The current study investigated differences in metacognition between high school gifted (n = 44) and typical (n = 23) students and examined local calibration accuracy as a potential mechanism for partially explaining superior exam performance by gifted students. Metacognition was measured using student self-reports of metacognitive awareness, acquired at the start of a semester-long course on biology, and students' global (pre- and posttest) and local (item-by-item) confidence judgments to assess monitoring across four biology exams over the course of one semester. Gifted students outperformed typical students on both local and global postdictive measures of calibration. However, there were no statistically significant differences in global predictive judgments or calibration bias. Local, item-by-item calibration accuracy partially mediated the relation between giftedness and exam performance. Implications for both theory and practice are discussed. (Contains 2 tables, 2 figures, and 1 note.)