Human Frequency Following Responses to Vocoded Speech: Amplitude Modulation Versus Amplitude Plus Frequency...

Area of Science:

• Auditory Neuroscience
• Speech Processing
• Cochlear Implants

Background:

• Current cochlear implant (CI) strategies primarily encode amplitude modulation (AM), limiting speech perception.
• A novel strategy incorporating frequency modulation (FM) alongside AM has shown behavioral benefits in preliminary studies.
• Understanding the neural basis of these benefits is crucial for optimizing CI technology.

Purpose of the Study:

• To investigate the neural representation of speech processed with AM alone versus AM + FM strategies using human frequency following responses (FFRs).
• To determine if encoding FM improves the neural representation of temporal fine structure (TFS) and envelope periodicity.
• To compare the fidelity of neural encoding for different numbers of spectral channels.

Main Methods:

• Recorded FFRs from normal-hearing adults listening to synthetic vowel sounds (/au/) processed with 2, 4, 8, and 16-channel AM and AM + FM vocoders.
• Analyzed FFRs for neural representation of envelope periodicity (fundamental frequency, F0) and TFS (formant structure).
• Utilized temporal, spectral, and correlation analyses to assess neural encoding strength and fidelity.

Main Results:

• The AM + FM strategy maintained stronger neural representation of envelope periodicity (F0) as channel count increased, unlike AM alone.
• A clear spectral peak at the stimulus F0 was consistently observed for AM + FM, but not AM, stimuli.
• Neural representation of TFS and time-varying spectral information (formants) was significantly better with the AM + FM strategy.

Conclusions:

• The AM + FM processing strategy elicits more robust brainstem neural activity, preserving key speech information (periodicity, TFS, spectral dynamics) compared to AM alone.
• These findings support behavioral observations of improved speech, speaker, and tone recognition with the AM + FM strategy.
• FFRs are a valuable tool for evaluating and optimizing signal processing strategies for cochlear implants.