CAJ | 학술논문

通过对类别学习中的阻碍效应进行系统研究，尝试性地提出了类别学习的双机制理论。三个实验分别考察：样例特征随机呈现时；定义特征固定呈现在样例首位时；以及刺激材料为知觉图形类别时，类别学习中的阻碍效应。三个实验的研究结果都发现：在类别学习中存在一定程度的阻碍效应，支持类别学习同时存在联结学习机制和认知学习机制的双机制观点。
통과대유별학습중적조애효응진행계통연구，상시성지제출료유별학습적쌍궤제이론。삼개실험분별고찰：양례특정수궤정현시；정의특정고정정현재양례수위시；이급자격재료위지각도형유별시，유별학습중적조애효응。삼개실험적연구결과도발현：재유별학습중존재일정정도적조애효응，지지유별학습동시존재련결학습궤제화인지학습궤제적쌍궤제관점。
Most models indicate that category learning is a kind of associated learning driven by reducing categorization errors. Accord- ing to this account, people solely learn features useful for categorization. By contrast, some researchers indicate that different from asso- ciation learning, category learning is a high level cognitive learning, and people like to learn more features than classification. A dual - processing account is advanced to describe the category learning mechanisms, which indicates that there are both an error - driven asso- ciative mechanism and a cognitive mechanism in category learning. According to the dual - processing account, the goal of associative mechanism in category learning is to learn as fewer features as possible, which is error - driven. Whereas, the goal of cognitive mecha- nism is to learn as more features as possible, and learning more features would be useful for future feature induction and so on. And these two mechanisms exist in category learning and affect features learning. Three experiments examined the dual - processing account in a blocking paradigm. In the blocking paradigm, participants learned one defining feature dimension, which was crucial for categorization. By accounting the blocking effect that the defining feature dimen- sion had on other features learning, we investigated the dual -processing mechanism in category learning. In Experiment 1, all features were presented randomly to investigate the blocking effect in category learning. The results showed that even some non - defining fea- tures were blocked by pre - learning of the defining features, which supports that the associative mechanism and cognitive mechanism both exist in category learning. In Experiment 2, defining features were defined on the top of exemplars to prevent participants from learning the non - defining features that, in Experiment 1, was caused by searching for defining features. Consistent with Experiment 1, the results supported the dual - processing account. Different with the sentence described categories in Experiment 1 and 2, figure cate- gories were adopted in Experiment 3 and similar results were found. Across all three experiments, the dual - processing account has been verified in different feature presentations and category materi- als. The associative mechanism predicted the blocking effect in category learning, which the cognitive mechanism predicted the learning of some non - defining features.