Neural and psychoacoustic insights into amplitude modulation processing and its link with speech perception in noise

Temporal modulations, especially amplitude modulations (AMs), play a fundamental role in speech perception in noisy environments, as highlighted by psychoacoustic research. However, the mechanisms underlying AM processing and its contribution to speech perception in noise remain unclear. The present study combined both behavioral psychophysics and electroencephalography (EEG) measures to investigate AM processing and its relationship to the perception of consonants in noise for young adults with normal hearing. Participants completed psychoacoustic measures of AM sensitivity, i.e., detection thresholds for AM fluctuations at 8 Hz, and susceptibility to AM masking (the effect on AM detection thresholds of interfering AM fluctuations carried by a masker) at 8 Hz. In addition, participants underwent an EEG recording measuring the neural AM following response (AMFR, or envelope following response) for AM tones modulated at 8 or 40 Hz presented without masking. Finally, they completed a consonant identification task in a stationary speech-shaped noise. Results showed a significant positive correlation between the signal-to-noise ratio of the AMFR recorded at 8 Hz and AM masking effects measured at the same rate. This finding is discussed in terms of both reduced selectivity of AM filters and elevated internal noise. Finally, no relationship between AM measures and consonant-in-noise perception was observed, suggesting that the mechanisms involved in AM processing may not fully capture the ability of adults with normal hearing to identify consonants in noisy backgrounds.NEW & NOTEWORTHY Results showed a significant correlation between the amplitude modulation (AM) following response (AMFR) and AM masking effect. Participants who are more susceptible to AM masking exhibit higher neural tracking of AM at the same fluctuation rate (8 Hz). This result might be explained both in terms of reduced selectivity of AM filters and elevated internal noise. Our results also showed no relationship between AM measures and consonant-in-noise perception, indicating that AM sensitivity mechanisms do not fully explain speech-in-noise perception skills in young normal-hearing adults.