Vowel processing by a model of the auditory periphery: a comparison to eighth-nerve responses

Area of Science:

• Auditory Neuroscience
• Computational Auditory Neuroscience
• Signal Processing

Background:

• Peripheral auditory system models are crucial for understanding sound perception.
• Existing models often simplify complex biological processes.
• Accurate modeling requires integrating anatomical and physiological data.

Purpose of the Study:

• To develop and validate a computational model of peripheral auditory processing.
• To simulate the conversion of acoustic signals to auditory neuron activity.
• To compare model predictions with experimental neural population responses.

Main Methods:

• Developed a modular model simulating peripheral auditory pathways from eardrum to auditory neurons.
• Incorporated published algorithms and experimental data, assuming linear basilar membrane mechanics.
• Compared model outputs (average localized synchronized rate) to neural recordings for vowels in quiet and noise.

Main Results:

• The model accurately predicted vowel formant representations in average localized synchronized rate (ALSR) responses.
• It successfully captured saturating characteristics of normalized average rate responses in quiet.
• Modeled ALSR responses were less robust than neural data when vowels were presented in noise.

Conclusions:

• The developed model provides a valuable tool for studying peripheral auditory processing.
• Linear basilar membrane assumption impacts noise robustness, highlighting areas for future model refinement.
• The model shows promise in predicting neural responses, particularly for vowel perception in quiet conditions.