PROCESSES OF SPEECH SEGMENTATION IN INFANCY

婴儿期的语音分段过程

• 批准号: 2857452
• 负责人: JAMES L MORGAN
• 金额: $ 22.7万
• 依托单位: BROWN UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-02-01 至 2003-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2857452
• 关键词: auditory stimulus; behavioral /social science research tag; clinical research; cognition; human subject; infant human (0-1 year); language development; longitudinal human study; neural information processing; semantics; speech; speech recognition; verbal learning

Learning about words is central to learning about language. To learn about the meanings and properties of words, infants must be able to recover recognizable word shapes from the continuous input speech that they hear. The difficulty of this task is readily apparent to anyone listening to speakers conversing in an unfamiliar language: what one hears is a babble lacking readily identifiable words. The development of word-shape recognition skills is prerequisite to further acquisition of language, for if a language learner cannot break input utterances into their constituent parts, it will be impossible to learn how these parts fit together or what individual parts mean. Despite the fundamental importance of word segmentation and word-shape identification to communicative development, research has only recently begun to explicate the nature and development of requisite cognitive and perceptual capacities. The studies proposed in this application seek to address questions including: How do the perceptual abilities recruited for word segmentation across the second half of the first year interact with developing lexical representations? How and when do infants succeed in disentangling multiple influences- arising from lexical properties, discourse factors, speaker affect, talker variability, and so forth- on prosodic characteristics of word shapes? To what degree does developing knowledge of language-particular patterns of word structure modulate early word-shape identification? Are individual differences in early word-shape recognition abilities predictive of aspects of later language development? What quantities and qualities of language experience are required for development of word segmentation and word-shape identification skills? How do these skills develop in infants with congenital sensorineural mild-to-severe hearing impairment? To these ends, the research proposed here will primarily use a word recognition extension of the conditioned head turning method to investigate 6- to 14-month-old infants' segmentation of fluent natural speech.

学习单词是学习语言的核心。 为了学习单词的含义和属性，婴儿必须能够从他们听到的连续输入语音中恢复可识别的单词形状。 对于任何一个听着说话者用一种不熟悉的语言交谈的人来说，这项任务的难度是显而易见的：一个人听到的是一种缺乏容易识别的单词的胡言乱语。 单词形状识别技能的发展是进一步习得语言的先决条件，因为如果语言学习者不能将输入的话语分解成其组成部分，就不可能学习这些部分如何组合在一起或各个部分的含义。 尽管分词和词形识别对交际发展至关重要，但研究直到最近才开始阐明必要的认知和感知能力的性质和发展。在这个应用程序中提出的研究试图解决的问题，包括：如何招聘的感知能力的分词在第一年的下半年与发展词汇表征的互动？ 婴儿是如何以及何时成功地从词汇特性、话语因素、说话者情感、说话者变异性等因素中分离出对词形韵律特征的多重影响的？ 对特定语言的单词结构模式的认识在多大程度上调节了早期的单词形状识别？ 早期词形识别能力的个体差异是否预示着后期语言发展的各个方面？ 发展分词和词形识别技能需要什么数量和质量的语言经验？ 先天性轻度至重度感音神经性听力障碍的婴儿如何发展这些技能？ 为此，本文提出的研究将主要使用条件转头方法的单词识别扩展来研究6至14个月大的婴儿对流利自然语音的分割。