基于语料库比较规则竞争模型和单词规则模型  

Comparison of Rules and Competition Model and Words and Rules Model

作　　者：赵亮[1] 

机构地区：[1]北京科技大学外国语学院,北京100083

出　　处：《北京科技大学学报（社会科学版）》2017年第4期32-37,共6页Journal of University of Science and Technology Beijing(Social Sciences Edition)

基　　金：教育部人文社会科学规划项目(编号为12YJC740147)部分研究成果

摘　　要：在语言习得领域,动词习得一直是备受争议的关键问题之一,尤其是英语动词过去式的习得机制。Yang提出的规则竞争模型(Rules and Competition Model)和Pinker提出的单词规则模型(Words and Rules Model)是其中比较有影响力的理论模型。文章分析了这两个理论模型的基本原理和习得机制,比较其异同点,并通过CHILDES语料库中儿童语言习得的真实语料来验证其解释力。结果显示:规则竞争模型更好地解释了输入频率的作用、动词和过去式的相似性和过度规则化错误;而单词竞争模型在解释不规则动词过去式成因方面表现得更好。Verb acquisition,especially the acquisition mechanism of English verb past tense,has always been one of the controversial key issues in language acquisition. Yang＇s Rules and Competition Model（ the RC model） and Pinker＇s Words and Rules Model（ the WR model） are two different influential theoretical models. This article has analyzed fundamental principles and language mechanisms of these two models,compared their similarities and differences,and tested their explanatory power on the basis of real language acquisition data selected from CHILDES（ Child Language Data Exchange System）. The findings of the study reveal that the RC model better explains the role of input frequency,the similarity of a verb and its past tense form,and overregularization errors while the WR model can better interpret the origin of irregular verb past tense.

关 键 词：动词过去式习得 单词规则模型 规则竞争模型 CHILDES语料库 

分 类 号：H0[语言文字—语言学]