Impulsive and Reflective Processes Related to Alcohol Use in Young Adolescents

Background: Dual process models suggest that the development of addictive behaviors is the result of interplay between impulsive and reflective processes, modulated by boundary conditions such as individual or situational factors. Empirical support for this model has been repeatedly demonstrated in adult samples [for a meta-analysis, see Ref. (1)]. The purpose of this study was to test these processes as they relate to emerging alcohol use in adolescents. Specifically, the interactive effects of several measures of impulsive and reflective processes and working memory capacity (WMC) are examined as predictors of changes in alcohol use among adolescents. It was expected that measures of reflective processes would better predict changes in alcohol use than measures of impulsive processes. Moreover, it was anticipated that WMC would moderate the relation between alcohol-specific impulsive and reflective processes and changes in adolescent alcohol use. Methods: The sample consisted of 427 adolescents (47.7% male) between 12 and 16 years of age (M = 13.96, SD = 0.78) who reported drinking alcohol at least once. Four measures of impulsive processes were included. Attentional bias for alcohol was assessed with a Visual Probe Test; approach bias toward alcohol was assessed with a Stimulus Response Compatibility (SRC) Test; and memory associations with alcohol were assessed with an Implicit Association Test (IAT) and a Word Association Test. Two measures of reflective measures were included: positive and negative expectancies. WMC was measured using a Self-Ordered Pointing Task. Results: Results showed that positive expectancies predicted changes in alcohol use, but this effect was qualified by an interaction with IAT scores. Moreover, SRC scores predicted changes in alcohol use only when negative expectancies were low. Attentional bias and word association scores did not predict changes in alcohol use. The relations between alcohol-specific processes or reflective processes and alcohol use were not moderated by WMC. Conclusion: Although there is empirical evidence for the validity of the model in predicting heavier alcohol use in adolescents, or alcohol abuse and dependence in adults, these observations do not generalize to a sample of normative, early adolescents. More specifically, results indicated that reflective processes are more important predictors of changes in alcohol use than impulsive process during adolescence.