This study seeks to investigate whether temptation declines over time; in particular, whether short waiting periods influenced the preference for healthy food choices. This project was previously completed prior to my awareness of the AEA RCT, in early 2015.
The abstract for the working draft can be found below:
Does temptation decline over time? Recent studies have highlighted the importance of Pavlovian cues for intertemporal decision making but less is known about how these cues change over time. To theoretically motivate the existence of these Pavlovian responses, an evolutionary principal-agent model is developed in which optimal temptation declines over time. In a laboratory experiment, subjects make a choice between bananas and chocolate, but some are informed 3 minutes prior that the choice will soon arrive. These treated subjects had no way to indicate their preferences in advance, yet were 28% more likely to choose a banana. Using effort induced by a preceding experiment, I also find no evidence of a resource- or ego-depletion effect.
I randomly treated individuals with a 3 minute waiting period before they were allowed to indicate their preference for a banana or a piece of chocolate (Reese's Peanut Butter Cups).
Intervention Start Date
2015-03-01
Intervention End Date
2015-05-30
Primary Outcomes (end points)
Choice of Banana, Chocolate, or Neither
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Subjects were allowed to determine whether they preferred a banana or a piece of chocolate. The only difference was whether there was a waiting period prior to the decision. There was a preceding experiment that involved real effort tasks, and that information is used to investigate resource or ego-depletion models.
Experimental Design Details
Randomization Method
Computer randomization of treatment
Randomization Unit
Individual level (in the laboratory)
Was the treatment clustered?
No
Sample size: planned number of clusters
422 Subjects
Sample size: planned number of observations
422 Subjects
Sample size (or number of clusters) by treatment arms
Approximately 211 Subjects for each treatment (waiting or not)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)