Retour vers liste

Détail de la contribution

Auteur: Sandrien VAN OMMEN

Co-Auteur(s): René KAGER, Utrecht University, the Netherlands

Segmentation to the beat, a cross-linguistic perspective

Abstract/Résumé: This study presents the results of a cross-linguistic experiment with Dutch, Turkish, Polish and Hungarian speakers. Previous studies have shown that listeners interpret stressed or strong syllables as potential beginnings of words in a.o. English and Dutch. This is interpreted as evidence for the Metrical Segmentation Hypothesis [1], which predicts that listeners have and use a parsing ability based on edge-aligned stress. However, most empirical evidence supporting this hypothesis comes from languages with (statistically dominant) word-initial stress. Evidence for a facilitatory effect of right-edge aligned stress is sparse (but see [2]), which leaves the question of how language-specific metrical segmentation is, open. The current non-word spotting experiment was designed to address this issue with a cross-linguistic comparison of Dutch (default penultimate stress, statistically variable), Polish (fixed penultimate stress) Turkish (regular word-final stress, with exceptions) and Hungarian (fixed initial stress). At the beginning of the experiment, participants learned two words in an artificial language. In the next stage, participants had to react as quickly as possible to these words, now embedded in strings of syllables. The stress pattern of strings preceding the target was manipulated, such that the only factor distinguishing the conditions was stress. Manipulation involved phonetic resynthesis of natural speech. The results, obtained by a mixed-effects model, show that there is an overall facilitatory effect of the native stress pattern both on the preceding context and on the target. A closer look at the data shows that the native pre-target stress pattern significantly facilitated target recognition in Turkish (p<.005) and Dutch (p<.05). For Polish and Hungarian there was no effect of stress pattern on the preceding context. The native stress pattern on the target, on the other hand, had a significantly facilitating effect for Turkish (p<.05) and Polish (p<.001), and this time no such effect was found for Dutch. For Hungarian, there was a trend towards facilitation of the native stress pattern on the target. The results support the hypothesis that listeners use native metrical cues in segmentation and invite us to further investigate the interlingual differences. [1]Cutler, A. & Norris, D. 1988, "The Role of Strong Syllables in Segmentation for Lexical Access", J. of Experimental Psychology: Human Perception and Performance, vol. 14, no. 1, pp. 113-121. [2]Kabak, B., Maniwa, K. & Kazanina, N. 2010, "Listeners use vowel harmony and word-final stress to spot nonsense words: A study of Turkish and French", Laboratory Phonology, vol. 1, no. 1, pp. 207-224.