Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/170969 
Erscheinungsjahr: 
2017
Schriftenreihe/Nr.: 
IZA Discussion Papers No. 10985
Verlag: 
Institute of Labor Economics (IZA), Bonn
Zusammenfassung: 
This paper examines the impact of a commonly experienced adverse cognitive state on decision making under uncertainty. Specifically, we administer an at-home sleep restriction protocol combined with random assignment to the time-of-day for decision making. Thus, we induce sleepiness in our subjects via sleep restriction as well as suboptimal time-of-day prior to administration of a Bayesian choice task. The specific task used discriminates between Bayesian choices that coincide with more simple reinforcement heuristic choices (in "Easy" trials) versus those that do not (in "Hard" trials), which is ideal given our underlying hypothesis that sleepy subjects are more likely to use simple heuristics. We first show that both circadian mismatch and sleep restriction significantly increase subjective sleepiness – this documents protocol validity. Our key behavioral results are that sleepy subjects are more likely to make a Bayesian inaccurate decision and more likely to make decisions consistent with a simple reinforcement heuristic, particularly in more cognitively difficult "Hard" trials. Secondary results show that stimulation of subject affect increased used of the simple decision heuristic but, when combined with sleep restriction, increased affect may increase task motivation and improve choice accuracy. These results offer new insights into the likely impact of sleepiness on decision making under uncertainty and highlight the potential negative impact on such cognitive states may have on accurate formation of probability assessments.
Schlagwörter: 
sleep restriction
sleep deprivation
reinforcement heuristic
Bayesian choice
experiments
JEL: 
C91
D81
D91
Dokumentart: 
Working Paper

Datei(en):
Datei
Größe
363.32 kB





Publikationen in EconStor sind urheberrechtlich geschützt.