摘要：Although there are plenty of studies about learning kanji, there are few if any that elucidate the key role and effect of the learners' personal visual cognitive processing in the learning process, specifically in the cognitive process that learners experience when guided to mentally divide the shapes of individual kanji into distinct parts as they view them. (We refer to these parts as the "visuospatial-temporal blocks", individually updatable visuospatial cognitive non-overlapping units). This study focuses on the question of whether or not guiding learners with a kanji-learning system that makes use of their personal visual cognition based on a suitably preestablished set of possible deconstructions (ways to divide a kanji into blocks) produces a significant effect on learner's recall of kanji's shapes and meanings. First, a novel approach to kanji-learning called "Learner's Visualization (LV) Approach" was conceived and formally proposed; the LV Approach includes a learner-guided technique that presents the user with a set of layouts representing how an individual kanji could be logically divided into blocks, allows the user to choose which layout he or she prefers (which layout currently appears the most logical or understandable to that user), and then guides the user in further exploration and learning about blocks within his or her chosen layout. This paper evaluates the merit of this novel approach for learners who are not accustomed to using a writing system similar to kanji, focusing specifically on the implications of taking into account learners' personal visual cognitive processing. The implementation of a learning support system based on the LV Approach and the positive experimental results of the proposed approach are discussed.
