Non-drinking as deviant! First attempts to use the Relational Responding Task to alter drinking identity in an experimental study.

Heavy drinking remains a serious problem among young adults and studies demonstrate that drinking identity plays a role. We investigated the efficacy of a training in altering drinking identity and explored the role of drinking identity in the urge to drink and amount of alcohol consumed in a taste task. We hypothesized that participants in the abstainer+me condition would report less urge to drink and would drink less alcohol than in the drinker+me condition. We explored whether participants’ drinking identity interacted with training condition in its relation to urge to drink and alcohol consumed in the taste task.

Young adults (N=100, Mage=20.6, SD=2.7) were randomized to one of two (abstainer+me vs. drinker+me) training conditions–an adaptation of the Relational Responding Task (RRT). Pre- and post-assessments of implicit drinking identity, self-reported drinking identity, urge to drink, and observed alcohol use were measured. Mixed ANOVA and hierarchical regression analyses were used.

We found no interaction between time and condition related to drinking identity, F(1, 97)=1.26, p=.264 or urge to drink, F(1, 96)=2.41, p=.124. Consumed alcohol in the taste task did not differ between conditions, U=1078.00, p=.303. Exploratory analysis indicated that the interaction between training condition and identity at post-assessment significantly predicted consumption of alcohol in the taste task. Simple slope analyses indicated that in the ‘abstainer+me’ condition, post-identity was not significant (p = .96), but in the ‘drinker+me’ condition post-identity was significant (B=.33, p=.01).

Our hypotheses were not confirmed with no differences between training conditions in drinking identity, urge to drink, or amount of consumption in the taste task. We failed to alter drinking identity associations, but importantly drinking identity predicted urge to drink and observed alcohol consumed in the drinker+me condition. Therefore, drinking identity is important to include when designing and informing (indicated) prevention approaches.