The current study was designed to examine the effect of metacognitive instruction strategies on the development of metacognition in grade 8 students and study variation in the development of metacognition in students with varying academic abilities. The study involved 80 grade 8 students from a public high school as a sample. A quasi-experimental non-equivalent control group research design was used to conduct research. The Junior Metacognitive Inventory Jr. MAI, developed by Sperling et al. (2002), was administered to assess metacognition in both experimental and control groups four times throughout the experiment, at baseline, and after each of the three stages. The experimental group was taught mathematical problem-solving using metacognitive teaching strategies (such as self-questioning, thinking aloud, modelling, and concept mapping). On the contrary, the control group received training using the traditional lecture method during the 18-week intervention. The results show that metacognitive instructional strategies used during the experiment significantly affect the growing metacognition, knowledge of cognition, and regulation of cognition of grade 8 students during mathematical problem-solving teaching. Furthermore, the metacognition of all subgroups was significantly enhanced after teaching through metacognitive instructional strategies. Mathematics teachers are recommended to spread awareness about metacognition and metacognitive instructional design to improve math problem-solving skills among elementary-level students.