For the enhance IAT/COMT model, main effects of COMT and implicit enhancement were not significant predictors of drinking, with the exception of a non-significant trend of COMT predicting drinks per week, p = .096 (see Table 3). Moderation was supported for drinks per week only; the COMT × enhance IAT interaction was consistent with predictions (p = .023). Enhance IAT scores showed stronger associations with drinking among COMT Met carriers compared to COMT Val/Val individuals (see Figure 1).
