Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group


First Advisor

Angela L. Duckworth


Why do people act self-controlled in some situations but not others? More specifically, why does it appear that an individual can be self-controlled in one domain (e.g., work) but impulsive in another (e.g., food)? This investigation tests a model that incorporates and explains both domain-specific and domain-general differences in impulsive behavior. Specifically, the model predicts that within-individual variation across domains is explained by subjective domain-specific appraisals of temptation and perceived harm, whereas domain-general impulsivity is explained by domain-general self-control strategies (e.g., pre-commitment) and resources (e.g., working memory). In Chapter 1, four studies test this model in adults. Studies 1 and 2 present the development and validation of a self-report questionnaire assessing impulsive behavior in six domains: work, interpersonal relationships, drugs, food, exercise, and finances. In Study 3, domain-specific appraisals of temptation and perceived harm are shown to explain within-individual variance in impulsive behavior, whereas domain-general self-control explains variance in domain-general impulsive behavior between individuals. Study 4 confirms that individuals in special interest groups (e.g., procrastinators) who are especially tempted in the target domain (e.g., work) are not likely to be more tempted in unrelated domains (e.g., food). Chapter 2 explores domain-specificity through the temporal discounting paradigm. Whereas self-report measures of impulsivity are sensitive to social desirability biases, choices in temporal discounting (sooner-smaller vs. later-larger rewards) are not as transparent. As predicted, temporal discounting is domain-specific, and domain-specificity in temptation partially explains domain-specificity in temporal discounting. Chapter 3 presents the development and validation of a domain-specific measure for children, motivated by the idea that some of the domains relevant for adults may not be relevant for children given that the average child presumably is either not attracted to certain temptations, does not perceive them as harmful, or does not frequently encounter them. For children, interpersonal and schoolwork impulsivity are shown to be correlated but distinct behavioral tendencies, demonstrating differentiated relationships with dimensions of childhood temperament, Big Five personality factors, and school outcomes. Collectively, these findings highlight the utility of a domain-specific approach, namely in terms of understanding psychological processes, improved prediction, and targeted interventions.

Included in

Psychology Commons