Felder Verena, Jönsson-Steiner Elisabet, Eulitz Carsten, Lahiri Aditi
Department of Linguistics, University of Konstanz, 78457 Konstanz, Germany.
Atten Percept Psychophys. 2009 Nov;71(8):1890-9. doi: 10.3758/APP.71.8.1890.
Languages such as Swedish use suprasegmental information such as tone, over and above segments, to mark lexical contrast. Theories differ with respect to the abstractness and specification of tone in the mental lexicon. In a forced choice task, we tested Swedish listeners' responses to words with segmentally identical first syllables differing in tonal contours (characterized as Accents 1 and 2). We assumed Accent 1 to be lexically specified for a subset of words and hypothesized that this specification would speed up word accent identification. As was predicted, listeners were fastest in choosing the tonally correct word when the accent was lexically specified. We conclude that the processing of surface tonal contours is governed by their underlying lexical structure with tonal specification.
像瑞典语这样的语言，除了音段之外，还使用诸如声调等超音段信息来标记词汇差异。关于心理词典中声调的抽象性和具体特征，不同理论存在差异。在一项强制选择任务中，我们测试了瑞典语听众对首音节在音段上相同但声调轮廓不同（分为重音1和重音2）的单词的反应。我们假定重音1在词汇上为一部分单词所特有，并假设这种特性会加快单词重音识别。正如所预测的那样，当重音在词汇上有明确界定时，听众选择声调正确单词的速度最快。我们得出结论，表面声调轮廓的处理受其具有声调特性的潜在词汇结构的支配。
