Comprehension in Naturalistic Contexts

自然主义背景下的理解

• 批准号: 6826266
• 负责人: Fernanda Ferreira
• 金额: $ 23.55万
• 依托单位: MICHIGAN STATE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-12-01 至 2006-11-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6826266
• 关键词: behavioral /social science research tag; clinical research; comprehension; differential semantics; human subject; language; language translation; neural information processing; psycholinguistics; speech recognition; syntax; verbal behavior

DESCRIPTION (provided by applicant): The proposed studies will focus on the effects of disfluencies on parsing and interpretation processes. The proposed experiments are based on the results of extensive preliminary work demonstrating the viability of the key assumptions. Disfluencies influence the operations of the syntactic parsing mechanisms by causing the parser to be committed longer to the structure it is in the process of building when the disfluencies are encountered. In addition, disfluencies tend to occur before certain types of constituents, and so they provide useful information to help the parser choose from among possible structural alternatives. A particular type of disfluency - corrections - may influence parsing because the misarticulated information is sometimes relevant to choosing between rival syntactic analyses of an ambiguous string. The proposed studies (ten in total, each consisting of 1-3 experiments) will lead to a better understanding of how the fundamental mechanisms of the language processing system accommodate and even take advantage of disfluencies in speech. They will also contribute towards the development of basic models of language processing, because disfluencies can be used as a tool to decide between rival architectures of the comprehension system. This research ultimately may contribute towards the creation of a more ecological approach to psycholinguistic questions - an approach that acknowledges the important role of linguistic structures in processing but pays more attention to the real properties of linguistic stimuli encountered in normal conversational contexts. The research is important for the development of basic models of language processing, but in addition the results would have important implications for studies of how individuals suffering from health-related cognitive disorders understand language, particularly in natural environments.

描述(申请人提供)：拟议的研究将集中在不流利对分析和口译过程的影响。拟议的实验是基于广泛的前期工作的结果，证明了关键假设的可行性。当遇到不流畅时，不流畅会导致解析器更长时间地致力于其构建过程中的结构，从而影响句法解析机制的操作。此外，不流利性往往发生在某些类型的成分之前，因此它们提供了有用的信息来帮助解析器从可能的结构备选方案中进行选择。一种特定类型的不流利--纠正--可能会影响句法分析，因为发音错误的信息有时与在歧义字符串的竞争句法分析之间进行选择有关。 这项拟议的研究(总共10项，每项实验由1-3个实验组成)将有助于更好地理解语言处理系统的基本机制如何适应甚至利用言语中的不流利。它们还将有助于语言处理基本模型的发展，因为不流利性可以被用作在理解系统的竞争体系结构之间做出决定的工具。这项研究最终可能有助于创建一种更生态的方法来处理心理语言学问题，这种方法承认语言结构在加工过程中的重要作用，但更关注正常对话语境中遇到的语言刺激的真实性质。这项研究对于开发语言处理的基本模式很重要，但除此之外，研究结果还将对研究患有健康相关认知障碍的人如何理解语言，特别是在自然环境中理解语言具有重要意义。