Hirata, Y., Whitehurst, E., & Cullings, E. (2007). Training native English speakers to identify Japanese vowel length contrast with sentences at varied speaking rates. Journal of the Acoustical Society of America, 121 (6), 3837–3845.
The study aims to examine the high phonetic variability hypothesis (Pisoni, 1995), which suggested that language learners could form perceptual categories more effectively if they are trained with material of higher variability. The main task in this study was to train the native-English-speaking subjects to identify Japanese vowel length of the target words. The subjects were first trained with slow-only, fast-only, or slow-fast materials, and then they were asked to identify the vowel length of the target words, which were embedded in sentences spoken with different rates (slow-rate, normal-rate, and fast-rate). The results showed that those who were trained with slow-fast materials improved the most, while those who were trained with slow-only and fast-only materials had smaller improvement. Therefore, the authors concluded that the present results supported the high phonetic variability hypothesis.
No comments:
Post a Comment