Objective: To determine whether fitness-center attendance established with the provision of weekly monetary incentives persisted after the discontinuation, or decreased frequency, of incentives. Participants: One hundred seventeen first-year college students participated during the 2011-2012 academic year. Methods: A randomized controlled trial with control, discontinued-incentive, and continued-incentive conditions was conducted. During fall semester, students in incentive conditions received weekly monetary payments for meeting fitness-center attendance goals. During spring semester, discontinued-incentive condition participants no longer received incentives, whereas continued-incentive condition participants received payments on a variable-interval schedule. ID-card attendance records tracked fitness-center attendance. Results: Goal completion decreased from 63% in the incentive groups during the fall semester to 3% in the discontinued-incentive condition, and 39% in the continued-incentive condition during the spring semester. There was not a significant interaction between condition and body mass index change, F(6, 332) = 0.67, p = 0.68. Conclusion: Incentive discontinuation resulted in students no longer meeting fitness-center attendance goals. A variable-interval reward schedule better maintained attendance.