Department of Psychology, University of Connecticut, 406 Babbidge Road, Unit 1020, Storrs, CT 06269 and Haskins Laboratories, New Haven, CT 06511, USA.
Neuropsychologia. 2011 Jun;49(7):1761-70. doi: 10.1016/j.neuropsychologia.2011.02.054. Epub 2011 Mar 4.
Evidence from event-related potential (ERP) analyses of English spoken words suggests that the time course of English word recognition in monosyllables is cumulative. Different types of phonological competitors (i.e., rhymes and cohorts) modulate the temporal grain of ERP components differentially (Desroches, Newman, & Joanisse, 2009). The time course of Chinese monosyllabic spoken word recognition could be different from that of English due to the differences in syllable structure between the two languages (e.g., lexical tones). The present study investigated the time course of Chinese monosyllabic spoken word recognition using ERPs to record brain responses online while subjects listened to spoken words. During the experiment, participants were asked to compare a target picture with a subsequent picture by judging whether or not these two pictures belonged to the same semantic category. The spoken word was presented between the two pictures, and participants were not required to respond during its presentation. We manipulated phonological competition by presenting spoken words that either matched or mismatched the target picture in one of the following four ways: onset mismatch, rime mismatch, tone mismatch, or syllable mismatch. In contrast to the English findings, our findings showed that the three partial mismatches (onset, rime, and tone mismatches) equally modulated the amplitudes and time courses of the N400 (a negative component that peaks about 400ms after the spoken word), whereas, the syllable mismatched words elicited an earlier and stronger N400 than the three partial mismatched words. The results shed light on the important role of syllable-level awareness in Chinese spoken word recognition and also imply that the recognition of Chinese monosyllabic words might rely more on global similarity of the whole syllable structure or syllable-based holistic processing rather than phonemic segment-based processing. We interpret the differences in spoken word processing between Chinese and English listeners as being due to morphosyllabic structural differences between the two languages.
事件相关电位 (ERP) 分析英语口语单词的证据表明，单音节英语单词识别的时间进程是累积的。不同类型的语音竞争词（即押韵词和同韵词）以不同的方式调节 ERP 成分的时间粒度（Desroches、Newman 和 Joanisse，2009）。由于两种语言的音节结构不同（例如，词汇声调），汉语单音节口语单词识别的时间进程可能与英语不同。本研究使用 ERP 在线记录大脑对口语单词的反应，以研究汉语单音节口语单词识别的时间进程。在实验中，要求参与者通过判断两张图片是否属于同一语义类别来比较目标图片和后续图片。在两张图片之间呈现口语单词，参与者在呈现期间不需要做出反应。我们通过以以下四种方式呈现与目标图片匹配或不匹配的口语单词来操纵语音竞争：首音不匹配、韵脚不匹配、声调不匹配或音节不匹配。与英语研究结果相反，我们的研究结果表明，三种部分不匹配（首音、韵脚和声调不匹配）同样调节 N400（在口语单词出现后约 400ms 时出现的负成分）的幅度和时间进程，而音节不匹配的单词比三种部分不匹配的单词更早且更强地诱发 N400。研究结果揭示了音节级别的意识在汉语口语单词识别中的重要作用，也意味着汉语单音节单词的识别可能更依赖于整个音节结构或基于音节的整体处理的整体相似性，而不是基于音素段的处理。我们将汉语和英语听众在口语处理方面的差异解释为两种语言之间的形态音节结构差异。
