Testing the robustness of daily associations of affect with alcohol and cannabis use

Etiological models of alcohol and cannabis use disorders hypothesize that people are more likely to use substances when experiencing heightened negative affect, yet recent Ecological Momentary Assessment (EMA) studies found no evidence for this daily association. To provide a robust understanding of whether and when affect regulation is supported in EMA, we tested within-person associations between affect and substance use across hundreds of statistical models in a diverse sample of young adults (n = 496) recruited from both college and community sources, aged 18-22 years (55.8% assigned female sex at birth, 44.2% assigned male sex at birth; 47.2% cisgender female, 43.8% cisgender male, 12.9% nonbinary/genderqueer/gender nonconforming, 4.0% transgender; 69.6% non-Hispanic White, 26.2% Asian, 6.7% African American, 8.5% Hispanic/Latino). Using specification curve analyses, we examined how different affect operationalizations, time scales, and moderators influenced these associations. For alcohol use, higher negative affect predicted decreased likelihood of drinking (median odds ratio = .95, p < .001), with 20.6% of specifications reaching significance. This counterintuitive pattern was strongest for sadness and when examining maximum daily negative affect. Surprisingly, and contrary to theoretical predictions, this negative association was slightly more pronounced among those with higher coping motives and at lower levels of alcohol use disorder symptoms. Positive affect showed a complex pattern, with high-arousal states like joviality strongly predicting increased drinking likelihood, whereas low-arousal states showed weaker associations. Neither affect type consistently predicted drinking quantity. For cannabis use, neither positive nor negative affect predicted use likelihood or quantity across specifications. These associations remained consistent regardless of substance use disorder severity or social context. Our findings challenge core assumptions of affect regulation models and suggest that, at least in young adults, the affect-substance use relationship is more nuanced than previously theorized, with implications for refining etiological models. (PsycInfo Database Record (c) 2026 APA, all rights reserved).