Department of Linguistics, Radboud University Nijmegen, Nijmegen, The Netherlands.
Brain Res. 2010 Apr 30;1328:79-88. doi: 10.1016/j.brainres.2010.02.053. Epub 2010 Mar 1.
An MEG experiment was carried out in order to compare the processing of lexical-tonal and intonational contrasts, based on the tonal dialect of Roermond (the Netherlands). A set of words with identical phoneme sequences but distinct pitch contours, which represented different lexical meanings or discourse meanings (statement vs. question), were presented to native speakers as well as to a control group of speakers of Standard Dutch, a non-tone language. The stimuli were arranged in a mismatch paradigm, under three experimental conditions: in the first condition (lexical), the pitch contour differences between standard and deviant stimuli reflected differences between lexical meanings; in the second condition (intonational), the stimuli differed in their discourse meaning; in the third condition (combined), they differed both in their lexical and discourse meaning. In all three conditions, native as well as non-native responses showed a clear MMNm (magnetic mismatch negativity) in a time window from 150 to 250 ms after the divergence point of standard and deviant pitch contours. In the lexical condition, a stronger response was found over the left temporal cortex of native as well as non-native speakers. In the intonational condition, the same activation pattern was observed in the control group, but not in the group of native speakers, who showed a right-hemisphere dominance instead. Finally, in the combined (lexical and intonational) condition, brain reactions appeared to represent the summation of the patterns found in the other two conditions. In sum, the lateralization of pitch processing is condition-dependent in the native group only, which suggests that language experience determines how processes should be distributed over both temporal cortices, according to the functions available in the grammar.
为了比较基于罗尔蒙德(荷兰)声调方言的词汇-声调对比和语调对比的处理,进行了一项 MEG 实验。一组具有相同音素序列但不同音高轮廓的单词,代表不同的词汇意义或话语意义(陈述与疑问),呈现给母语为荷兰语的本族语者以及非声调语言的标准荷兰语的对照组。刺激物采用不匹配范式排列,分为三种实验条件:在第一种条件(词汇)中,标准刺激与偏差刺激之间的音高轮廓差异反映了词汇意义的差异;在第二种条件(语调)中,刺激在话语意义上有所不同;在第三种条件(组合)中,它们在词汇和话语意义上都有所不同。在所有三种条件下,母语者和非母语者的反应都在标准和偏差音高轮廓发散点后 150 至 250 毫秒的时间窗口内显示出明显的 MMNm(磁失配负性)。在词汇条件下,母语者和非母语者的左颞叶都发现了更强的反应。在语调条件下,对照组观察到了相同的激活模式,但母语者组没有观察到,而是表现出右半球优势。最后,在组合(词汇和语调)条件下,大脑反应似乎代表了其他两种条件中发现的模式的总和。总之,只有在本族语者群体中,音高处理的偏侧化是条件依赖的,这表明语言经验决定了根据语法中的可用功能如何将处理过程分配到两个颞叶中。