The purpose of this study was to determine if task, coping, and scheduling self-efficacy (SE) for exercise could be influenced using guided imagery interventions in an experimental design controlling for overt exercise experiences. Healthy women (N = 205, Mage = 31.47) who did not exercise regularly were randomly assigned to guided imagery sessions targeting a specific type of SE or to a nutrition information control group. All participants attended a 12-week cardiovascular exercise program. The influence of the various imagery interventions on the three types of self-efficacy over time were assessed with two doubly multivariate ANOVAs: one from baseline to 6 weeks and the other from 6 to 12 weeks. The analyses were conducted for 61 participants who completed the exercise intervention. The first analysis demonstrated that task, coping, and scheduling SE were differentially influenced over time in response to the different imagery interventions. The results of the second analysis were non-significant, revealing that the main changes in SE were observed within the first half of the 12-week intervention. This study demonstrates that task, coping, and scheduling SE can be seen as independent from one another and that mental imagery interventions are an effective means for influencing exercise-related cognitions.