025-03-24 – 2025-12-31
Countries
Private
JEL codes
C91, D83, D84, D91, G41
Registry record
http://www.socialscienceregistry.org/trials/15607

Abstract

We conduct a belief formation experiment with unobservable transitions in the latent state. Individuals observe a dataset of sequential signals, and the underlying state that generates these signals go through unobservable stochastic transitions (as in a "hidden Markov" environment). We test whether asking individuals to develop a narrative (a specific hypothesis of the state transition history) as an interpretation for each observed dataset can increase the extremeness and volatility of their beliefs by causing narrative developers to overweight best-fit narratives (interpretations). We also test whether the communication of developed narratives to others (the "narrative receivers") generates similar distortions in the beliefs of narrative receivers.

Experimental design

Participants take part in a belief updating task, where they see data on a hypothetical company and state their mean belief about the company's probability of success. In a Developer treatment, participants additionally provide a narrative before stating their mean belief. In a Receiver treatment, participants see data and a narrative when stating their mean belief.

Design details

Randomization method
Participants are randomized into treatment on the experimental session level. Further randomization of company data shown to participants is conducted in office by a computer.
Randomization unit
Randomization into treatment on the experimental session level. Company data assignment is randomly conducted on the individual level. Assignment of participants to companies is balanced across treatments.
Clusters
240 participants
Observations
6000 (25 per participant)
Arms
80 participants in Baseline, 80 participants in Developer, 80 participants in Receiver.
Primary outcomes
Extremeness of stated belief about a hypothetical company's probability of success, Volatility of stated belief about a hypothetical company's probability of success, Extremeness of narrative-implied estimate of a hypothetical company's probability of success, Volatility of narrative-implied estimate about a hypothetical company's probability of success.
Secondary outcomes
See the attached pdf.

Cited by

No citation data available for this registration yet.

Keywords