The basic behavioral process of operant conditioning contributes to problem behaviors in psychological disorders. Escape from aversive situations in depression, the rewarding effects of drugs in substance abuse, and the receipt of caregiver attention for disruptive behavior in intellectual or developmental disabilities are just a few examples of operant reinforcement contingencies that perpetuate undesirable behavior. Behavioral treatment strategies often introduce alternative sources of reinforcement for a desirable alternative behavior. Although treatments can be effective, alternative reinforcement removal can trigger relapse of the problem behavior, called resurgence. Persistence in alternative reinforcement treatments and resurgence can be understood from the prospective of behavioral momentum theory, which predicts greater operant persistence and resurgence when there is a greater history of reinforcement associated with the context in which an operant response occurs. Shahan and Sweeney incorporated resurgence into the framework of behavioral momentum theory, and the proposed model makes explicit qualitative and quantitative predictions that are tested in this dissertation. Chapter 1 provides the background and significance of resurgence of operant behavior, and gives an introduction to behavioral momentum theory and the quantitative model of resurgence. Chapter 2 reports two recently published experiments that show increased time with alternative reinforcement treatment reduces subsequent resurgence in an animal model with pigeon subjects. The study presented in Chapter 3 examined how persistence and resurgence may be affected when alternative reinforcement is delivered in a novel context. This experiment, which used rat subjects, integrated and compared the animal model of resurgence with another operant relapse phenomenon, renewal, in which context change alone is known to induce relapse of a previously reduced response. Chapter 4 describes a study with college undergraduates as participants that tested the feasibility of a brief, three-alternative, forced-choice procedure as a human operant model of resurgence. Despite procedural manipulations of the length of training and probability of reward for choice of the target stimulus, resurgence was never consistently observed. Chapter 5 provides an integrative discussion of these research topics.