Interactions between self-reported alcohol outcome expectancies and cognitive functioning in the prediction of alcohol use and associated problems: A further examination

A recent debate regarding the theoretical distinction between explicit and implicit cognitive processes relevant to alcohol-related behaviors was strongly shaped by empirical findings from dual-process models (Moss & Albery, 2009; Wiers & Stacy, 2010; Moss & Albery, 2010). Specifically, as part of a broader discussion, Wiers & Stacy (2010) contended that alcohol-related behaviors are better predicted by self-reported alcohol expectancies for individuals with good executive control and verbal abilities relative to those without such abilities. The purpose of the current paper is to further test whether self-reported alcohol outcome expectancies are moderated by measures of cognitive functioning. Using multiple indices of alcohol use, alcohol-related consequences, self-reported alcohol outcome expectancies, and cognitive functioning, both cross-sectional and longitudinal analyses were conducted in a prospective sample of 489 individuals at varying risk for alcohol use disorders. Results from a series of regression analyses testing interactions between self-reported alcohol expectancies and cognitive functioning showed minimal support for the hypothesized pattern discussed by Wiers and Stacy, 2010 regarding self-reported alcohol outcome expectancies. The overall rates of significance were consistent with Type I error rates and a substantial proportion of the significant interactions were inconsistent with previous findings. Thus, the conclusion that cognitive measures consistently moderate the relation between self-reported alcohol expectancies and alcohol use and outcomes should be tempered.