Burns, Rachel J2014-06-302014-06-302014-05https://hdl.handle.net/11299/163660University of Minnesota Ph.D. dissertation. May 2014. Major: Psychology. Advisor: Alexander John Rothman. 1 computer file (PDF); vii, 160 pages, appendices 1-8.Offering financial incentives that are contingent upon the performance of a specified health behavior has emerged as a popular intervention strategy. However, the types of incentives that are most effective in changing behavior and the mechanisms through which these effects occur have not been identified. This study tests a theoretical framework that specifies two dimensions along which incentives may vary (reinforcement procedure: positive, negative; schedule: fixed, variable). Negative reinforcement was expected to be more effective than positive reinforcement because people are loss averse. Variable schedules were expected to be more effective than fixed schedules because they buffer against habituation to the incentive. A 5-week randomized controlled trial randomly assigned 153 participants to one of the four conditions that emerge when reinforcement procedure and schedule are crossed or a no-incentive control condition. Incentives were contingent upon meeting a specified walking goal. A host of psychological variables, including the perceived value of the incentive, were measured throughout the study so that mediation could be tested. Rates of walking were greater in the incentive conditions compared to control; the incentives did not differentially affect behavior. The perceived value of the incentive predicted behavior, but was not affected by the type of incentive. Walking rates dropped markedly during a 2-week follow-up. Future work should seek to definitively determine if the incentive categories differ from each other and continue to test mediational models. This study underscores the value of systematically and directly comparing theoretically grounded types of incentives and testing putative mediators that underlie the effect of specific types of incentives on behavior.en-USThesis or Dissertation