The study aimed at identifying the role of metacognitive experiences (ME), such as feeling of difficulty, estimate of solution correctness and estimate of effort exerted, in the formation of one's self-concept in the mathematics domain and in self-regulation. The idea was that when one enters an achievement situation, one's self-concept in the respective domain will influence the ME evoked in response to the task at hand. The ME, in their turn, will influence and shape one's task-specific self-concept and through it one's more general self-concept in the respective academic domain. To test these predictions a longitudinal study was designed, in which 172 students of 5th and 6th grade participated. On the first testing, they were given a questionnaire tapping their self-concept in mathematics and two verbal mathematical problems. Before and after problem solving, ME were measured on a 4-point scale. After that, a questionnaire addressing their self-concept in verbal mathematical tasks was administered. On the second testing, one week later, the same process was repeated, with two more mathematical problems, which were more difficult than the initial two. On the third testing, the same procedure was followed with all 4 mathematical problems. Students' ability in mathematics was represented by their respective school marks. Path analysis confirmed the predictions stated. It was found that exisiting self-concept influences ME which, along with information from task processing, feed back on self-concept and update it. Thus self-regulation becomes more adaptive to reality demands.

Key words: metacognitive experiences, self-concept, self-regulation