Auditory Mechanisms: Processes and Models

PREFACE.

COMMITTEES.

DATA DIPS AND PEAKS (WITH APOLOGIES TO ELLA FITZGERALD).

PREVIOUS PUBLICATIONS FROM THIS SERIES OF WORKSHOPS.

PHOTOLEGEND.

CONFERENCE PARTICIPANTS.

CONTENTS.

We recorded guinea-pig, basilar-membrane (BM) motion, and cat, single auditory-nerve-fiber (AN) responses to clicks, with and without electrical stimulation of medial-olivocochlear (MOC) efferents. In both BM and AN responses, MOC stimulation inhibited almost completely at low click levels. However at moderate-to-high click levels, MOC inhibition was small on the first half cycle and built up over many cycles in BM click responses, but was large on the first half cycle and negligable in the second cycle in AN click responses. The data support the hypothesis that OHCs produce or influence a motion which bends inner-hair-cell stereocilia and can be inhibited by MOC efferents, a motion that is present through most, or all, of the cochlea, but that is not apparent in basal-turn BM motion. These data, from normally-working cochleas, highlight the need to shift the conceptual paradigm for cochlear mechanics to one in which the classic BM traveling wave is not the only motion that excites AN fibers.

The mechanisms for deflecting inner hair cell (IHC) stereocilia have not been identified experimentally. Here, we describe a deflection mechanism which is capable of mechanically coupling somatic electromotility of the outer hair cells (OHCs) directly to the IHC stereocilia. The description is based on our recent discovery [1] that in response to intracochlear electrical stimulation, the apical surface of the IHC and the lower surface of the overlying tectorial membrane (TM) exhibit antiphasic motion of similar amplitudes for stimulus frequencies up to at least 3 kHz. This results in a pulsatile motion of the fluid surrounding the IHC stereocilia. Based on well-known physical principles of fluid flow between narrowly spaced elastic plates, we show that the fluid motion is amplified relative to that of the two boundary membranes and that this motion is capable of bending IHC stereocilia.

The high sensitivity of human hearing is believed to be achieved by cochlear amplification. The basis of this amplification is thought to be the motility of mammalian outer hair cells (OHCs), i.e., OHCs elongate and contract in response to acoustical stimulation. This motility is made possible by both the cytoskeleton beneath the OHC plasma membrane and the motor protein prestin distributed throughout the plasma membrane. However, these factors have not yet been fully clarified. In the present study, therefore, attempts were made to observe the ultrastructure of the cytoskeleton of guinea pig OHCs and to identify the motor protein prestin expressed in the plasma membrane of Chinese hamster ovary (CHO) cells by atomic force microscopy (AFM). Results indicate that the OHC cytoskeleton is comprised of circumferential actin filaments and spectrin cross-links and that particle-like structures with a diameter of 8–12 nm which exist in the plasma membrane of the prestin-expressing CHO cells are most likely be prestin.

A number of unusual and reversible changes take place in the cochlear mechanics with the administration of the drug furosemide. The purpose of the present paper is to understand and model the effect of furosemide on the cochlear mechanical response.

Outer hair cells (OHCs) have an intracellular turgor pressure, but maintain a cylindrical shape because of their elastic lateral wall cytoskeleton. In vitro, changing the osmolarity of the extracellular fluid produces changes in OHC morphology, compliance, and force production. We sought to determine the effects of changing perilymph osmolarity on cochlear function in vivo. After perfusing hypoosmotic perilymph (260 mOsm) through the guinea pig cochlea, compound action potential thresholds and distortion product otoacoustic emissions thresholds decreased. These effects reversed after washout with artificial perilymph of normal osmolarity (300mOsm). The opposite effects were seen when hyperosmotic perilymph (340 mOsm) was perfused. We then created a mathematical model of cochlear mechanics that included several of the unique nanoscale biophysical properties of the OHC, allowing us to simulate the effect of OHC turgor pressure on electromotility. The magnitude of the measured effects was consistent with the predictions of our mathematical model. These findings suggest that changing perilymph osmolarity modulates cochlear function by affecting OHC electromotility via changes in cell turgor pressure.

