Neural Processing of Noise-Vocoded Speech Under Divided Attention: An fMRI-Machine Learning Study

Aug 3, 2026

Real-life speech perception often occurs under acoustic degradation and competing demands on attention, yet the neural interaction between degraded-speech processing and concurrent cognitive tasks remains poorly defined. This study used a dual-task paradigm combining fMRI with machine learning to characterize brain networks supporting noise-vocoded sentence perception during divided attention. Twenty-five participants performed a visuomotor recognition task while listening to four- and eight-band noise-vocoded speech, with machine learning identifying difficulty- and dual-task-related activations across frontotemporal regions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Abstract

Introduction

Background on degraded speech perception

Dual-task evidence and attentional effects

Noise vocoding and prior neuroimaging findings

Human Brain Mapping RESEARCH ARTICLE OPEN ACCESS Neural Processing of Noise-Vocoded Speech Under Divided Attention: An fMRI-Machine Learning Study Han Wang1,2 | Rongru Chen3 | Josef Schlittenlacher3 | Carolyn McGettigan3 | Stuart Rosen3 | Patti Adank3 1Clinical Systems Neuroscience Section, Department of Developmental Neurosciences, Great Ormond Street Institute of Child Health, University College London, London, UK | 2Department of Neurosurgery, Great Ormond Street Hospital for Children NHS Foundation Trust, London, UK | 3Department of Speech, Hearing and Phonetic Sciences, University College London, London, UK Correspondence: Han Wang ([han.wang.19@ucl.ac.uk](han.wang.19@ucl.ac.uk)) Received: 4 April 2025 | Revised: 16 July 2025 | Accepted: 25 July 2025 Funding: This work was supported by University College London (Grant No. 156630) . Keywords: divided attention | dual task | functional magnetic resonance imaging | machine learning | neuroimaging | noise-vocoded speech ABSTRACT In real-life interaction, we often need to communicate under challenging conditions, such as when speech is acoustically degraded. This issue is compounded by the fact that our attentional resources are often divided when we simultaneously need to engage in other tasks. The interaction between the perception of degraded speech and simultaneously performing additional cognitive tasks is poorly understood. Here, we combined a dual-task paradigm with functional magnetic resonance imaging (fMRI) and machine learning to establish the neural network supporting degraded speech perception under divided attention. We presented 25 human participants with noise-vocoded sentences while they engaged in a concurrent visuomotor recognition task, employing a factorial design that manipulated both speech degradation and task difficulty. Participants listened to eight-band (easier) and four-band (more difficult) noise-vocoded sentences, while the Gabor task featured two difficulty levels, determined by the angular discrepancy of the target. We employed a machine learning algorithm (Extreme Gradient Boosting, XGBoost) to evaluate the set of brain areas that showed activity predicting the difficulty of the speech and dual tasks. The results illustrated intelligibility-related responses in frontal and cingulate cortices and bilateral insulae induced by divided attention. Machine learning further revealed modality-general and specific responses to speech and visual inputs, in a set of frontotemporal regions reported for domain-general cognitive functions such as attentional control, motor function, and performance monitoring. These results suggest that the management of attentional resources during challenging speech perception recruits a bilateral operculofrontal network also associated with processing acoustically degraded speech. 1 | Introduction Everyday listening often involves processing acoustically degraded speech (e.g., a poor telephone signal), yet listeners can maintain successful recognition in such suboptimal scenarios (McGettigan et al. 2014; Shannon et al. 1995) . However, the cognitive functions (e.g., learning and attention; Carroll 1993) and their underlying neural substrates supporting degraded speech processing remain largely unexplored. In some cases, listeners must process such speech in distraction (e.g., chatting while driving), and recent studies using dual tasks show that degraded speech processing remains robust while attention is split between concurrent tasks (Gennari et al. 2018; Hunter and Pisoni 2018; Wang et al. 2023) . Nevertheless, speech perception becomes less accurate for a hard compared to an easy concurrent task (e.g., visual search among different number of distractors; Mattys et al. 2014), suggesting a role of attention in speech processing. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, pro

FAQs

What research question does the study address?

How the brain processes noise-vocoded (acoustically degraded) speech when attentional resources are divided by a concurrent task.

How were participants and tasks designed in the experiment?

Twenty-five participants listened to noise-vocoded sentences while performing a concurrent visuomotor recognition task; the design manipulated speech degradation and task difficulty.

What methods were used to analyze brain activity?

fMRI measured neural responses, and an explainable machine learning approach (XGBoost) evaluated which brain areas predicted speech difficulty and dual-task conditions.

Aug 3, 2026

Upload by Genevieve