Young adults who engage in heavy alcohol consumption are more likely to experience both current and future alcohol-related problems. Heavy episodic drinking is associated with risky behaviors, such as having multiple sex partners and a variety of negative academic, physical, and social consequences[2, 3]. Additionally, these drinking patterns predict concurrent and future alcohol use disorders. Brief interventions may lead to reductions in the drinking behavior of young adults, but these reductions appear to be modest. Examining variables that moderate intervention success may help to predict who will benefit from treatment. One potential class of moderators consists of mental processes that are automatic—motivational responses that are spontaneously activated in the presence of an alcohol cue (in contrast to deliberative consideration of whether to consume). The current study examined whether automatic alcohol-approach associations would act as a moderator of a brief alcohol intervention.
Brief motivational interventions (BMIs) represent a class of treatments designed to reduce heavy drinking, in part through increasing motivation and commitment to change. Although there are differences between specific BMIs, they typically involve one or two sessions and consist of elements such as individualized feedback of drinking behavior and negative alcohol-related consequences, emphasis on the responsibility of the individual to make any change, offering advice to change, offering options for making a change, enhancing the individual’s self-efficacy for change, and conducting the intervention in an empathic style. In BMIs, motivation to reduce drinking behavior is elicited through a number of strategies such as discussing discrepancies between the individual’s values (e.g., academic performance) and actual behavior (e.g., frequently neglecting homework because of drinking alcohol). A key objective is to enhance motivation and commitment to change by increasing the salience of negative alcohol-related consequences and highlighting their incongruence with personal standards and valued outcomes.
Early research indicated that BMIs may be a useful treatment for hazardous drinking among college students[9–12]. Recent reviews suggest that BMIs for alcohol and other substance use yield beneficial effects, but that these effects are modest in size. In an effort to identify those who may benefit from motivational interventions for alcohol use, investigators have begun to explore the role of a number of moderators. Regarding participant variables, there is mixed evidence for gender as a moderator of BMIs, with some studies supporting it and others finding nonsignificant results[9, 15]. Other potential moderators, such as self-regulation skills and tendency to engage in social comparison, have failed to predict who will benefit from BMIs.
There is growing support for the view that alcohol use may be influenced by two systems of psychological processes: (1) an automatic, non-volitional, impulsive system, and (2) a resource dependent, volitional, reflective system[16–18]. This model suggests that, for heavy drinkers, the typical response to an alcohol cue is an automatically activated disposition to approach and consume and that sufficient motivation to restrain and sufficient self-control resources are required to inhibit this automatic appetitive response. From this perspective, although BMIs may elicit motivation to change drinking, strong automatic alcohol-approach associations may impede the ability to translate this motivation into actual change. Thus, one class of moderators that may be particularly important to explore consists of implicit measures of automatic approach responses to alcohol. In contrast to explicit measures, which require participants to directly introspect and report on mental content (e.g., outcome expectancies), implicit measures are designed to assess automatic processes related to mental content by using indirect methods such as reaction time tasks. A number of implicit measures have been used to assess the role of automatic attentional biases and mental associations in substance use behavior. For example, attentional biases toward alcohol cues are related to heavier alcohol consumption[21–23] and have been shown to predict treatment success in individuals motivated to abstain or restrain alcohol and other drug use[25, 26].
One widely-used implicit measure of automatic affective associations with target stimuli (such as alcohol) is the Implicit Association Test (IAT). The IAT is a categorization task that allows inferences about the relative associative strength between two concepts through reaction time performance (see the Methods section for more procedural detail on the IAT). Research with young adult drinkers indicates that automatic alcohol-affect associations measured by the IAT demonstrate a medium effect-size relation with alcohol use, even when controlling for explicit measures of alcohol-affect associations[28–30]. Of particular importance for clinical research, the IAT has been shown to predict failure to control alcohol consumption[31, 32].
Theoretical accounts suggest that automatic mental processes can impede the ability to change addictive behaviors. The research reviewed above supports this theory by providing evidence that the strength of automatic appetitive responses (i.e., attentional biases and approach associations[31, 32]) to alcohol cues predict alcohol consumption despite motivation to restrain use. It follows that the effects of increasing motivation to change drinking behavior with BMIs may be moderated by automatic processes. Specifically, strong automatic alcohol-approach associations may impede the ability to translate motivation to change drinking into actual reductions of alcohol consumption.
The current study was designed to examine whether a measure of automatic alcohol-approach (relative to alcohol-avoid) associations would moderate the efficacy of a BMI in a sample of hazardous drinkers. It was hypothesized that automatic alcohol-approach associations would moderate intervention effects such that students exposed to a BMI would show greater reductions in amount of alcohol consumed per occasion if they had weaker automatic alcohol-approach (relative to alcohol-avoid) associations. We also examined whether automatic alcohol-approach associations would moderate the relation between individual differences in motivation to change drinking and subsequent drinking quantity. We predicted that, among those with high motivation to change, subsequent reductions in consumption would occur in those with weak alcohol-approach associations but not in those with strong alcohol-approach associations.