Assessment and feedback play an instrumental role in an individual's learning process. Continued assistance is required to help students learn better and faster. This need is especially prominent in engineering laboratories where students must perform a wide range of tasks using different machines. One approach to understanding how students feel towards using certain machines is to assess their affective states while they use these machines. Affective state can be defined as the state of feeling an emotion. The authors of this work hypothesize that there is a correlation between students' perceived affective states and task complexity. By adopting the Wood's complexity model, the authors propose to assess how the correlations of perceived affective states of students change while they perform tasks of different complexity. In this study, each student performs a "hard" and an "easy" task on the same machine. Each student is given the same tasks using the same materials. Knowledge gained from testing this hypothesis will provide a fundamental understanding of the tasks that negatively impact students' affective states and risk them potentially dropping out of STEM tracks, and the tasks that positively impact students' affective states and encourage them to engage in more STEM-related activities. A case study involving 22 students using a power saw machine is conducted. Perceived affective states and completion time were collected. It was found that task complexity has an effect on subjects' affective states. In addition, we observed some weak correlation between some of the perceived affective states and laboratory task performance. The distribution of correlation between affective states may change as the tasks change. With the knowledge of the relationship between task complexity and affective states, there is the potential to predict students' affective states before starting a given engineering task.