Mutations in Col11a2 and Tecta cause a decrease in hearing sensitivity in both humans and mouse models. To determine if these mutations also cause changes in the material properties of the tectorial membrane (TM), equilibrium stress-strain relations of TMs from normals and mutants were measured by applying osmotic pressure using polyethylene glycol (PEG). Results were consistent with a TM model having two components: a compliant component in which compression is resisted by electrostatic repulsion of TM macromolecules, and a component that is not compressed by PEG. While the mutations caused some change in the fixed charge concentration of the compliant components (in mmol/L: –9.9 for normals, –8.7 for Col11a2 –/–, and –5.4 for TectaY 1870C/+) the fractional volume of compliant component showed even larger changes (58% for normals, 42% for Col11a2 –/–, and 11% for TectaY 1870C/+). The simplest interpretation of these results is that the TM normally acts as a composite material with two functionally-different regions. Mutations of the TM alter the relative proportions of these two regions and thereby alter TM mechanics.

Two-tone distortion was measured in the intracochlear pressure in the basal turn of the gerbil cochlea, close to the sensory tissue, where the local motions and forces of the organ of Corti can be detected. The characteristics of the distortion reflect several stages of cochlear mechanics, including the nonlinear process that generates distortion, and single-tone cochlear tuning of both the primaries and the distortion products (DPs). In this contribution we present results that illustrate these processing stages. We show (i) the relatively large size of the 2f₁-f₂ component at its own best place, indicating that it traveled there, (ii) that the combined effects of nonlinear processing and cochlear tuning determine the shape of the family of DPs produced by one primary pair, (iii) that the tuning of an individual DP (e.g., 2f₁-f₂) reflects the single-tone tuning of both the primaries and the DP and (iv) that the DP measured at one longitudinal location is often composed of a sum of a local component and a component that emanated from a remote location.

The absence of knowledge of the mechanical response in the mid-frequency region of any species cochlea is due to the difficulty of accessing anything other than very basal or apical regions. The 6kHz region of the chinchilla cochlea was studied to determine whether the assumption that the transfer functions measured in the base are log scale invariant across the length of the cochlea. Measurements show that the 6 kHz region is at least as sensitive as the 9-20 kHz region. Input/output growth rates as low as 0.1 dB/dB. Two-tone suppression maps are consistent with those obtained at higher frequencies. Simultaneous measurements of DPOAEs and basilar membrane mechanics provides a basis for speculation as to OAE origin as either nonlinear (wave-fixed) or reflection (place-fixed). Growth of the 2f1-f2 distortion product as a function of the level of the primary tones generally supports the formula used for optimizing DPOAE production.

Driving point stiffnesses of the reticular lamina with its supporting structures and the tectorial membrane (TM) were determined with a piezoelectric sensor. Measurements were made at several radial positions and at four locations along the cochlea from base to apex. Furthermore, using a stiff probe, static images of the stepwise indentation of the reticular lamina (RL) were captured to monitor relative displacements of structures within the organ of Corti. Stiffness values at the RL approximately matched the stiffness values of the TM for each of the locations along the length of the cochlea. Reticular lamina moved like a rigid lever with its pivot point at the pillar cells' heads. Moreover, reticular lamina displacement was slightly greater than outer hair cell (OHC) or Deiters' cell cups displacement. While basilar membrane displecements were below the detection threshold of the system, the lower two thirds of the Deiters cells were compressed the most.

Otoacoustic emissions have been commonly believed to be generated in the cochlea and emitted through backward-traveling waves. A recent study (Ren, 2004), however showed that there is no detectable backward traveling wave and that the stapes vibrates earlier than the basilar membrane (BM) at the emission frequency. These findings indicate that a cochlearfluid compression wave is responsible for backward propagation of the emissions. This study contradicts with a widely accepted view that the delay of the otoacoustic emissions is approximately two times the forward traveling wave delay. In this study, the emission was measured in the ear canal, at the stapes, and at different locations on the BM. It was found that the slope of the phase-frequency curve measured from an apical location is always steeper than that measured from basal locations. Derived from the distance between two measured locations and their phase difference, the propagation velocity demonstrates that the BM vibration at the emission frequency propagates from base to apex through the observed region. Moreover, the emission group delay measured at the stapes is less than twice the traveling wave delay.

