Lee, C.-Y., Tao, L., & Bond, Z.S. (2008) Identification of acoustically modified Mandarin tones by native listeners. Journal of Phonetics, 36, 537–563.
This study investigated the relationship between the amount of acoustic information provided and the effectiveness of tonal identification in Mandarin. The specific question is how native Mandarin listeners identify the tones with limited F0 information. To meet the research goal, tonal contours were modified in the following four ways – intact, center-only, silent-center, and onset-only. Intact condition presented the complete F0 contour of the tone; center-only excluded the beginning and end parts of the contour, whereas silent-center preserved only the two parts; onset-only included the beginning part of the contour. Each target syllable was then embedded into two carrier sentences to create match/mismatch conditions with the preceding tones. Three experiments were conducted accordingly. In the first experiment, subjects only listened to the target syllables with four modification conditions. Results showed that T3 and T4 were rather unaffected by modification effect, whereas T1 relied more on the onset and offset tonal portions and T2 requires the center tonal part for better identification. The effect of preceding tone was further examined in the following two experiments. In Experiment 2, the original carrier sentences were played along with the target syllables, and in Experiment 3, the sentences and targets of match and mismatch conditions were cross-spliced. In both experiments, the effects of preceding tone were not significant. However, it could still be observed that with contexts, confusions between T1/T4 and T2/T3 were reduced to a certain degree. Based on these experimental results, the authors thus concluded that the effect of the amount of acoustic information was tonally dependent. In addition, the extrinsic context information helped eliminate the confusions among tones.
No comments:
Post a Comment