Introduction: Most research assumes driving under the influence of cannabis (DUIC) is the outcome of deliberate decision-making, neglecting the possibility that many DUIC episodes may be unplanned. The Prototype Willingness Model (PWM) conceptualises two decision-making pathways: a 'reasoned action' pathway that is conscious, rational and goal directed and a 'social reaction' pathway that is semi-conscious and primed by social and situational cues. Purpose: This study investigated the applicability of both PWM pathways, and subjective norms and Perceived Behavioural Control (PBC) from the Theory of Planned Behaviour, to DUIC decision making among cannabis users and non-users. Method: Australian drivers (N = 357, M = 24.54 years, SD = 11.01 years; 74.2% female; 44% never used) completed a 20-minute online questionnaire. Three hierarchical multiple regressions were conducted. These analyses tested theoretically deduced hypotheses as well as exploratory relationships between the predictors and dependent variables on different PWM pathways (intentions/expectations, willingness), and the influence of injunctive norms and PBC. Step 1 included control variables (gender, general attitudes towards cannabis use, and cannabis user group); steps 2 and 3 added the PWM constructs depending upon the pathway being examined (perceived vulnerability, descriptive norms, prototype favourability and similarity) and injunctive norms; PBC was entered in step 4. Results: The regressions accounted for between 14.8% and 17.9% of the variance in intentions, expectations and willingness to DUIC. Consistent significant positive predictors in each regression were descriptive norms and injunctive norms; counterintuitively, overall attitudes was a consistent negative predictor. Against theoretical expectations, perceived vulnerability predicted willingness only, and prototypes were not significant predictors in any analysis. Conclusion: Descriptive and injunctive norms may be important targets for interventions to prevent DUIC. Further research is needed to confirm if more positive general attitudes towards cannabis are predictive of lowered DUIC intentions, expectations and willingness.