Sound-evoked responses of the basilar membrane are shown to be influenced by electrical stimulation of the medial olivocochlear efferent system. Both fast (τ~50ms) and slow (τ~10s) effects can be observed in the basal turn of the cochlea. Differences between the fast and slow effects imply that outer hair cells can influence basilar membrane motion in at least two ways. Differences between the effects observed on the basilar membrane and in the auditory nerve (as assayed using compound action potential recordings in the same cochleae) imply that outer hair cells influence more than just basilar membrane motion.

Nonlinearity in cochlear transduction is responsible for the amplification and compression in normal hearing. Distortion products generated from cochlear nonlinearity can be modulated with a bias tone, and the modulation patterns resemble the derivatives of the sigmoid-shaped transducer function. The resting position and inflection point of the cochlear transducer with optimal gain were indicated by quasi-static modulation patterns of even-order distortions. With a high-level bias tone, temporal modulation patterns revealed the dynamic behavior of cochlear transducer. Within one period of the bias tone, two typical modulation patterns formed a hysteresis loop. In force-displacement relation, the counterclockwise traversal of hysteresis represents energy gain. These results suggest that the nonlinearity of cochlear dynamics presents in four aspects: compression, suppression, distortion, and hysteresis.

The question of how the somatic electromotility of outer hair cells (OHCs) influences the amplification process in the cochlea is still unanswered. To investigate this, we measured the vibration patterns of the organ of Corti in response to intracochlear electrical stimulation on the reticular lamina (RL), basilar membrane (BM) and the overlying tectorial membrane (TM). Using a laser-Doppler-vibrometer, amplitude and phase responses at altogether 32 different positions from the inner sulcus cells to the Hensen's cells on the RL, BM and upper and lower surfaces of the TM were measured.

Low-pass filtered amplitudes were found at all positions. Additionally, a resonance and antiresonance were found in the basal turn on the RL and TM. The BM vibration pattern exhibited a CF-independent resonance in all measured turns. Phase was independent of radial position on the TM but not on the RL and BM. This results in a stimulus-dependent modulation of the width of the subtectorial space. A chloride-channel blocker (9-AC) was used as a control for the influence of the measured electromotility on the organ of Corti vibration. The results suggest that the somatic electromotility of the OHCs is capable of being coupled directly to the stereovilli of the inner hair cells (IHCs).

Changes in postspikes excitability of the auditory nerve fibers (ANF) (refractoriness and adaptations) are estimated by responses recovery functions (RF), receiving by the double pulse's method. The method was modified to share RF into stochastic and deterministic components. The simulation shows that the refractoriness has no influence on the fine temporal structure of ANF responses, when the stochastic component exceeds the deterministic one. The conditions appear when stimuli act either in isolation or in noise and when intensity approaches the threshold of the ANF.

We recorded the responses of auditory nerve (AN) fibers of cats to irregularly spaced tone complexes. Recordings from fibers with low characteristic frequencies (CF<4 kHz) enabled us to reconstruct the cochlear transfer characteristics over the entire range of frequencies that excite these low-CF fibers. In three cats, coverage of CFs was sufficiently dense to allow unambiguous unwrapping of phase across fibers. We determined amplitude and phase in the apex as a function of both CF and stimulus frequency. These panoramic maps of cochlear vibration are a physiological couterpart to Bekesy's measurements in cadaver cochleas.

There is evidence that intracellular chloride ions modulate prestin function. This study investigated how chloride ions (Cl^-) influence cochlear sensitivity and OHC electromotility. Lowering perilymph Cl^- concentration greatly reduced the magnitude of basal turn BM motion and the electrically evoked otoacoustic emissions. These effects could be modulated by tributyltin (TBT, a Cl^- ionophore). The results indicate that the electrochemical drive for Cl^- is important for normal activity of prestin.

