Speech coding in the auditory nerve: III. Voiceless fricative consonants

Area of Science:

• Neuroscience
• Auditory Perception
• Speech Processing

Background:

• Understanding how the auditory system processes complex speech sounds like fricatives is crucial for explaining auditory perception.
• Previous research has focused on vowel processing, but fricative consonant processing remains less understood.

Purpose of the Study:

• To investigate the neural encoding of synthetic voiceless fricative consonants in auditory-nerve fibers.
• To determine how auditory nerve fiber responses relate to the spectral characteristics of fricatives.
• To assess the effectiveness of neural processing schemes for extracting fricative formant frequencies.

Main Methods:

• Recording auditory-nerve fiber responses in anesthetized cats to synthetic fricatives (/x/, /s/, /ʃ/, /f/).
• Presenting stimuli at two sound pressure levels (60 and 75 dB SPL).
• Analyzing discharge patterns using peristimulus time (PST) histograms and their power spectra.

Main Results:

• Neural discharge rates were highest in characteristic-frequency (CF) regions corresponding to stimulus energy.
• Higher stimulus levels yielded more distinct discharge rate profiles at the onset.
• Power spectra of PST histograms showed components near CFs up to 3-4 kHz and low-frequency components.
• Formant frequencies generally did not align with major response components, except for those below 3 kHz.

Conclusions:

• Auditory nerve fiber responses encode spectral energy of fricatives, particularly at lower frequencies.
• Extracting formant frequencies from fricative stimuli using neural discharge patterns is difficult.
• Current neural processing models effective for vowels are insufficient for fricative formant extraction.