Implicit affective associations predict snack choice for those with low, but not high levels of eating disorder symptomatology

The central purpose of the current study was to examine whether eating disorder symptomatology moderates the extent to which implicitly and explicitly measured affective associations with fruits, or the feelings individuals associate with them, predict food choice. Participants (N = 107) completed both implicit and explicit measures of affective associations with fruits, a self-report of eating disorder symptomatology. In a subsequent snack selection task, they selected either a granola bar or fruit as a measure of their food choice behavior. Logistic regression analyses revealed that eating disorder symptomatology moderated the relation of implicit affective associations on behavior, OR= 0.27, p =.024, 95% CI (0.085, 0.84). A test of the simple effects indicated that implicit affective associations predicted snack choice at or below mean symptomatology levels [OR= 2.073, p =.02, 95% CI (1.12, 3.84)], but not at high levels [OR= 0.56, p =.28, 95% Cl (0.18, 1.74)]. The effects of explicit measures on eating behavior were not moderated by eating disorder symptomatology (ps >.05). These findings advance our understanding of both the relation of implicit affective associations on behavior, as well as the ways in which eating disorder symptomatology may impact the decision-making process, thereby perpetuating disorder-related eating behavior. As such, they have implications for both health decision-making theory, as well as for the study and treatment of eating disorders. (C) 2014 Elsevier Ltd. All rights reserved.