The change of operating point (OP) of outer hair cell mechano-transduction could be determined from the change of the 2^nd harmonic of the cochlear microphonic (CM) following a calibration to determine its initial value. To perturb the OP, a constant force was applied to the bony shell of the cochlea using a blunt probe. orce applied over the scala tympani increased the OP. During constant force of the CM underwent a slow partial recovery toward the initial level. Removing the force again initiated a change of 2^nd harmonic, which returned to the control level. These data indicate an active mechanism for OHC transduction OP to dynamically be controlled at its normal position.

An optical coherence tomography (OCT) imaging system operating at 1310nm wavelength was used to image the organ of Corti. The spatial resolution of the system was ~13 μm. The reflectivity of light intensity for the organ of Corti was ~10^-5 ( a mirror defines a reflectivity of 1.00). Operating the system as a low-coherence interferometer allowed us to measure localized vibration of the basilar membrane in the organ of Corti, driven mechanically with a piezo stack coupled to the sample chamber. We were able to detect vibrational signal at 16kHz from three different locations in the organ of Corti. The chamber was vibrated by ±~1nm. The normalized detected signal (vibrational signal amplitude divided by the intensity of reflected light) from all three locations were approximately constant for equal amount of vibration.

Both an animal and mechano-acoustic model of superior semicircular canal dehiscence (SCD) were developed. The animal model demonstrates that middle-ear input admittance and stapes velocity increase and that cochlear potential decreases in response to sound after introducing an SCD. These changes are consistent with the 'third-window' hypothesis as illustrated by a mechano-acoustic model of the effects of SCD on audition.

There is a thin bony pedicle joining the lenticular plate to the rest of the long process of the incus. We have previously presented a brief review of its anatomy; new histological observations in cat; and a simplified finite-element model of the long process, pedicle, lenticular plate, incudostapedial joint and stapes head in the cat. Low-frequency simulations suggested that there may be more flexibility in the pedicle than in the incudostapedial joint itself. In this paper the modelling work is extended: a 3-D model for the cat is shown which has a more realistic geometry, and a model for the human is presented.

We describe extension of otoacoustic emission technique intended to directly register change in cochlear mechanical baseline. A two tone probe/masker experiment is described in human subjects. Instead of just calculating distortion products from the averaged responses, low frequency variation in ear canal pressure is obtained and the pattern of positive and negative summating/adaptive behaviour versus frequency and level of the masker is reminiscent of two-tone suppression contours. The technique largely eliminates middle ear considerations.

The mode of vibration of the stapes is predominantly piston-like but at higher frequencies, rotations about the long and short footplate axis are also observed. An experiment was performed to verify whether the non-piston components influence the pressure produced in the cochlea. First the pressure in the scala vestibuli behind the footplate was measured using a micro-pressure sensor while a microphone recorded the pressure produced by the sound source in the ear canal. Then the motion of the stapes was measured under different angles and all 3D stapes motion components were calculated. Piston motion and tilt of the footplate could thus be correlated with vestibular pressure in the same ear. With the present experimental data we can also directly calculate individual cochlear input impedances and find the width of the transmission band and the time delay in the middle ear.

The purpose of this study was to measure both middle ear stiffness and basilar membrane stiffness for the bottlenose dolphin (Tursiops truncatus) and compare these results with similar measures in other mammalian species. It was found that the point stiffness of the bottlenose dolphin basilar membrane has a gradient from 20 N/m near the base to 1.5 N/m near the apex and the middle ear has a stiffness of 1.37 × 10⁶ N/m. These values are considerably higher than those reported for most terrestrial mammals, yet consistent with species specialized for high-frequency hearing.

To delineate the cellular mechanisms underlying the cochlear active process, we have developed an active in vitro preparation of the cochlea from the clawed jird. The amplitude and phase of the active mechanical and electrical responses of this preparation accord with those obtained in vivo. Analysis of the resonant properties of the exposed cochlear segment discloses two principal modes of oscillation: a second-order mode whose resonant frequency is set by the bulk volumetric stiffness of the segment and the fluid mass loading it, and a less prominent traveling-wave mode whose resonant frequency more closely matches the best frequency in vivo.

