P17Session 1 (Monday 12 January 2026, 15:00-17:30)Divergent cortical speech tracking at the cocktail party: Preserved in hearing aid users, impaired in cochlear implant users
Background: Listeners with hearing impairment, including users of hearing aids (HAs) and cochlear implants (CIs), often struggle with speech comprehension in noisy backgrounds. Speech comprehension might be enhanced if the focus of attention was known and the corresponding speech signal could be enhanced. Recent research showed that the attentional focus can be recorded from EEG by considering the neural tracking of the speech envelope. However, this neural response has been predominantly studied in typical hearing subjects, leaving a critical gap in our understanding of how it functions in individuals with hearing loss.
Methods: We recorded EEG data while the participants attended to one of two competing talkers, a female and a male one, in a free-field acoustic environment. In contrast to prior studies, HA users as well as CI users used their personal, clinically-fitted devices. Cortical speech tracking was assessed through linear backward and forward models that related the EEG data to the speech envelope (1-8Hz). Moreover, we assessed speech comprehension using 3AFC multiple choice questions and subjective listening effort.
Results: Behaviorally, we found a significant increase in listening effort from TH to HA and further to CI users. Furthermore, comprehension scores were significantly decreased in CI users. Neurally, the backward model revealed that the envelope tracking of the HA and TH group did not differ significantly. As a result, the attention decoding accuracies in both groups were indistinguishable, reaching a maximal accuracy of 82% for 60s decision windows. In contrast, the CI group showed a profound deficit in attentional modulation, reaching a maximal decoding accuracy of 63% on a 60s decision window. The forward model further supported these findings. TRFs and encoding scores were largely similar between TH and HA users, whereas CI users’ impaired cortical speech tracking was reflected in weaker attentional modulation of both encoding scores and temporal response functions.
Conclusion: This study reveals divergent neural envelope tracking between HA and CI users in a challenging multi-speaker environment. These patient-specific differences underscore the need to tailor future neurofeedback in hearing instruments to the unique neural processing of each user group.