Despite the common belief among practitioners that a happy worker is a productive worker, researchers have been struggling to understand the causality between satisfaction and performance for decades. This study attempts to bring clarity to current understanding through an experiment with repeated measures of satisfaction and performance. A total of 143 participants repeatedly performed a task based on the Stroop test, with their objective performance and task satisfaction measured each time. Two different types of feedback (high/low performance) were randomly assigned to participants in order to manipulate perceived performance. The data were analyzed using a path analysis. The results support the hypothesized influence of task satisfaction on task performance and of perceived task performance on task satisfaction.