We developed a simulation code in the Matlab environment for the study, using the Monte Carlo method, of cellular phenomena involving diffusion, buffering, extrusion and release of Ca²⁺. In particular we simulated the entry of Ca²⁺ at individual presynaptic active zones (hotspots) of auditory and vestibular hair cells, where Ca²⁺ plays a fundamental role in the transduction of mechanical stimuli, due to sound or acceleration, into electrical signals to be sent to the brain. The realistic reconstruction, in three dimensions, of the cellular boundaries and the derivation of the virtual fluorescence ratio ΔF/F₀ (equivalent to the one computed from fluorescence microscopy experiments) allowed us (i) to directly compare simulations to experimental data, (ii) to supply an estimate of the equivalent concentration of Ca²⁺ reactants (buffers) and (iii) to show how the mass action law hypothesis brakes down because of the local non equilibrium of the system.

Recent in vitro and in vivo data have drawn attention to the presence of high-frequency electro-mechanical resonances in the electrically evoked response of the cochlear partition and analogous resonances in isolated cochlear outer hair cells (OHCs). Resonances in isolated OHCs are similar to those present in damped piezoelectric structures and therefore it has been suggested that the behavior may result from the interplay of electro-mechanical potential energy and mechanical kinetic energy. In OHCs, the total potential energy includes both mechanical and electrical terms associated with the lateral wall while the kinetic energy accounts for the inertia of the moving fluids and tissues entrained by the moving plasma membrane. We applied first principles of physics to derive a model of OHC electromechanics consisting of an electrical cable equation directly coupled to a mechanical wave equation. The model accounts for the voltage-dependent capacitance observed in OHCs by means of a nonlinear piezoelectric coefficient. The model predicts the presence of electromechanical traveling waves that transmit power along the axis of the cell and underlie highfrequency resonance. Findings suggest that the subsurface cisterna (SCC) directs current from the transduction channels to the lateral wall and slows the phase velocity of the traveling wave. Results argue against the common assumption of space-clamp in OHCs under physiological or patch clamp conditions. We supplemented the traveling wave model with an empirical description of transduction current adaptation. Results indicate that the so-called RC paradox isn't paradoxical at all; rather the capacitance of the OHC may work in concert with transduction current adaptation and electro-mechanical wave propagation to achieve a relatively flat frequency response.

Outer hair cells in the cochlea have motility which directly uses electrical energy available at their plasma membrane. It has been shown that this motility can be reasonably explained by a simple two state model in which two states differ in charge and membrane area but not in the mechanical compliance. The model leads to a biphasic dependence of the axial stiffness analogous of gating compliance. However, the experimentally observed axial compliance monotonically increases with depolarization. Such observation appears to be explained by assuming that a large compliance of the state with smaller membrane area. However, such an assumption leads to incorrect tension dependence of the motor. It is found that this inconsistency is associated with the condition that increased membrane tension reverses the size of membrane areas of the two states. To avoid this paradox, the compliance of the state with smaller membrane area must decrease as membrane tension increases. That means that the axial compliance that is monotonic with respect to voltage can be predicted only if turgor pressure is less than 0.1 kPa, somewhat less than reported estimates.

The OHC motor, likely comprised of prestin and other associated proteins intrinsic to the cell's lateral membrane, presents sensitivity to chloride ions. We have been studying the effects of intra and extracellular chloride on many of the biophysical traits of the OHC motor through evaluations of the cell's nonlinear capacitance. Here we review some of our recent observations, including interactions between the motor's tension dependence and C1 flux through the lateral membrane. Additionally, we report on our efforts to estimate intracellular C1 in intact OHCs, and on our estimates of the motor's chloride sensitivity in intact OHCs. These data are helping us to understand how the cochlea amplifier is managed in vivo. Finally, we illustrate how prestin can be used to identify the presence of the environmental toxin tributyltin that can leach from toxin-treated marine structures, including sonar domes.

