Department of Speech, Language, Hearing Sciences, University of Minnesota, Minnesota, USA.
Rotman Research Institute, Baycrest Centre, Ontario, Canada.
Neuroreport. 2021 Mar 24;32(5):373-377. doi: 10.1097/WNR.0000000000001605.
Phonological encoding, a subprocess in speech production, proceeds incrementally from word onset to offset. However, the incremental nature of encoding for multisyllabic words is unclear, and limited research has examined neural activity underlying the seriality of the process. In the present study, we investigated the timing of encoding between and within syllables of bisyllabic words using a data-driven cluster-based analysis of electroencephalography (EEG) data. In a phoneme-monitoring task, young adults covertly named pictures of bisyllabic words with a prespecified target phoneme present or absent. Target phonemes in target-present trials were distributed among four serial positions of the word concept: first syllable (S1) onset or offset and second syllable (S2) onset or offset. Upon covert naming, participants responded to target presence via button press or withheld responses for target absence. Neuroelectric activity during task performance was recorded using EEG and analyzed using cluster-based permutation testing. Faster response times and differences in neural activity were observed for monitoring targets at S1 onset than S2 onset, and for monitoring targets at S1 onset than S1 offset. No differences were found between monitoring targets at S2 onset and S2 offset. Our study supports the incremental nature of phonological encoding in bisyllabic words. Furthermore, the neural findings confirm that the serial time course of encoding in bisyllabic words extends to phonemes within the first, but not the subsequent syllable. Findings may have implications for current models of speech production.
语音编码是言语产生的一个子过程，从单词的起始到结束逐步进行。然而，对于多音节词的编码的递增性质尚不清楚，并且有限的研究已经检查了支持该过程的序列性的神经活动。在本研究中，我们使用基于脑电图 (EEG) 数据的聚类分析，研究了双音节词中音节之间和音节内的编码时间。在语音监测任务中，年轻人在目标音存在或不存在的情况下，对双音节词的图片进行内隐命名。目标音在目标存在的试验中分布在单词概念的四个连续位置：第一个音节（S1）的起始或结束和第二个音节（S2）的起始或结束。在进行内隐命名时，参与者通过按钮按压来响应目标的存在或在目标不存在时抑制反应。使用 EEG 记录任务执行期间的神经电活动，并使用基于聚类的置换检验进行分析。与监测 S1 起始处的目标相比，监测 S1 结束处的目标的反应时间更快，并且监测 S1 起始处的目标的神经活动也更快。在监测 S2 起始处和 S2 结束处的目标之间未发现差异。我们的研究支持双音节词中语音编码的递增性质。此外，神经发现证实，双音节词中编码的序列时间进程延伸到第一个音节内的音位，但不延伸到随后的音节内的音位。这些发现可能对当前的言语产生模型产生影响。
