Florine L. Bachmann1, Jens Hjortkjær1,2
Hearing Systems Section, Department of Health Technology, Technical University of Denmark, Kgs. Lyngby, Denmark
2Danish Research Centre for Magnetic Resonance, Centre for Functional Diagnostic Imaging and Research, Copenhagen University Hospital, Denmark

The application of EEG-based linearized encoding models for measuring neural responses to continuous speech has recently been extended from cortical to subcortical auditory processing. The subcortical response to continuous speech approximates the auditory brainstem response (ABR) to clicks and holds the potential to complement current clinical hearing assessments. This approach also enables the simultaneous assessment of subcortical and cortical responses, which could further shed light on the effects of age and hearing loss on the auditory system. However, for both clinical and research applications, it is crucial to investigate the reliability of such speech-derived subcortical responses in a clinical population. We therefore measured subcortical responses to continuous speech six times in a participant sample of 11 (5 female) older (68.45 ± 7.19) experienced bilateral hearing aid users with symmetric mild to moderate hearing loss. On different days, participants were presented with either identical or different speech materials to investigate consistency. Conventional click-ABRs and frequency-following responses (FFRs) were also assessed for comparison with the traditional methods. We discuss test-retest reliability of the latency and amplitude of wave-V components in the speech-ABR and its relation to conventional click-ABRs. We further discuss potential clinical applications, such as the evaluation of noise-reducing hearing assistive device algorithms or device fitting.