Fast adaptation, the rapid termination of transduction current that follows a mechanical stimulus, correlates temporally with hair-bundle force production that could boost bundle displacement in response to weak stimuli. We have presented data in favor of the "release model", which proposes that Ca²⁺ lengthens a mechanical linkage in the transduction apparatus, reducing gating-spring tension and allowing channels to close. To determine the molecule responsible for fast adaptation, we introduced the Y61G mutation into the Myo1c genomic locus using gene targeting. We show here that both fast and slow slipping adaptation are faster in Y61G knock-in hair cells as compared to C57BL/6 control cells. Although these results show that mechanical activity of myosin-1c is required for fast adaptation in vestibular hair cells, additional controls are required to ensure that the differences in adaptation rate are not due to strain differences between the two mouse lines.

Membranes show bidirectional energy conversion in that their mechanical strain and electrical polarization are coupled. Changes in transmembrane potential generate mechanical force and membrane deformation results in charge movement. The coefficients for the electromechanical transduction and the mechano-electrical transduction have the same magnitude, satisfying Maxwell reciprocity and suggesting a piezoelectric-like mechanism. Outer hair cell models that include its piezoelectric behavior indicate that charge movement occurs at frequencies that span the mammalian hearing range. Experiments have confirmed the modeling results, including the presence of resonances at high frequencies. Another set of experiments have measured electrically evoked pN forces in long (>10 μm) cylinders of cellular membrane having radii < 150 nm. Highly curved membrane cylinders are found in all hair cells, specifically in their stereocilia and in the fusion pore that forms when a synaptic vesicle joins the presynaptic membrane. High frequency membrane-based electro-mechanical conversion of the receptor potential may contribute to the stereocilia amplifier motor. It may also contribute to the synaptic amplifier required for assuring the temporal precision of neurotransmitter release. Evolutionary pressures for high frequency amplification may have led to the development of mammalian somatic electromotility through an expansion of both the structure and function of the stereocilia bundle motor down the lateral wall of the primal outer hair cell.

Electromotile length change and force generation by outer hair cells is thought to be essential for a sensitive and sharply tuned mammalian cochlea. Outer hair cell stiffness is in turn important for effective transmission of force to the organ of Corti. Maturation of stiffness in outer hair cells during development may therefore be an important factor in the onset and maturation of hearing. We compared the mechanical properties of developing and adult gerbil outer hair cells using calibrated glass fibers. We found that specific compliance of outer hair cells increased dramatically up to the onset of hearing and decreased after that. We examined F-actin, spectrin and prestin synthesis in developing and adult gerbil outer hair cells. While the F-actin content of the lateral wall decreased progressively during postnatal development, both spectrin and prestin increased close to the onset of hearing. These results strongly support the hypothesis that the cortical lattice and the prestin content of the lateral wall membrane influence outer hair cell mechanical properties.

The electromechanical properties of the Outer Hair Cell (OHC) have been reformulated in terms of acoustic variables. It is anticipated that the acoustic variable formulation will be more useful for incorporating OHC electromechanics into cochlear micromechanics. For guidance on the interdependency of the acoustic and electrical quantities and to aid physical intuition we also present piezoelectric circuit diagram for the OHC.

The cochlear outer hair cell (OHC) has a unique property of electromotility, which is critically important for the sensitivity and frequency selectivity during the mammalian hearing process. The underlying mechanism could be better understood by examining the force generated by the OHC as a feedback to vibration of the basilar membrane. In this study, we propose a model to analyze the effect of the constraints imposed on OHC on the cell's high-frequency active force generated in vitro and in vivo. The OHC is modeled as a viscoelastic and piezoelectric cylindrical shell coupled with viscous intracellular and extracellular fluids, and the constraint is represented by a spring with adjustable stiffness. We found that constrained OHC can achieve a much higher corner frequency than free OHC, depending on the stiffness of the constraint. We also analyzed cases in which the stiffness of the constraint was similar to that of the basilar membrane, reticular lamina, and tectorial membrane and found that the force per unit transmembrane potential generated by the OHC can be constant up to several tens of kHz.

