This study was concerned with the mechanism(s) via which performance monitoring affects task performance. To investigate whether an individual's concerns over the evaluation of his or her performance underlie the effects of performance monitoring (independent of contingent consequences) on the distribution of effort across tasks, the study examined whether the way in which individuals performed two tasks was differentially affected by the way in which supervisors monitored performance on those tasks. Subjects working in a group setting performed two different tasks for 90 minutes, their instructions being to perform as well as possible on each task. When supervisors periodically monitored the work output of each individual within the group on one of the two tasks, subjects perceived that the supervisor was evaluating and comparing individual performance, and produced more on the monitored task at the expense of the unmonitored task. When, however, the supervisor's monitoring focused on group rather than individual products, subjects perceived that the supervisor was less likely to be able to evaluate and to compare individual performance, and their performance on the two tasks was the same, regardless of which task was monitored. Path analysis supported the interpretation that the effects of monitoring on performance were mediated by evaluation concerns associated with performance on each task.