Suppr超能文献

协同发音和词级误读的电生理反应。

Electrophysiological responses to coarticulatory and word level miscues.

机构信息

School of Communication Sciences and Disorders, Ebron College, The University of Western Ontario, London, Ontario, Canada.

出版信息

J Exp Psychol Hum Percept Perform. 2011 Aug;37(4):1275-91. doi: 10.1037/a0023506.

Abstract

The influence of coarticulation cues on spoken word recognition is not yet well understood. This acoustic/phonetic variation may be processed early and recognized as sensory noise to be stripped away, or it may influence processing at a later prelexical stage. The present study used event-related potentials (ERPs) in a picture/spoken word matching paradigm to examine the temporal dynamics of stimuli systematically violating expectations at three levels: entire word (lexical), initial phoneme (phonemic), or in coarticulation cues contained in the initial phoneme (subphonemic). We found that both coarticulatory and phonemic mismatches resulted in increased negativity in the N280, interpreted as indexing prelexical processing of subphonemic information. Further analyses revealed that the point of uniqueness differentially modulated subsequent early or late negativity depending on whether the first or second segment matched expectations, respectively. Finally, it was found that word-level but not coarticulatory mismatches modulated the later-going N400 component, indicating that subphonemic information does not influence word-level selection provided no lexical change has occurred. The results indicate that acoustic/phonetic variation resulting from coarticulation is preserved in and influences spoken word recognition as it becomes available, particularly during prelexical processing.

摘要

协同发音线索对口语识别的影响尚未得到很好的理解。这种声学/语音变化可能会被早期处理,并被视为需要去除的感觉噪音,也可能会影响到词汇前阶段的处理。本研究使用事件相关电位(ERPs)在图片/口语匹配范式中,系统地检查了在三个层面上违反期望的刺激的时间动态:整个单词(词汇)、初始音素(音位)或包含在初始音素中的协同发音线索(次音位)。我们发现,协同发音和音位不匹配都会导致 N280 中负向增加,这被解释为预先处理次音位信息。进一步的分析表明,独特性的关键点根据第一个或第二个片段分别与预期相匹配,从而分别调节随后的早期或晚期负向。最后,发现仅词级但非协同发音的不匹配会调制后续的 N400 成分,表明只要没有发生词汇变化,次音位信息就不会影响词级选择。研究结果表明,协同发音产生的声学/语音变化在口语识别中得到保留并产生影响,特别是在词汇前处理阶段。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验