Brainstem correlates of cochlear nonlinearity measured via frequency-following responses (FFRs): A neural marker of “hidden hearing loss” or individual variation in central auditory processing?

Publication Date

February 2018

Document Type


Publication Title

Abstracts of the Annual MidWinter Research Meeting of the Association for Research in Otolaryngology



First Page


Last Page



Scalp-recorded frequency-following responses (FFRs) are EEG-based potentials used to characterize the pre-attentive, subcortical encoding of complex sounds. In the present study, we aimed to determine if FFRs carry information regarding cochlear health (nonlinearities) and putative diagnostic information on “hidden hearing loss” (HHL) that has heretofore remained elusive to measure in humans. Based on the premise that normal (healthy) cochlear transduction is characterized rectification and compression, we reasoned that these nonlinearities would create measurable harmonic distortion in the FFR spectrum in response to pure tone input. In normal hearing individuals, we compared conventional indices of cochlear nonlinearity, via distortion product otoacoustic emission (DPOAE) I/O function, to the total harmonic distortion measured from their FFRs (FFRTHD). FFRTHD quantifies the relative amplitude of response at harmonic (cochlear generated) distortion components to that at the stimulus frequency; HHL was predicted to produce a more linearized response resulting in FFRs at only the stimulus frequency and thus weaker FFRTHD. Physiologic measures were then compared to QuickSIN scores as poorer speech-in-noise perception in the presence of a normal audiogram has been suggested as an indicator of preclinical neurodegeneration and HHL. Analysis of DPOAE I/O functions and the FFRTHD revealed listeners with higher cochlear compression thresholds had lower neural distortion, (i.e., more linearized responses), thus linking cochlear and brainstem correlates of auditory nonlinearity. Importantly, FFRTHD was negatively correlated with QuickSIN scores whereby listeners with more nonlinear FFRs (higher FFRTHD) were better at perceiving speech in noise (i.e., lower QuickSIN scores). We infer that individual differences in SIN and FFR nonlinearity even in normal hearing individuals may reflect early HHL not captured by normal audiometric evaluation. Nevertheless, future studies in hearing impaired individuals and animal models are necessary to confirm diagnostic utility of FFRTHD and its relation to neural degeneration/synaptopathy in human listeners.