The present study investigated incremental theorists' metacognitive monitoring and study behaviours during a word-pair learning task. Sixty-five participants (38 female; aged 18-64 years, M = 33.92, SD = 14.40) studied English-Spanish word-pairs of differing difficulty (easy, moderate and difficult) and made judgments of learning (JOLs). Participants selected half of the word-pairs for restudy, before completing an initial cued-recall test to demonstrate the accuracy of their JOLs and study decisions. They then completed a restudy phase, incorporating a within-in subjects honour/dishonour manipulation, to assess the optimality of their study decisions. A final cued-recall test was administered, followed by Dweck's (1999) Theories of Intelligence Scale. Endorsing more incremental views of intelligence was marginally associated with poorer calibration for difficult word-pairs, (r = .235, p = .062), and was negatively associated with metacognitive accuracy for easy word-pairs during the study choice phase (r = -.260, p = .036). No further impairments were evident for the other difficulty-levels of these statistics, nor on over-confidence, under-confidence or resolution. Participants' study choices reflected a Discrepancy Reduction approach, and in general, did not appear to be optimal. Future research is required to further clarify these findings in reference to a larger entity theorist comparison group.