Human frequency-following responses: representation of second formant transitions in normal-hearing and hearing-impaired listeners.

Auditory nerve single-unit studies have demonstrated that phase-locking plays an important role in the neural encoding of the spectrum of speech-like sounds. Recently, it has been reported that the phase-locked activity underlying the scalp-recorded human frequency-following response (FFR) also encodes the first two formants of several steady-state vowels and the time-variant frequency presented in tonal sweeps. The purpose of this study was to determine (1) if FFR can encode the time-varying second formant transitions in synthetic stop consonant stimuli in normal-hearing and hearing-impaired listeners, (2) if hearing-impairment causes degradation of this neural representation, and (3) if the degraded representation is correlated with reduced consonant identification in hearing-impaired listeners. FFRs were obtained from normal-hearing and hearing-impaired listeners in response to several synthetic stop consonants. The results demonstrated that the FFR did encode the second formant transition in normal-hearing listeners. However, FFR encoding was severely degraded in most of the hearing-impaired listeners. Further, comparison of identification and FFR data for individual hearing-impaired listeners appears to suggest that degradation in the neural representation of the second formant transition may be accompanied by reduction in identification performance.