Previous research has indicated that an intervention called "exam wrappers" can improve students' metacognition when they are using wrappers in more than one course per academic term. In this study, we tested if exam wrappers would improve students' metacognition and academic performance when used in only one course per academic term. A total of 86 students used either exam wrappers (an exercise with metacognitive instruction), sham wrappers (an exercise with no metacognitive instruction), or neither (control). We found no improvements on any of three exams, final grades, or metacognitive ability (measured with the Metacognitive Awareness Inventory, MAI) across conditions. All students showed an increase in MAI over the course of the semester, regardless of condition. We discuss the challenges of improving metacognitive skills and suggest ideas for additional metacognitive interventions.