LANGUAGE LEARNABILITY AND LANGUAGE DEVELOPMENT

语言学习和语言发展

• 批准号: 3315429
• 负责人: STEVEN PINKER
• 金额: $ 13.99万
• 依托单位: MASSACHUSETTS INSTITUTE OF TECHNOLOGY
• 依托单位国家: 美国
• 财政年份: 1983
• 资助国家: 美国
• 起止时间: 1983-09-01 至 1989-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/3315429
• 关键词: child (0-11); child behavior; child psychology; language development; learning; speech; verbal behavior

This project has three aims. The first is development, refinement, and testing of an explicit theory of language acquisition in children. The theory, building on earlier work, will if successful account for the development of language from the start of syntax learning through the acquisition of complex grammatical structures; it will be consistent with linguistic theory and the important facts of language development in children; it will deal with exotic languages as well as English; and it will learn despite noisy, complex, misinterpreted, or nonstandard inputs. A computer simulation of the theory, using "constraint satisfaction" algorithms from artificial intelligence research, will test its adequacy and claims about the kinds of information necessary and sufficient for language acquisition to work. The second aim is to search computerized databases of children's speech for "possible but nonoccurring errors", such as violations of subjacency or dative movement. These searchers will be used to test claims about what sorts of linguistic constraints children are born with. The third part of the project is an experimental investigation of certain of children's rule-learning mechanisms and their cognitive correlates. Children will be taught novel words in isolation to see if they use the semantics of a word to guess its syntactic privileges (e.g. action word = verb; agent = subject). Interactions between conceptual structure and language development will be examined by studies examining the acquisition of constraints on the causative construction, children's use of a "thematic" or spatial schema to conceptualize nonspatial events such as changes of state, and their acquisition of spatial and nonspatial prepositions. Results of the project could yield new methods of assessing children's ability to produce specific grammatical constructions, insight into the language learning mechanisms that can be used in remediation programs for language-disabled individuals, specific techniques to teach new words and constructions to children, and quantitative measures of language development.

这个项目有三个目标。一是发展、完善和 对儿童语言习得的显性理论进行测试。这个 建立在早期工作基础上的理论，如果成功，将解释 语言的发展从句法学习开始一直到 获得复杂的语法结构；它将与 语言理论与《论语》语言发展的重要事实 儿童；它将处理外来语言和英语；它 即使有噪音、复杂、误解或非标准的输入，也会学习。 使用“约束满足”对该理论进行计算机模拟 来自人工智能研究的算法，将测试其充分性 以及关于必要和充分的信息种类的声明 语言习得才能奏效。第二个目标是用计算机进行搜索 儿童语言数据库中的“可能但不会发生的错误”，如 作为对潜伏或与格运动的侵犯。这些搜索者将是 用来测试关于孩子们有什么样的语言限制的说法 与生俱来。项目的第三部分是实验研究。 儿童的某些规则学习机制及其认知 相互关联。孩子们将被孤立地教授新词，看看是否 他们使用一个词的语义来猜测它的句法特权(例如 动作词=动词；施事=主语)。概念之间的相互作用 结构和语言发展将通过以下研究来检验 儿童对使役结构的制约习得 使用“主题”或空间图式来概念化非空间事件 例如状态变化，以及它们对空间和非空间的获取 介词。该项目的结果可能会产生新的评估方法 儿童产生特定语法结构、洞察力的能力 可用于补救的语言学习机制 为语言障碍人士提供的课程，教授的具体技巧 对儿童的新单词和新结构以及对儿童的量化测量 语言发展。