Does Cue Context Matter? Examining the Specificity of Cue-Related Activation of Positive and Negative Alcohol Expectancies Jeffrey D. Wardell and Jennifer P. Read State University of New York at Buffalo Consistent with the Encoding Specificity principle, positive alcohol expectancies may be activated by cues in drinking contexts because they are more closely associated with these cues in memory than are negative expectancies. However, there is little research examining the specificity of cue-induced alcohol expectancy activation. This study investigated the relative activation of positive and negative expectancies in response to positive and negative cue contexts. We also examined whether these effects were stronger for heavier and more problematic drinkers. College student drinkers were randomly assigned to listen to vignettes describing either positive or negative drinking scenarios (cue exposure). Participants also completed an implicit measure of alcohol expectancy activation (modified Stroop task) both before and after the cue exposure, as well as self-report measures of alcohol use and alcohol-related problems. We found that alcohol-related problems moderated the effects of cue condition on expectancy activation, such that specific activation of positive relative to negative expectancies in response to positive cues was observed only for drinkers with higher levels of alcohol problems. No differences in activation of positive versus negative expectancies were observed for more problematic drinkers in the negative cue condition or for less problematic drinkers in either cue condition. The results are partially consistent with the Encoding Specificity principle, showing that positive contextual cues can selectively activate positive alcohol expectancies for more problematic drinkers. Findings may have implications for interventions that target automatic expectancy processes, suggesting potential utility in targeting specific expectancies in specific contexts. Keywords: alcohol expectancies, alcohol problems, implicit, cue, Stroop Problematic alcohol use (i.e., heavy drinking that is associated with hazardous outcomes) is prevalent among young adults and can result in injury, social and occupational problems, and death (Wechsler & Nelson, 2008). Understanding why some individuals persist in prob- lematic drinking behavior despite negative consequences is an impor- tant challenge for alcohol researchers. From a basic learning perspec- tive, the negative outcomes of problematic drinking should punish this behavior, and individuals should learn to reduce their alcohol use as a result. However, this is often not the case. Indeed, some types of negative alcohol consequences even predict later increases in drinking behavior (Read, Wardell, & Bachrach, 2013). Research on alcohol expectancies has helped to shed light on this phenomenon. Alcohol expectancies are beliefs about the likely out- comes of drinking alcohol that are learned and stored in memory (Goldman, Del Boca, & Darkes, 1999). Heavy drinkers tend to have both greater positive expectancies (e.g., mood improvement) and greater negative expectancies (e.g., feeling ill) compared with light drinkers (McMahon, Jones, & O’Donnell, 1994), perhaps because they had more experience with a range of drinking outcomes. But, positive alcohol expectancies are stronger predictors of drinking be- havior than are negative expectancies (Leigh, 1989; Rohsenow, 1983; Stacy, Widaman, & Marlatt, 1990). Thus, positive expectancies may motivate heavy drinking even when an individual expects that nega- tive outcomes may follow. However, the processes through which positive alcohol expectan- cies come to dominate the prediction of drinking behavior are in need of further investigation. One important line of research in this regard is the study of automatic memory processes related to alcohol expec- tancies. Consistent with an associative network model of memory, experiences with alcohol may result in the formation of memory associations between alcohol and concepts representing outcomes of drinking alcohol (e.g., happy, sick). These associations are thought to 1 There is some debate regarding the appropriate use of the term “alcohol expectancy.” Some researchers believe this term should be used only to describe explicit, propositional “if. . .then” statements, whereas others be- lieve it can be applied to describe memory associations that are an under- lying component of explicit expectancies (Moss & Albery, 2010; Wiers & Stacy, 2010). To remain consistent with research that has used the same assessment approach as the present study, we use the term alcohol expec- tancy broadly to describe memory representations of commonly endorsed outcomes of drinking alcohol, which can be assessed using both explicit and implicit measures (Goldman & Darkes, 2004; Goldman et al., 2006). This article was published Online First October 7, 2013. Jeffrey D. Wardell and Jennifer P. Read, Department of Psychology, State University of New York at Buffalo. We thank Drs. Mark Goldman and Jack Darkes for making their alcohol expectancy materials available for us to use in this study. We also thank Jenna Dellacio, Joshua Helak, Alysia O’Rourke, and Serena Wong for their assistance with data collection and preparation of study materials, as well as Drs. Stephen Tiffany, Craig Colder, and James Sawusch for their contributions to this study’s conceptualization. While conducting this study, the authors were supported in part by Grant R01AA016564 from the National Institute of Alcohol Abuse and Alcoholism awarded to Jennifer P. Read. Correspondence concerning this article should be addressed to Jeffrey D. Wardell, Department of Psychology, 206 Park Hall, State University of New York at Buffalo, Buffalo, NY 14260. E-mail: jwardell@buffalo.edu This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Experimental and Clinical Psychopharmacology © 2013 American Psychological Association 2013, Vol. 21, No. 6, 457– 466 1064-1297/13/$12.00 DOI: 10.1037/a0033967 457