PURPOSE: Approximately 35-50% of individuals who start an exercise program have been reported to drop-out within the first few months (e.g., Dishman & Buckworth, 1997). Suggestions have been made that affective experience of exercise is linked to adherence, but only a limited amount of research has been conducted in this area. Also, none of these studies measured the influence of affective experiences during exercise on future choice. Identifying exercise that results in a positive affective experience may increase adherence. A theory that has found support in other fields is the Peak-End hypothesis (Kahneman et al., 1993). It suggests that individuals' positive affective experience and subsequent decisions about a behavior are derived from the peak affective moment and the final moments while engaged in the behavior. An episode with a pleasurable peak at the end will be remembered as positive even if it is longer than another episode with no pleasurable peak at the end (a concept known as Duration Neglect) Therefore, the purpose of this study was to determine whether adding a positive end to an unpleasant exercise bout will influence the choice to repeat an exercise bout and to test whether exercise duration plays a role in this choice. METHODS: 27 overweight and low-active young adults (14 males, mean age 26 yrs) completed an incremental treadmill exercise test to determine ventilatory threshold (VT). They then completed two counterbalanced exercise sessions: one at 10% above VT for 20 minutes and one at 10% above VT for 20 minutes followed by 2.5 mph, 0% grade for 5 minutes. Given that exercising above VT has been shown to elicit predominantly negative responses and exercising below VT elicits a positive affective response, these workloads were designed to result in a peak negative end affect and a peak positive end affect, respectively. Feeling Scale scores were measured before, during, and after exercise in both the sessions. In a final session, participants were asked to choose to repeat one of the two exercise bouts. RESULTS: Participants were twice as likely to choose to repeat the exercise bout that ended positively over the one that ended negatively, even though it was longer and involved more work overall. CONCLUSION: The results support the Peak-End and Duration neglect hypotheses in an exercise setting. To promote adherence, exercise prescriptions should put emphasis on a pleasurable (i.e. reduced intensity) end to each exercise bout.