Finally, to examine the substance-related factors driving the alpha effects, we ran several additional tests. First, we ran a bivariate spearman correlation of alpha oscillatory power with number of AUD symptoms. This showed a significant correlation such that a greater number of AUD symptoms was associated with a weaker alpha response (r(42) = 0.32, p = 0.034). To determine whether this relationship was driven by the amount of substance use, we next performed multiple regression with AUD and CUD symptoms, and AUDIT/CUDIT-C scores as predictors on alpha activity. This showed that the number of AUD symptoms remained the only significant predictor above and beyond the other variables (t(39) = 5.53; p = 0.024; Fig. 3).
