Linking Language Comprehension to Production Patterns

将语言理解与生产模式联系起来

• 批准号: 7109216
• 负责人: MARYELLEN C MACDONALD
• 金额: $ 17.33万
• 依托单位: UNIVERSITY OF WISCONSIN-MADISON
• 依托单位国家: 美国
• 财政年份: 2005
• 资助国家: 美国
• 起止时间: 2005-08-15 至 2009-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7109216
• 关键词: behavior test; behavioral /social science research tag; choice; clinical research; comprehension; data collection methodology /evaluation; developmental psychology; human subject; language development; learning; memory; neural information processing; performance; questionnaires; reading; speech; stimulus interval; syntax; verbal behavior; videotape /videodisc

DESCRIPTION (provided by applicant): This project tests specific hypotheses about dependencies between language comprehension and production, which are typically studied independently. The PI's production-distribution-comprehension (PDC) account holds that utterance planning choices during language production yield distributional patterns in the language, in which certain syntactic structures co-vary with particular word choices, messages, and discourse environments. Comprehenders, through statistical learning during prior comprehension experiences, become highly sensitive to these patterns, and this sensitivity guides comprehension processes. Thus, many aspects of comprehension can ultimately be traced to task demands related to language production. Specific aims of the project include: (1) Link syntactic structure choice in production to mechanisms of sentence planning. (2) Compare the PDC account of comprehension to alternative views of relative clause interpretation. (3) Test the causal relations between production constraints, distributional patterns in the language, and comprehension performance. (4) Relate adult sentence comprehension to statistical learning. (5) Test the current limits of constraint-based models of language comprehension. The PDC approach offers a significant alternative to other views and also may inform language acquisition research by illuminating the role of distributional patterns in child language acquisition. The work also can inform language therapies for brain injured patients in several ways. First, sources of production difficulty are precisely investigated, as are accommodations that unimpaired speakers make in the face of this difficulty. Second, the project investigates the relationship between prior experience with a syntactic construction and comprehension difficulty, which can have implications for the amount and nature of practice that should be provided to patients to improve their comprehension of certain sentence types.

描述（由申请人提供）：该项目测试了有关语言理解和生产之间依赖关系的特定假设，这些假设通常是独立研究的。 PI的生产分布理解（PDC）认为，语言生产过程中的话语计划选择在语言中产生分布模式，其中某些句法结构与特定的单词选择，信息，信息和话语环境共同变化。通过先前理解经验期间的统计学习，理解者对这些模式变得高度敏感，这种敏感性指导了理解过程。因此，理解的许多方面最终都可以追溯到与语言生产有关的任务需求。该项目的具体目的包括：（1）生产中的句法结构选择与句子计划的机制。 （2）将理解的PDC帐户与相对从句解释的替代观点进行比较。 （3）测试生产约束，语言分配模式和理解性能之间的因果关系。 （4）将成人句子理解与统计学习联系起来。 （5）测试基于约束的语言理解模型的当前限制。 PDC方法提供了其他观点的重要替代方法，也可以通过阐明分布模式在儿童语言获取中的作用来为语言获取研究提供信息。这项工作还可以通过多种方式为脑受伤的患者提供语言疗法。首先，对生产难度的来源进行了精确研究，面对这一困难，没有受损的演讲者的住宿也是如此。其次，该项目调查了先前与句法结构的经验与理解难度之间的关系，这可能对应提供给患者的实践数量和性质有影响，以提高他们对某些句子类型的理解。