Statistical Word Learning in Children with Language Disorders

语言障碍儿童的统计词汇学习

• 批准号: 8594469
• 负责人: Eileen Katherine Haebig
• 金额: $ 2.94万
• 依托单位: UNIVERSITY OF WISCONSIN-MADISON
• 依托单位国家: 美国
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-08-01 至 2015-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8594469
• 关键词: Acoustics; Address; Bees; Behavioral; Characteristics; Child; Communication; Conflict (Psychology); Cues; Data; Development; Educational process of instructing; Exhibits; Exposure to; Eye; Frequencies; Goals; Hearing; Heterogeneity; Impairment; Intervention; Investigation; Knowledge; Label; Language; Language Development; Language Disorders; Language Therapy; Learning; Linguistics; Link; Literature; Machine Learning; Maps; Measures; Methods; Nature; Participant; Pattern; Performance; Phase; Population; Probability; Process; Quality of life; Reading; Research; Speech; Statistical Methods; Stream; Testing; Time; Vocabulary; Work; autism spectrum disorder; classical conditioning; gaze; innovation; movie; novel; public health relevance; skills; social; sound; specific language impairment; theories

DESCRIPTION (provided by applicant): Many children with autism spectrum disorders (ASD) and specific language impairment (SLI) have difficulties learning words, which negatively impacts academic performance, functional independence, and quality of life. Despite the importance of vocabulary development, relatively little is known about the learning mechanisms that underlie these deficits. Furthermore, the overlap of language characteristics in children with ASD and SLI is not well understood. An understanding of language-learning mechanisms in children with ASD and SLI has important theoretical implications upon which effective language therapies rely. Although statistical learning (i.e., sensitivity to linguistic regularities) is accpted as a robust language-learning mechanism in multiple domains of language in typical development, very little is known about statistical learning in children with ASD and SLI. This study will address this gap in knowledge by examining the impact of statistical learning in word learning. Specifically, this study will assess how children link knowledge of native language sound patterns to meaning. The contribution of statistical learning in word learning in children with atypical language development will be assessed through three studies. Specific Aim 1 will investigate prerequisite skills in word learning. Study 1a will assess the use of statistical learnng through a segmenting task (i.e., segmenting sound units from a continuous speech stream with the use of implicit transitional probability cues) without word learning. Study 1b will test associative word learning of novel labels paired with novel objects without segmentation. Because children with ASD and SLI have difficulties learning words and have difficulties consolidating linguistic information in word learning, it is important to test component skills in word learning. Once prerequisite skills have been assessed, Study 2, addressing specific aim 2, will assess whether children with ASD and SLI are able to map meaning onto newly segmented words. A significant relationship between prerequisite skills and specific aim 2 is anticipated; however, comparison among the different groups will allow for identification of potentially different abilities in word learning skills in the two populations with atypical language development. Both implicit and explicit teaching and testing methods will be used to capture subtle real-time learning information while placing minimal behavioral demands on the children. Within- and between- group analyses will assess learning in children with ASD, SLI, and typical development. New statistical methods for analyzing eye-gaze data will be employed to address potential analytical difficulties given the heterogeneity and possible lack of within-subject stability in children with ASD and SLI. Additionally, because of known social deficits in children with ASD, the studies will use non-social tasks. Findings from this study will contribute important information to our theoretical understanding of language acquisition in children with atypical language development.

描述（由申请人提供）：许多自闭症谱系障碍（ASD）和特定语言障碍（SLI）的儿童学习单词困难，这会对学业表现，功能独立性和生活质量产生负面影响。尽管词汇发展的重要性很重要，但对这些缺陷的学习机制知之甚少。此外，儿童的语言特征的重叠 ASD和SLI不太了解。对ASD和SLI儿童的语言学习机制的理解具有有效语言疗法的重要理论意义。尽管统计学习（即对语言规律性的敏感性）被视为典型发展中多个语言领域的强大语言学习机制，但对于ASD和SLI儿童的统计学习知之甚少。这项研究将通过检查统计学习在单词学习中的影响来解决这一差距。具体而言，这项研究将评估儿童如何将母语声音模式的知识联系起来。在具有非典型语言发展儿童单词学习中统计学习的贡献将通过三项研究评估。特定目标1将研究单词学习中的先决条件。研究1a将通过分段任务（即，通过使用隐式过渡概率提示从连续的语音流中分割声音单元对统计学习的使用）评估统计学习的使用。研究1b将测试新标签与新物体配对而无需分割的新标签的关联单词学习。由于ASD和SLI的儿童在学习单词中遇到困难，并且在单词学习中的语言信息巩固了困难，因此测试单词学习中的组成技能很重要。一旦评估了先决条件的技能，研究2，解决特定目标2的研究将评估患有ASD和SLI的儿童是否能够将含义映射到新分段的单词上。预计先决条件和特定目标2之间存在显着的关系；但是，不同群体之间的比较将允许识别两个人群中具有非典型语言发展的两个人群中潜在不同的能力。隐式和明确的教学和测试方法都将用于捕获微妙的实时学习信息，同时对儿童的行为需求最少。组内和组之间的分析将评估ASD，SLI和典型发展儿童的学习。鉴于ASD和SLI儿童的异质性和可能缺乏受试者的稳定性，将采用用于分析眼睛目光数据的新统计方法来解决潜在的分析困难。此外，由于ASD儿童的已知社会缺陷，研究将使用非社会任务。这项研究的结果将有助于重要 信息以我们对具有非典型语言发展儿童的语言获取的理论理解。