Grasp all, lose all: information overload under ambiguity

How to cite

Santis, Gianmarco De. 2025. "Grasp all, lose all: information overload under ambiguity." AEA RCT Registry. AEARCTR-0015591. https://doi.org/10.1257/rct.15591-1.0

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

Gianmarco De Santis*

* Primary investigator

Registry ID
AEARCTR-0015591
DOI
https://doi.org/10.1257/rct.15591-1.0
First registered
Last updated
Status
In development
Trial dates
2025-04-07 – 2025-10-05
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/15591

Abstract

Consumers often face decisions between ambiguous options, where the available information about potential outcomes can influence their choices. This study examines how the amount of information regarding outcome probabilities affects decision-making. We investigate whether increasing information reduces ambiguity and alters choice patterns. Through a laboratory experiment, participants explore different levels of probability information before selecting a lottery. Our findings contribute to a deeper understanding of how information availability shapes decision processes in uncertain environments.

Experimental design

The experiment is based on lotteries, divided into two types: transparent lotteries and opaque lotteries. There are up to five possible prizes, each associated with a probability of winning. In transparent lotteries, participants are informed of the probabilities (though these may include some margin of error). In opaque lotteries, the probabilities remain unknown to participants, who can only estimate them based on collected information (which may also be subject to error). For transparent lotteries, errors stem from inaccuracies in the probabilities displayed to participants. For opaque lotteries, errors arise from the information participants gather through sampling. Participants are tasked with choosing between pairs of lotteries, which can consist of:
- Two transparent lotteries;
- Two opaque lotteries;
- One transparent lottery and one opaque lottery.
When selecting between opaque lotteries, participants do not initially know the probabilities associated with the prizes. However, they can draw samples freely to gather information before making their choice.

Design details

Randomization method
The order in which the lotteries are presented to participants is randomized by a computer, ensuring that each participant encounters a unique sequence of lottery pairs.
Randomization unit
Individual unit
Clusters
0
Observations
At least 150 university students
Arms
At least 150 university students
Primary outcomes
choice between lotteries (risk and ambiguity preferences); measure of overload of information

Cited by

No citation data available for this registration yet.

Keywords