The mechanical properties of cellular membranes can be studied by forming a long, thin, bilayer tube (a tether) from the membrane surface. Recent experiments on human embryonic kidney and outer hair cells (OHCs) have demonstrated that the force needed to maintain a tether at a given length depends upon the transmembrane potential. Since the OHC tether force is highly sensitive to the holding potential, these results suggest that the unique electromechanical properties of the OHC membrane contribute to the voltage response of the tether. Here we develop a theoretical framework to analyze how two proposed mechanisms of OHC electromotility, piezoelectricity and flexoelectricity, affect tether conformation. While both forms of coupling are predicted to lead to experimentaly observable changes in tether force, piezoelectric coupling is predicted to cause an increase in tether force with depolarization while flexoelectric coupling is predicted to lead to a decrease in force. The results of this analysis indicate tether experiments can provide insight into electromechanical behavior of the OHC membrane.

Construction of pseudotransducer functions in prestin knockout (KO) mice reveal that the outer hair cell (OHC) transducer appears to function normally, which allows one to study nonlinearities in mice that do not show frequency selectivity or amplification. Measurement of harmonic distortion indicates that the second harmonic exceeds the third in homozygotes and their controls. In addition, intermodulation distortion shows that the cubic difference tone (CDT), 2f1-f2, is ~20 dB down from f1 in both KO and wildtype (WT) mice. However, in contrast to controls where the cubic exceeds the quadratic difference tone (QDT, f2-f1), f2-f1 and 2f1-f2 are similar in magnitude for KO mice. Because KO mice also exhibit two-tone suppression, these results support the idea that cochlear nonlinearity persists in the absence of low thresholds and sharp tuning and that the hair cell transducer is the primary source of cochlear distortion.

Cells isolated from the organ of Corti were held in the microchamber configuration and their mechanical impedance was measured in a broad frequency range (0.48-50 kHz). This was achieved with a more sensitive version of a previously presented method employing magnetically actuated probes. Preliminary data suggest that axially loaded outer hair cells exhibit a purely viscoelastic impedance composed of a spring and a frequency dependent damper. The spring constant compares with known values.

Overexposure to intense sound damages outer hair cells (OHCs) and causes the loss of cochlear amplification, resulting in permanent hearing loss. However, OHCs are protected from such exposure by prior sublethal conditioning. Although the protective mechanisms is thought to be based on the modification of the cell structure which reduces the mechanical damage of cells, it is not clear whether the cell structure becomes more flexible or more rigid by such modification. In this study, therefore, the mechanical properties of OHCs in mice with/without heat stress were measured by atomic force microscopy (AFM). As a result, it was found that conditioning by heat stress causes an increase of Young's modulus of OHCs in mice 3–12 h after heat stress.

No abstract received.

No abstract received.

The cationic amphipath chlorpromazine (CPZ) is postulated to selectively partition into the inner leaflet of the plasma membranes and modulate the electromotile behavior of cochlear outer hair cells (OHCs). We used an optical tweezers system to characterize the mechanical properties of OHCs plasma membrane (PM) through the formation and analysis of membrane tethers in the presence and absence of CPZ. We observed characteristic force relaxation when the tethers were formed and maintained at a constant length for extended periods. This relaxation process was modeled using a 2^nd order Kelvin body that provided stiffness, membrane viscosity-related measurements, and relaxation time constants, which collectively indicated an overall biphasic nature of relaxation. Our results with CPZ strengthen the hypothesis linking the drug's effect to reducing the mechanical interaction between PM and cytoskeleton.

To clarify the mechanics of the cochlea, the investigation of the outer hair cell motility in the organ of Corti is crucial. In this study, a finite-element model of the organ of Corti of the gerbil cochlea including the OHCs was constructed, and the magnitude of the force generated by the OHC motility was estimated. Consequently, the maximum value of the force generated by the OHC motility was obtained to be 150 nN. The phase of the neural excitation relative to the basilar membrane motion was then estimated, and it was found that the OHC motility did not affect the phase of the neural excitation.

