Abstract
BACKGROUND: Attitudes toward driving after drinking are strongly predictive of drinking and driving behavior. This study tested working memory capacity (WMC) as a moderator of the association between attitudes and drinking and driving behavior. Consistent with dual process models of cognition, we hypothesized that the association between perceived danger and drinking and driving would be stronger for individuals with higher WMC. METHODS: Participants (N = 161) enrolled in larger alcohol administration study were randomly assigned to an alcohol (n = 57), placebol (n = 52), or control (n = 52, not included) beverage condition. Past-year frequency of driving after drinking and WMC were assessed at baseline. Attitudes were assessed by asking participants to rate the perceived danger of driving at their current level of intoxication twice on the ascending limb (AL1, AL2), at peak breath alcohol concentration (BrAC), and twice on the descending limb (DL1, DL2). RESULTS: Analyses across the BrAC curve indicated that the hypothesized interaction was observed for the alcohol but not placebo condition. Analyses for each assessment point indicated that the interaction was significant for the ascending limb and peak BrAC. In the alcohol condition, for those higher in WMC, lower perceived dangerousness was strongly associated with increased driving after drinking (AL1: incident rate ratios [IRR] = 5.87, Wald's χ(2) = 12.39, p = 0.006, 95% CI [2.19, 15.75]; AL2: IRR = 8.17, Wald's χ(2) = 11.39, p = 0.001, 95% CI [2.41, 27.66]; Peak: IRR = 5.11, Wald's χ(2) = 9.84, p = 0.002, 95% CI [1.84, 14.16]). Associations were not significant at low WMC. CONCLUSIONS: Results suggest that individuals higher in WMC are more likely to act consistently with their explicit attitudes toward drinking and driving. Findings may have implications for existing drinking and driving interventions and suggest the potential for novel interventions targeting implicit associations or WMC.