Emotion-dependent Loss Aversion

Last registered on April 02, 2024
View Trial History

Pre-Trial

Trial Information

General Information

Title
Emotion-dependent Loss Aversion
RCT ID
AEARCTR-0013274
Initial registration date
March 30, 2024

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
April 02, 2024, 11:22 AM EDT

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

Locations

Country
United States of America
Region
Arizona

Primary Investigator

Name
Yiwei Qu
Affiliation
University of Arizona
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Clément Staner
PI Affiliation
Université Libre de Bruxelles
Contact Investigator

Additional Trial Information

Status
On going
Start date
2024-03-26
End date
2024-04-30
Keywords
Behavior, Lab
Additional Keywords
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
How do past failures shape future decisions? This article experimentally investigates the dynamic role of negative emotions on investment behaviour. To do this, we first augment traditional reference-dependent models to integrate the effect of past emotions on utility. Emotions stemming from past failures linger and affect utility negatively, but success brings emotional relief. Using these portable principles we derive testable implications without imposing functional form assumptions on the reference-dependent utility function.
External Link(s)

Registration Citation

Citation
Qu, Yiwei and Clément Staner. 2024. "Emotion-dependent Loss Aversion." AEA RCT Registry. April 02. https://doi.org/10.1257/rct.13274-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2024-03-26
Intervention End Date
2024-04-30

Primary Outcomes

Primary Outcomes (end points)
Within the same decision task: the difference between subjects' choices of effort levels when they are frustrated and when they are not.

Between different tasks: the difference in the difference above.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We employ a within-subject design. Subjects first face six different decision tasks. In each task, there is a list of gambles. Subjects need to exert effort to acquire the gamble. In general, the more effort they exert, the higher the expected value would be from the gamble. The tasks differ in whether the effort increases the chance of winning or the return of winning. Subjects choose one gamble from each decision task.

Then, subjects do a short filler task (be distracted), after which they receive either a high payment or a low payment. We consider the subjects who receive a low payment to be frustrated.

Last, subjects are presented with the same six decision tasks again. They can make choices the same as or different from their previous choices, depending on themselves. In total, they choose 12 gambles. We draw one gamble for them to work on.
Experimental Design Details
Not available
Randomization Method
We do within-subject design and only have one treatment. Hence, there is no need to randomize subjects into different treatments. However, we do have some lottery outcomes during the study. These randomizations are done by the Z-tree computer program.
Randomization Unit
Individual.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
Depending on how emotion affects people's utility (whether he is frustration-aversion, and whether he is prudent), subjects can be divided into 4 types. For each type, the amount of frustrated subjects is 28. On average half of the subjects will be frustrated. Therefore, for each type, the expected sample size is 56.
Sample size: planned number of observations
For all 4 types to have enough power, we need 56*4=224 observations.
Sample size (or number of clusters) by treatment arms
For each type, the expected sample size is 56. For all 4 types to have enough power, we need 56*4=224 observations.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
For each type, with 56 samples, effect size 0.5, and significant level 0.05, we can achieve power 0.805. Notice that we do now know ex-ante how many percentages of subjects will fall in each type. Therefore, the above power can be achieved only after we categorize people into their corresponding types.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
University of Arizona IRB
IRB Approval Date
2024-01-23
IRB Approval Number
STUDY00003924
Analysis Plan