Discrimination of mono-syllables in sentence context: the case of Japanese listeners' perception of /ba/-/da/ continuum

摘要

It is widely known that native speech sounds are perceived according to the phonetic categories of the native language: the process is called categorical perception. The hypothesis of categorical perception assumes that two stimuli are perceived to be different only when these are identified as different. Therefore, the best discrimination performance (discrimination peak) is observed when the paired stimuli are members of different categories. Traditionally, categorical perception has been tested using monosyllables presented in isolation. However, my recent research has shown that the discrimination peak is not observed when the mono-syllabic stimuli are embedded in a sentence. This recent research investigated the categorical perception in a sentence using an AXB paradigm with a stimulus interval of 300 ms: The relatively short interval may be one of the factors that suppressed the discrimination peak. To clarify this point, the current study conducted a perceptual experiment on Japanese listeners using an AXB paradigm with a stimulus interval of 1 s. The results suggest that for the perception of a Japanese /ba/-/da/ continuum by Japanese listeners, the length of the interval has only a small impact.