Poster #: 122
Topic: Speech and language (incl. deficits)
Thursday, Sep 10, 2015
On the role of different acoustic-phonetic cues in encoding voicing in Russian and German: a cross-linguistic MMN study with native and non-native stop consonants
1Sprachwissenschaft, University of Konstanz, Konstanz, Germany
2University of Konstanz, Konstanz, Germany
The voicing contrast of stop consonants in Russian and German is based on different laryngeal features. German speakers differentiate between aspirated and unaspirated stops, whereas native speakers of Russian depend upon voice onset time (VOT, negative vs. positive). This may result in misperceptions. Taking an English example, the Russian pronunciation of the word “pair” would sound like “bear” to a German listener, and vice versa. The best explanation would be that listeners automatically rely on acoustic cues used in their mother tongue, categorizing non-native stops into native categories. Thus, for Russian listeners aspirated and unaspirated stops with positive VOT, which form a contrast in German, would fall within the category of voiceless stops. For German listeners, stop consonants with negative and positive VOT (both unaspirated), forming a contrast in Russian, would be categorized as voiced stops. To test this prediction, we measured the mismatch negativity (MMN) using the roving standard design on three types of contrasts in CV-syllables for Russian and German participants. The MMN differed between the groups, not only in the traditional latency window, but also later up to 600 ms. The German group showed early and late MMNs for contrasts with aspirated stops as deviants. For the Russian group, only the late MMNs showed an influence of the native language cues. This indicates that perceptual assimilation of non-native sounds seems to be based on native language-specific acoustic cues. Moreover, the reliance on different acoustic cues by Russian and German listeners affects the temporal dynamics of MMN effects.