Symposia
Transdiagnostic
Leigh C. Brosof, M.S.
Student
University of Louisville
San Francisco, California
Cheri Levinson, Ph.D.
Associate Professor
University of Louisville
Louisville, Kentucky
Perfectionism is a transdiagnostic vulnerability factor for psychopathology, such as anxiety and eating disorder symptoms. One dimension of perfectionism related to maladaptive outcomes is concern over mistakes (CM), or the excessive worry over making errors. Understanding the mechanistic processes through which CM leads to psychopathological symptoms will aid in creating precision interventions. One pathway may include repetitive negative thinking (RNT), or maladaptive repetitive thoughts regarding the consequences of past or future behavior. Individuals high in CM both excessively worry about future mistakes, as well as ruminate on past mistakes (Egan et al., 2014; Handley et al., 2014). These RNT processes may spur an increase in negative affect (Ehring & Wakins, 2008) and maladaptive behaviors (e.g., binge eating). In the current study (N = 196 women), we experimentally manipulated CM (High [n = 61]), Low[n = 70], and Control [n = 65]) to examine the causal pathways between perfectionism, RNT, anxiety, and eating behavior. We expected that the high CM condition would predict RNT, which in turn would predict greater increases in anxiety and eating, compared to the low and control CM conditions.
Participants were randomly assigned to the High, Low, or Control CM condition. After random assignment, participants completed a computerized attentional task and then were left in a room for 10 minutes to eat snacks. We tested the effect of the High CM condition, as well as trait-by-CM condition interaction on RNT. Then, we tested whether RNT mediated the High CM condition and anxiety and food intake.
A one-way ANOVA showed a significant difference between conditions on change of CM due to the manipulation, F(2, 156)=10.82, p< .001, partial h2=.12. There were no significant main or interaction effects of Condition on RNT, ps ³ .205. There was a significant main effect of trait CM on RNT (b = .31, p = .005, partial r = .25). RNT did not mediate the relationship between anxiety and Condition (p = .531), or trait CM (p = .196), nor the relationship between food intake and Condition (p = .620) or trait CM (p = .345).
It may be that RNT is more related to trait CM, rather than changes of CM in the moment, as well as longer-term anxiety and eating behaviors. These findings may be due to the temporal nature of RNT, which often involves worrying about events in the future or ruminating on events in the past. Future research should continue to use experimental design to elucidate the temporal pathways through which CM and RNT lead to psychopathological symptoms.