The purposes of this study were (1) to testity the truth of the viewpoints that "metacomponents are the important components of intelligence", and (2) to examine the effects of the training of cognitive strategies on the imcrement of intelligence scores. Study 1 examined the truth of the viewpoints that "metaconponents are the important components of intelligence." Forty graduate students and 70 undergraduate freshmen were served as subjects and the differences of their metacomponents were compared. The results showed that there were significant differences between the graduate students and freshmen in scores of metacomponents. There were also significant differences between them in their thinking process modes. While the graduate students tended to adopt top-down thinking process in problem solving, the freshmen tended to adopt bottom-up thinking process. The results revealed that both the "scholastic experience" and the modes of "thinking process" may influence metacomponents. In study 2, 105 undergraduate freshmen were arranged to participate one of three cognitive strategies training, the results of MANOVA showed that the scores of internal intelligence (especially on metacomponents) of two trained groups were significantly higher than those of control group. This implied that it is possible to increase intelligence scores through instruction of cognitive strategies.