Date of Award

January 2017

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Alan King


The present study tested two theoretical models of binge eating (i.e., escape theory and the affect regulation model of binge eating) by combining an experimental design with ecological momentary assessment. After undergoing a negative affect induction, 139 participants, all of whom were current binge eaters, were instructed to eat a pint of ice cream (binge eating condition) or to eat as much as they would during a typical snack (control condition). Participants assigned to the binge eating condition ate either with (binge eating + self-awareness condition) or without (binge eating condition) a visible video camera pointed in their direction. Participants rated their negative affect in the laboratory at four different time points and in their natural environments via Palm Pilot in the four following hours. A mixed measures analysis of variance and generalized linear mixed effects model were used to test escape theory and affect regulation model, respectively. When compared to immediately prior to eating, guilt did not decrease during eating, though having a higher BMI was associated with reductions in guilt during eating. Guilt increased among both the binge eating and binge eating + SA conditions immediately following eating when compared to during eating. Both the control and binge eating conditions experienced postprandial reductions in guilt over the course of the four hours following the laboratory eating episode. However, the binge eating condition experienced these reductions at a faster initial rate with more slowing over time when compared to the control group. Findings provide preliminary support for escape theory in the context of higher BMI and are consistent with affect regulation model, as more rapid decreases in guilt are thought to negatively reinforce binge eating and distinguish affect regulation following binge eating from the mood enhancing effects of normal eating episodes. Future studies should be conducted with a greater sample size and aim to enhance ecological validity to ensure eating in the laboratory represents binge eating episodes occurring outside of the laboratory.