The Acquisition of Grammatical Categories by Infants

像名词和动词这样的语法范畴是语言结构的基石，控制着单词的顺序、重排和修饰方式。因此，为了学习一门特定语言的结构，孩子必须首先确定类别的结构：名词是什么，动词是什么，形容词是什么，等等。首席调查员最近的研究表明，孩子们可以使用单词的线性顺序来得出关于类别的结论（例如，在英语中，单词“you”后面和单词“it”前面的单词通常是动词）。然而，许多关键问题仍然存在。例如，有很多种模式可以用来对单词进行分类；目前尚不清楚婴儿注意并使用哪些特定模式来形成类别。也不知道在结构上与英语非常不同的语言中的分类是否可以用类似的方式学习。因此，目前还不能确切地了解婴儿有什么样的学习机制，以及他们进行什么样的计算来对单词进行分类。这个项目通过对成人和婴儿的一系列实验来解决这些问题，使用真实和人工语言来研究非常小的孩子是如何对单词进行分类的。同时，该项目还对与英语在词序特性上有很大不同的语言进行计算分析。这个项目解决了关于儿童语言学习基本方面的学习过程和表征的关键问题。了解儿童最初是如何对单词进行分类的，对于了解他们如何获得语言的句法以及如何学习单词的含义至关重要。这些问题的答案可能对语言习得领域产生重大的理论影响，并可能为发育性语言障碍和成人第二语言习得提供理论依据。

Grammatical categories like noun and verb are the building blocks of structure in language, controlling the way in which words are ordered, rearranged, and modified. Hence, in order to learn how a particular language is structured, a child must first determine how the categories are structured: what are the nouns, the verbs, are there adjectives, etc. Recent research by the Principal Investigator has shown that children can use patterns in the linear order of words to draw conclusions about categories (e.g., words following the word ''you'' and preceding the word ''it'' are often verbs in English). However, many critical questions remain. For instance, there are many kinds of patterns that potentially could be used to classify words; it is not known which particular patterns infants attend to and use to form categories. Nor is it known whether categories in languages that are structurally very different from English could be learned in similar ways. Consequently, it is not yet understood precisely what kinds of learning mechanisms infants have at their disposal, and what kinds of computations they carry out to categorize words. This project addresses these issues in a series of experiments with adults and infants, using both real and artificial languages, to investigate how very young children categorize words. Concurrently, the project undertakes computational analyses of languages that are very different from English in how their word-order properties. This project addresses crucial questions about the learning processes and representations involved in fundamental aspects of language learning in children. Understanding how children initially categorize words is crucial to understanding how they acquire the syntax of their language, and how they learn word meanings. The answers to these questions could have a significant theoretical impact on the field of language acquisition and will potentially inform theories on developmental language disabilities, and second language acquisition in adults.