The derived-band envelope following response and its sensitivity to sensorineural hearing deficits

Abstract

The envelope following response (EFR) has been proposed as a non-invasive marker of synaptopathy in animal models. However, its amplitude is affected by the spread of basilar-membrane excitation and other coexisting sensorineural hearing deficits. This study aims to (i) improve frequency specificity of the EFR by introducing a derived-band EFR (DBEFR) technique and (ii) investigate the effect of lifetime noise exposure, age and outer-hair-cell (OHC) damage on DBEFR magnitudes. Additionally, we adopt a modelling approach to validate the frequency-specificity of the DBEFR and test how different aspects of sensorineural hearing loss affect peripheral generators. The combined analysis of simulations and experimental data proposes that the DBEFRs extracted from the [2-6]-kHz frequency band is a sensitive and frequency-specific measure of synaptopathy in humans. Individual variability in DBEFR magnitudes among listeners with normal audiograms was explained by their self-reported amount of experienced lifetime noise-exposure and corresponded to amplitude variability predicted by synaptopathy. Older listeners consistently had reduced DBEFR magnitudes in comparison to young normal-hearing listeners, in correspondence to how age-induced synaptopathy affects EFRs and compromises temporal envelope encoding. Lastly, OHC damage was also seen to affect the DBEFR magnitude, hence this marker should be combined with a sensitive marker of OHC-damage to offer a differential diagnosis of synaptopathy in listeners with impaired audiograms.

Envelope reconstruction of speech and music highlights stronger tracking of speech at low frequencies

“…We focused particularly on reconstructions using the 2–16 Hz, 4–32 Hz, and 8–64 Hz models which produced the best reconstructions on average ( Fig 6C ). We then subtracted the spectra from null spectra generated by randomizing the phases in the reconstructions for each subject and averaging these randomized reconstructions (see [ 48 , 49 ]) ( Fig 6D ). From these adjusted spectral values we identified the peaks occurring at the tempo of the music (see Materials and Methods ) as well as 2 – 4x the tempo, since the peak energy may occur at multiples of the expected musical beat frequency of the music based on the acoustics [ 50 ] or neural activity following subcortical processing [ 51 ].…”

“…Then a null distribution of power spectral densities was created by shuffling the phases in each of the reconstructions, averaging the randomized reconstructions, and computing the power spectral density. This technique is based on methods to quantify magnitudes of peaks in frequency-following responses [ 48 , 49 ], where randomizing the phases of each signal and then averaging captures the spectra associated with the noise floor. 100 null spectra were computed for each reconstruction.…”

Towards personalized auditory models: predicting individual sensorineural-hearing-loss profiles from recorded human auditory physiology

“…Cochlear TL Model (Verhulst et al, 2018) Simulate DP amplitudes (sLDP ) of each pole function at given primary levels and frequencies, i.e., sLDP (f2, L2): However, removing all AN fibers from an IHC in the model would functionally correspond to IHC damage. The CF dependence of the AN population was considered in two steps: (1) Following the CF-dependent AN distribution observed in rhesus monkey (Valero et al, 2017;Keshishzadeh et al, 2020), we applied a non-uniform NH AN fiber population. (2) CF-specific AN-damage profiles were simulated.…”

“…cochlear synaptopathy (CS), (Kujawa and Liberman, 2009;Furman et al, 2013;Sergeyenko et al, 2013;Shaheen et al, 2015). However, applying the same AEP markers for CS diagnosis in humans has yielded mixed success, since AEP amplitudes can be affected by (i) other coexisting SNHL aspects such as outer-hair-cell (OHC) damage (Don and Eggermont, 1978;Gorga et al, 1985;Herdman and Stapells, 2003;Verhulst et al, 2016;Chen et al, 2008;Keshishzadeh et al, 2020) and (ii) subject-specific factors such as age, gender, and head-size (Trune et al, 1988;Mitchell et al, 1989;Hickox et al, 2017). Moreover, the sensitivity of AEPs to different degrees of OHC-loss and CS is unclear, and a direct quantification of AN fiber damage through histopathology is impossible in live humans .…”

“…However, similar to the ABR wave-V, EFR generators have latencies associated with IC processing (Purcell et al, 2004), thus differences in central auditory processing may reflect on the EFR magnitude to mask individual synaptopathy differences (Chambers et al, 2016;Möhrle et al, 2016;Parthasarathy et al, 2019a,b). To address these issues, relative EFR and ABR metrics were proposed in several studies to cancel out subject-specific factors and isolate the CS component of SNHL in listeners with coexisting OHC-loss: ABR wave-I amplitude growth as a function of stimulus intensity (Furman et al, 2013), ABR wave-I -V latency difference (Coats and Martin, 1977;Elberling and Parbo, 1987;Watson, 1996), the wave-V and I amplitude ratio (Gu et al, 2012;Schaette and McAlpine, 2011;Hickox and Liberman, 2014), EFR amplitude slope as a function of modulation depth Guest et al, 2018), the derived-band EFR (Keshishzadeh et al, 2020), or the combined use of the ABR wave-V and EFR (Vasilkov and Verhulst, 2019). While these relative metrics are promising, it is not known how OHC-loss and CS differentially impact AEPs.…”