Lebert, Jules Marie; (2025) Neural mechanisms of listening in noise. Doctoral thesis (Ph.D), UCL (University College London.

Preview

Text Lebert_10206652_thesis.pdf Download (40MB) | Preview

Abstract

Auditory scene analysis is the process by which the brain separates and makes sense of complex acoustic environments, allowing focus on specific sounds amidst competing signals. This thesis investigates the neural mechanisms underlying auditory scene analysis and listening in noise, combining novel behavioural paradigms, electrophysiological recordings, and advanced data analysis techniques. The research focuses on how noise impacts neural encoding in the primary auditory cortex and develops a complex behavioural task to study auditory stream segregation in ferrets. Using a vowel discrimination task, this thesis demonstrates that background noise significantly impairs the encoding of vowel identity at both single-unit and population levels in ferret A1. Time-resolved decoding analysis reveals a dynamic adaptation to noise over stimulus presentation, with partial recovery of vowel encoding during sustained portions. Analysis of neural population dynamics shows that noise introduces structured changes in neural activity, increasing manifold dimensionality while concentrating variance along fewer principal components. Furthermore, this thesis introduces a novel behavioural paradigm, requiring ferrets to detect a target word within a stream of words presented concurrently with a non-target noise stream. Through machine learning analysis of behavioural data, an interaction is uncovered between the presence of a non-target stream and the timing of the target word relative to trial onset. This interaction suggests active stream segregation processes, with performance improving over the course of a trial as animals better separate target and background streams. Moreover, preliminary electrophysiological data obtained from simultaneous recordings from auditory and prefrontal cortices suggest that this task provides rich potential for exploring the neural correlates of selective attention. This research advances the understanding of auditory processing in complex acoustic environments and establishes new methodologies for investigating the neural basis of auditory scene analysis in animal models.

Download activity - last month

Export as XMLJSONCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item