Effects of manipulating the amplitude of consonant noise portion on subcortical representation of voice onset time and voicing perception in stop consonants

Purpose: The purpose of this study was to investigate whether speech perception would reflect small latency changes in subcortical speech representation. Method: Twelve native Japanese listeners participated in the experiment. Those listeners participated in speech identification task and auditory brainstem response (ABR) measurement using /d/–/t/ continuum stimuli varying in voice onset time (VOT) with manipulation of the amplitude of initial noise (consonant) portion, the duration of which corresponded to VOT. Results: Increasing the noise portion amplitude lengthened subcortical representation of VOT, which is the latency difference between ABRs synchronizing to the onsets of initial noise and following periodic (vowel) portions (VOT _ABR ) and made listeners likely to perceive the stimuli with ambiguous VOT as a voiceless stop /t/. In addition, the amount of VOT _ABR lengthening was close to that of the VOT boundary shortening. Conclusion: A few milliseconds of difference in subcortical speech representation are important for the perception of speech sounds with ambiguous acoustic cues.