Replicated a previous procedure developed by the authors for reducing driving and extended it to increase its potential for wide use. Two groups, reinforcement and self-recording, consisting of a total of 17 undergraduates, received a month-long experimental condition counterbalanced by 2 month-long baselines. As in the previous study, the 9 reinforcement group Ss received individual monetary prizes scaled in terms of percent reduction in miles driven. The systematic replication confirmed the previous result (20% reduction) and provided a more stringent test. Ss were more representative of the general population and there were more checks on possible rule violations. The use of a weekly FI reinforcement schedule contingency, as well as reinforcement per se (the 1st extension), increased short-term control. A 2nd extension was the sizable effect of a leader. Four Ss from the reinforcement group were from the 2nd author's class and 5 were from a nonparticipating instructor's class. When the leader variable was combined with self-recording of driving mileage, the 3rd extension, the self-recording group averaged a 10% reduction even though the monetary prizes were solely for record keeping. It is suggested that the components of the present study could be combined and implemented in an industrial or institutional setting where a leader naturally exists.