http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
趙明翰,高成龍,趙炳煥 대한민국 학술원 2007 학술원논문집 : 인문, 사회과학편 Vol.46 No.1
본 연구는 담화 이해와 추리에 관한 여러 실험들의 결과를 검토했고, 잘 알려진 실험 문장들을 변형하여 직관을 검토했고 또한 지식 구성을 간단히 하여 공명이론 기반의 시뮬레이션도 수행하였다. 실험 자료, 직관 자료 및 시뮬레이션 결과를 잘 설명하는 관점이 기억기반 접근임을 논의했으며, 더 나아가 이 접근에서 가정되는 작업기억과 장기기억간의 유연한 상호작용의 시각에서 개인차의 연구들을 고찰했다. 개인차 연구들의 핵심 결과는 비숙련 독자들도 숙련 독자들과 마찬가지로 관련 지식이 있으며, 더 나아가 관련 지식이 장기기억에서는 활성화되었음에도 불구하고 쓰이지 않을 수 있다는 것이었다. Zoh와 Ahn[2006]은 모순 정보를 담고 있는 문장들이 인과와 대초 추리 실험문장들 사이에 있을 때 비숙련 독자들도 숙련 독자처럼 인과와 대초추리를 하는 결과를 보고했다. 이 결과를 비숙련 독자들도 모순정보가 있을 때 더 세밀한 명제를 구성하여 전체적으로 정합성을 찾으려고 시도를 했으며 이 모순 문장들에 관한 누적된 처리 기술이 실험 문장으로 전이된 것으로 해석하였다. 결론적으로 Zoh와 Ahn의 연구가 담화 이해 및 추리와 같은 복잡한 인지 과제의 처리에서 핵심적인 역할을 하는 작업기억의 용량이 제한은 있으나 고정되어있는 것이 아니라 신축적이라는 것을 시사한다고 논의하였다. This study reviewed several experimental results on discourse comprehension and inference, examined our intuitions on the modified sentences constructed by changing one word in the well-known sentences, and carried out the computer simulation on the basis of the resonance model. We discussed that memory-based approach seemed to account for the results on the experiments and simulations pretty well. Furthermore, the studies on individual differences on inference were examined from memory-based approach’s viewpoint of the flexible relation between working memory and long-term memory. The crucial results were that less skilled readers was not ready to use the knowledge related to the inference, even though they had the knowledge, which was activated in long-term memory. Zoh and Ahn[2006] showed that less skilled readers made causal and contrastive inferences only when fillers with inconsistent information were used. This result was interpreted as showing that less skilled readers tried to construct more constrained propositions to seek the global coherence on the face of inconsistent information, and seemed to accumulate the knowledge and skills related to the inference, which were finally transferred to the comprehension of experimental sentences. We discussed that Zoh and Ahn’s study indicated that working memory for such complex cognitive tasks as discourse comprehension and inference waxes and wanes.
조명한 서울대학교 어학연구소 1977 語學硏究 Vol.13 No.2
While the study of language development in children has rich historical tradition, especially, the last decade has seen a much more rapid development in subtelety of research methods, depth of theoretical issues, and abundance of new findings. This paper aimed to review five current questions about language development in order to suggest some promising directions. First, any effort to try to apply a certain grammar to child language has only taken researchers to endless distributional detail of doubtful generality, and never sufficiently complete. Therefore, 'the goal of creating a type of grammar that is suitable for both adult and child language remains. The two cannot be considered in isolation from each other, for eventually the child's language becomes that of the adult.' Second, both of the findings that reveal the gradual continuity of language development and the tremendous effects of parental speech on children strongly discourage the innateness hypothesis. Third, a lot of characteristics in child speech has proved to have some generality which supports the universality hypothesis, but not innateness. The D.I. Slobin's principle, 'new forms express old functions, and new functions are first expressed by old forms', is compatible with cognitive prerequisites for the syntactic development. Fourth, recent studies on the structure and/or acquisition of word meaning have focused attention on the role played by prototypes or best exemplars in the internal structure of natural categories to which words refer. A new paradigm of reference that radically departs from the traditional view of all-or-none categorization is emerging. Fifth, efforts to teach chimpanzees language are remarkably in progress. The chimpanzees have become capable to demonstrate stage I child language, so that the species-specific and species-uniform assumption for human language is open to doubt.