Understanding how responses become prepotent is essential for understanding when inhibitory control is needed in everyday behaviour. We investigated prepotency in the grass-snow task--in which a child points to a green card when the experimenter says "snow" and a white card when the experimenter says "grass". Experiment 1 (n = 548, mean age = 53.5 years) investigated the response method effect--whether pointing is prepotent because it is habitual. Experiment 2 (n = 560, mean age = 53.5 years) investigated the response set effect--whether responses are prepotent because the child plans to make them in the task. Experiment 2 also provided evidence that children could remember the rules in the task. Experiment 3 (n = 530, mean age = 53.4 years) produced further evidence that children could remember these rules. We found no evidence for the response method and response set effects, suggesting that prepotency in the grass-snow task is more "stimulus-driven" than in tasks previously studied. The implications of our findings are discussed in relation to other developmental inhibitory tasks and to children's reliance on inhibitory control to regulate their everyday behaviour. (Contains 4 tables and 1 note.)