The result obtained by Brown et al.(1977)emphasizing that younger subjects could recall idea units of a story in terms of its importance while failing to rate them, was investigated. In experiment I, though the numbers of idea units and levels were less than in Brown's experiment, our younger subjects failed to rate them correctly. Such result suggested that the cause of this phenomenon was not relevant to the procedural problem in which they divided the total units into a certain number of units per level. Therefore in experiment II, a simulation of the cognitive process likely to be necessary for the performance of the rating task, from an analytical point of view in a structured Japanese cognitive frame, "Ki-Sho-Ten-Ketsu", was given to younger subjects to assist them in monitoring explicitly. Such training improved the rating indicating that the cause of their rating failure was likely to have resulted from a lack of knowledge of the metacognitive system, (inability to monitor their own cognitive process).