Doctoral Dissertation Research: Morphological Decomposition in Arabic

当我们听到一个单词时，许多因素会影响我们判断所听到的确实是我们语言中的一个单词的速度和准确性。当听者看到一个单词时，词频是对听者行为的一个特别显著的预测；出现频率较高的单词比出现频率较低的单词被识别得更快。此外，词缀和前缀等单个单词部分的使用频率也会影响对单词整体的识别。这些现象对于存储和加工的心理语言学模型很重要，该模型代表了人们如何习得一门语言的单词，将它们储存在脑海中，并在他们必须感知或产生它们的时候提取它们。这些模式为语言教学以及语言障碍的诊断和治疗提供了有效的方法。然而，关于频率在单词识别中的作用的研究在很大程度上局限于印欧语言，这导致对人类学习、识别和产生语言的机制的了解不完整。这项心理语言学研究的主要目的是探索频率在约旦安曼所说的阿拉伯语口语方言的单词识别中的作用。这项研究需要心理语言学实验来测量对各种口语单词的反应，这些反应基于它们所包含的部分的频率而不同。此外，这些研究包括收集个人语言史，以及从网络、聊天和广播领域收集文本和文字记录。这项研究将有助于更丰富地理解说阿拉伯语的人如何确定他们方言中的单词，以及该单词的哪些属性影响他们确定单词的速度和准确性。确定哪些单词部分对它们在头脑中的存储和访问做出了最大贡献，这将导致向非母语人士教授口语阿拉伯语方言的更明智的方式。集中于哪些单词实际上属于这种特定方言的可靠数据将有助于词典编写和词典建设，加强阿拉伯方言口语现有的稀缺资源。

When we hear a word, many factors affect the speed and accuracy with which we decide that what we have heard is indeed a word of our language. Word frequency is a particularly salient predictor of a listener's behavior when presented with a word; words which are more frequent are recognized more quickly than words which are less frequent. Moreover, the frequency of individual word parts such as suffixes and prefixes also have an effect on the recognition of the word as a whole. These phenomena are important for psycholinguistic models of storage and processing which represent how people acquire words of a language, store them in their minds, and retrieve them when they must perceive or produce them. Such models inform effective methods for teaching language as well as diagnosis and treatment of language disorders. However, studies focusing on the role of frequency in word recognition have been largely limited to Indo-European languages, which results in an incomplete picture of the mechanisms which allow humans to learn, recognize and produce language. The main goal of this psycholinguistic research is to explore the role of frequency in word recognition for a colloquial dialect of Arabic spoken in Amman, Jordan. This research entails psycholinguistic experiments measuring reactions to a variety of kinds of spoken words which differ based on the frequencies of the parts that they contain. Furthermore, these studies involve the collection of personal linguistic histories, as well as the gathering of texts and transcripts from Web, chat, and broadcast domains. This research will contribute to a richer understanding of how Arabic speakers determine what constitutes a word of their dialect, and what properties of the word affect the speed and accuracy with which they do so. Determining which word parts contribute most to their storage and access in the mind will result in more informed ways of teaching colloquial Arabic dialects to non-native speakers. Robust data focusing on which words actually belong to this particular dialect will contribute to dictionary-writing and lexicon-building, strengthening the scarce resources which exist for spoken Arabic dialects.