The purpose of this symposium was to explore, via discussing the problems of children's Tsumazuki, new possibilities of relating educational psychological researches to instructional practices. The "Tsumazuki" in Japanese (abbreviated as Tsu) Iiterally means stumbling, but it is used to mean a wide variety of children's failures, from failure in an arithmetic computation up to failure in school life, and even in one's whole life. Saheki presented a paper titled : "Tsumazuki , its causes and countermeasures-a cognitive psychological onsideration". He classified Tsu into three categories ; 1. Tsu in behavior, 2. Tsu in understanding, and 3. Tsu in inability to make Tsu ; according to him, the third is the most serious. Causes of Tsu should be analyzed in correspondence with its two distinct yet interrelated phases : knowledge representation and knowledge generation. In relation to the latter, hestressed the need to study the formative process of the sense that a particular knowldge is real. The sense of reality is generated by an active schema, varying one's perspectives through vigorous activities, rather than by a passive schema locating things in one's already established frame of reference. Also he stressed the necessity of encouraging the wholistic active schema in educational psychological researches. Komabayashi's presentation was titled : The relationship between how an instruction is and how Tsumazuki is grasped-should studies of Tsumazuki be limited to those of errors ?" He pointed out that the Tsu in instruction is usually interpreted as children's errors, negatively valued. However, he contended, Tsu has two meanings ; the negative one of failure in development and the positive one of initiation in development, of which he stressed the latter's importance. He proposed, introducing John Holt's views, that the educational psychologists should studv children's errors, while cooperating with practicing teachers in developing new teaching materials.