EEG reflects subjective but not behavioral hearing in noise performance

Ever since the first description of the "cocktail party problem" in 1953, much effort and progress have been devoted to understanding the factors that contribute to hearing difficulties in noise. Hearing difficulties in noise are multifold, and as such have been assessed using a range of objective and subjective measures. Here, we explored three distinct measures of hearing in noise: tone-in-noise (TIN) detection, speech reception threshold (SRT) in noise, and the self-reported Speech, Spatial and Qualities of Hearing Scale (SSQ). We evaluated how these measures relate to auditory brainstem response, cortical electroencephalogram (EEG), behavioral (e.g., frequency modulation, FM detection), and self-reported factors. FM detection threshold emerged as a common factor across the predictive models for TIN, SRT, and SSQ, consistent with the possibility that temporal fine structure sensitivity is broadly related to different aspects of hearing in noise. In contrast, EEG P3 amplitude was only retained in the SSQ model and not in the TIN or SRT models, suggesting that cortical EEG measures may be more closely related to subjective than objective hearing in noise outcomes. Together, these findings suggest that different measures of hearing in noise capture partially distinct aspects of auditory function.