Salience in Choice Under Risk: An Experimental Investigation

How to cite

Vitaku, Valon; Castillo, Marco. 2022. "Salience in Choice Under Risk: An Experimental Investigation." AEA RCT Registry. AEARCTR-0009744. https://doi.org/10.1257/rct.9744-1.0

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

Valon Vitaku* , Marco Castillo

* Primary investigator

Registry ID
AEARCTR-0009744
DOI
https://doi.org/10.1257/rct.9744-1.0
First registered
Last updated
Status
On going
Trial dates
2021-10-01 – 2023-05-31
Countries
Private
JEL codes
Criteria for Decision-Making under Risk and Uncertainty, Neuroeconomics
Registry record
http://www.socialscienceregistry.org/trials/9744

Abstract

We study choice under risk by presenting lotteries in a spinning wheel. A prize in a spinning wheel presents a state, and the outcome of the state depends on a binary lottery choice. We maintain marginal distributions of lotteries identical and change joint distributions. Our design allows us to isolate the effect of salience and we are able to rule out other theories of choice under risk for inconsistent behavior. We complement the experiment with eye-tracking measures which help us tease out the shape of salience. These measures allow us to grasp how states that ”stand out” guide choice under risk. We also present choices between lotteries in separate wheels which shed light on how an endogenous decision-making process may lead to a different tolerance towards risk. Our results have significant implications for the insurance market. Benevolent regulation surrounding the choice architecture could significantly improve consumer welfare.

Experimental design

We randomly assign subjects within a session to one of three treatments (uncorrelated, correlated joint, correlated separate). Subjects are faced with binary choice lotteries presented in a random order. Lotteries are presented in wheels. For the correlated treatments, we only change the presentation of lotteries but the underlying correlation is the same. Wheels spin in practice rounds (non-incentivized) to illustrate the nature of correlation if there is any and lottery outcomes are resolved. Following practice rounds, one of 20 incentivized choices (which are unresolved) is randomly selected at the end of the experiment to determine subject payout.

Design details

Randomization method
Randomization done by computer
Randomization unit
Individual
Clusters
300
Observations
300
Arms
100 subjects per treatment (uncorrelated, joint, separate)
Primary outcomes
Preference reversals, Risk-taking, Violations of First-Order Stochastic Dominance

Cited by

No citation data available for this registration yet.

Keywords