Three experiments with single subject research designs were conducted to see whether habituation occurred as human participants completed physical activity. Experiment 1 had an arrangement where participant’s rates of stationary cycling were recorded throughout 20 minute sessions. Instead of habituation, stationary cycling rates increased over time, likely to be due to a rule that controlled behaviour derived from previous reinforced responding in the same setting as the experimental context. Experiment 2 was arranged similarly to Experiment 1 with an added concurrent task to disrupt the behaviour chain causing the rule-governed behaviour seen in Experiment 1. Despite this concurrent task, rates of stationary cycling again increased due to rule-governed behaviour. Experiment 3 had stationary cycling rates associated with a salient reinforcer that could lose its effectiveness to reinforce over time and thus habituate. Instead, Experiment 3 resulted in stable rates of stationary cycling consistent with task adherence, a rule-governed behaviour. With the failure of habituation to predict the results of this study, they are instead discussed from the more applicable perspective of motivating operations.