A simple electrical lumped-element model simulates intra-cochlear sound pressures and cochlear impedance below 2 kHz

Abstract

Low-frequency sounds displace large parts of the basilar membrane (BM) and can have a modulating and possibly disturbing effect on hearing at other frequencies. A better understanding of the transfer of such sounds onto the BM is therefore desirable. Lumped-element models have previously been employed to determine the low-frequency acoustic properties of the cochlea. Although helpful in illustrating schematically the role of the helicotrema, BM compliance, and the round window on low-frequency hearing, these models, when applied quantitatively, have not been able to explain experimental data in detail. Building on these models, an extended electrical analog requires just 13 lumped elements to capture, in surprising detail, the physiologically determined frequency-dependence of intra-cochlear pressure and cochlear impedance between 10 Hz and 2 kHz. The model's verification is based on data from cat, guinea pig, and humans, who differ principally in their low-frequency cochlear acoustics. The modeling data suggest that damping within the helicotrema plays a less prominent role than previously assumed. A resonance feature, which is often observed experimentally near 150 Hz in these animals and near 50 Hz in humans, is presumably a phenomenon local to the apex and not the result of a standing wave between stapes and helicotrema.

The effect of the helicotrema on low-frequency loudness perception

“…This indicates that the hearing organ's sensitivity to very low frequencies might be underestimated by the standard. (6) The observed characteristics of iMETFs and ELCs (besides their difference in steepness) are explained by the model of the apical cochlea proposed by Marquardt and Hensel (2013). It suggests that the helicotrema is a dominant factor in shaping and decreasing perceptual sensitivity to LF-sounds, with individual differences presumably arising from cochlear anatomical differences.…”

“…As mentioned earlier, an important difference between ELC and iMETF is that the latter is measured using DPOAE, which are generated more basally, near the characteristic places of the primary tones in the second cochlear turn (Gaskill and Brown, 1996), whereas the behaviorally obtained ELC is dominated by the large vibrations in response to LF-sound at the most compliant, apical, end. If the resonance involves the fluid mass in the helicotrema (Marquardt and Hensel, 2013), one would therefore expect it to be more pronounced in the ELC than in the iMETF. This was not generally observed, probably because the BM response shows little frequency tuning to LF-tones, which excite therefore almost the entire BM (e.g., see modelling by Schick, 1994) so that the resonance feature is evident even in METFs obtained from the 1st turn of animal cochleae (Dallos, 1970;Nedzelnitsky, 1980).…”

“…Recently, Marquardt and Hensel (2013) showed that a simple lumped-element model of the apical cochlea can account for the physiologically observed METFs from both animals and humans, including the often observed nonmonotonic resonance feature. The abruptly increasing slope below the resonance was explained by the shunt impedance of the helicotrema.…”

Investigation of Mechanisms in Bone Conduction Hyperacusis With Third Window Pathologies Based on Model Predictions

“…This means that other AC pathways that may become important when the ossicular chain transmission is restricted is omitted in the current model. The model is similar to other lumped-element models of the inner ear where the stimulation is by AC ( 31 , 40 , 41 ). Moreover, the model could predict intra-cochlear sound pressures obtained experimentally in cadaveric temporal bones when the stimulation was a sound pressure in the ear canal ( Figures 3A,B ).…”

“…Based on the impedances of the scala vestibuli and scala tympani ducts ( Z SVD and Z STD , Figure 2B ), and the cochlear impedances in Figure 1 , all sections' contributions to U SV and U ST are computed and summed resulting in a final contribution of the volume velocity from U SV and U ST ( Figure 2B ). The impedance of the helicotrema ( Z H , Figures 1 , 2 ) is taken from Marquardt and Hensel ( 31 ). The volume velocity source of the vestibule ( U V ) is computed similar as in the previous model based on the length of the vestibule (5.8 mm) and an elliptic cross-sectional surface area (radius 1.55 and 2.45 mm).…”