Enhancing Speech Perception in Noise Through Home-Based Competing Talker Training.

Abstract

Objectives: This study aimed to evaluate the effectiveness of a competing talker training paradigm (2TT-Flemish). The primary objectives were the assessment of on-task learning and the transfer to untrained tasks.

Design: A total of 60 participants (54-84 years, mean age = 69.4) with speech-in-noise problems participated in the study. The study used a randomized controlled design with three groups: an immediate training group, a delayed training group, and an active control group. The immediate training group trained from the very beginning, while delayed training started after 4 weeks. The active control group listened to audiobooks for the first 4 weeks. All participants underwent 4 weeks of competing talker training. Outcome measures included speech perception in noise, analytical tasks (modulation detection and phoneme perception in noise), and inhibitory control. In addition, a listening-posture dual task assessed whether training freed up cognitive resources for a concurrently performed task. Finally, we assessed whether training induced self-reported benefits regarding hearing, listening effort, communication strategies, emotional consequences, knowledge, and acceptance of hearing loss. Outcome measures were assessed every 4 weeks over a 12-week period. The present study aimed to investigate the effectiveness of competing talker training in a stratified randomized controlled trial.

Results: Overall compliance to the training was good and increased with age. We observed on-task improvements during the 4 weeks of training in all groups. Results showed generalization toward speech-in-noise perception, persisting for at least 4 weeks after the end of training. No transfer toward more analytical tasks or inhibitory control was observed. Initial dual-task costs in postural control were reliably reduced after competing talker training suggesting a link between improved listening skills and cognitive resource allocation in multitask settings. Our results show that listeners report better knowledge about their hearing after training.

Conclusions: After training with the 2TT-Flemish, results showed on-task improvements and generalization toward speech-in-noise. Improvements did not generalize toward basic analytical tasks. Results suggest that competing talker training enables listeners to free up cognitive resources, which can be used for another concurrent task.