Speaker: Mark Saddler

Lab: McDermott

Title: The role of temporal coding in everyday hearing: evidence from machine learning

Abstract: Neurons can encode information in the timing of their spikes in addition to their firing rates, but the role of these codes remains debated. The precision of spike timing is arguably greatest in the auditory nerve, whose action potentials are phase-locked to the fine-grained temporal structure of sound. We investigated the perceptual role of this precise temporal coding by optimizing machine learning models to perform everyday hearing tasks from simulated auditory nerve input. Our findings illustrate the utility of supervised machine learning for linking peripheral coding to real-world perception and clarify conditions in which prostheses that fail to restore high-fidelity temporal coding (e.g., cochlear implants) could in principle restore near-normal hearing.