Calcium buffers help shape and localize cytoplasmic Ca²⁺ transients in excitable cells. We have measured the concentrations of calbindin-D28k, calretinin, parvalbumin-α and parvalbumin-β, four endogenous calcium-buffering proteins in rat cochlear hair cells during development. In hearing animals, the inner hair cells (IHCs) have a tenth of the calcium buffering capacity provided by these proteins compared with the outer hair cells (OHCs) where the cell body contains levels equivalent to more than 5 mM calcium-binding sites. Overall, buffer concentrations decrease in IHCs and increase in OHCs during cochlear maturation.

To understand and model the contribution of outer hair cells to cochlear function, a full knowledge of both forward and reverse transduction is needed. To determine the limitations imposed by forward transduction, we have characterized mechanotransducer (MET) currents and hair bundle mechanics of outer hair cells in isolated coils of neonatal rats. In response to step deflections of the hair bundle, MET currents activated and then rapidly adapted with a sub-millisecond time constant that depended on extracellular calcium and cochlear location. The adaptation time constant imposed a first order high-pass filter at a corner frequency that, when corrected to in vivo conditions, was similar to the cell's characteristic frequency. Hair bundle mechanics determined by stimulation with a flexible fiber, exhibited a pronounced non-linearity that developed with the same time course as fast adaptation. We propose that the MET channels will high-pass filter the input signal and provide positive mechanical feedback to augment motion of the tectorial membrane. The time course of the feedback varies with cochlear location.

Using a novel technique for rapid time-resolved confocal microscopy, we acquired image sequences showing the sound-evoked motion of inner and outer hair cell stereocilia bundles in the apical, low-frequency regions of the guinea pig cochlea. Motion of structures of interest was analyzed by optical flow computation. Sound stimulation at 80 - 100 dB SPL and 200 Hz led to deflection of both inner and outer hair cell stereocilia. The deflection was linearly related to the displacement of the reticular lamina. However, deflection was smaller for inner hair cell stereocilia. Phase differences were also found: deflection of inner hair cell stereocilia led that of the outer hair cell by 44 degrees on average. It was previously shown that apical inner hair cells have 10 – 16 dB larger AC receptor potentials than outer hair cells (Russell & Sellick, 1983; Dallos, 1985). Our results suggest that inner hair cells are equipped with transducer channels of higher conductance or higher density than are the outer hair cells. In view of the tiny deflection that inner hair cells are supposed to detect, it might also be that active processes acting at the level of their stereocilia are necessary near threshold.

Cochlear outer hair cells (OHCs) are involved in a mechanical feedback loop in which the fast somatic motility of OHCs is required for cochlear amplification. Alternatively, amplification is thought to arise from active hair bundle movements similar to that in non-mammalian hair cells. We measured the voltage-evoked hair bundle motions in the gerbil cochlea to determine if such movements are also present in mammalian OHCs. The OHCs displayed a large hair bundle movement that was not based on mechanotransducer channels but originated in somatic motility. Significantly, bundle movements were able to generate radial motion of the tectorial membrane in situ. This result implies that the motility-associated hair bundle motion may be part of the cochlear amplifier.

Using a novel experimental technique that combines optical trapping with patch-clamp and fluorescence photometry, we provide preliminary evidence that native biological membranes are capable of electrically-induced piconewton level force generation in the absence of specialized transmembrane proteins such as prestin. Force generation is dependent on membrane tension and the transmembrane electrical potential. Salicylate diminishes and prestin enhances force generation.

The mechanically gated transduction channels of vertebrate hair cells tend to close in ~1 ms following their activation by hair bundle deflection. This fast adaptation is correlated with a quick negative movement of the bundle (a "twitch"), which can exert force and may mediate an active mechanical amplification of sound stimuli in hearing organs. We used an optical trap to deflect bullfrog hair bundles and to measure bundle movement while controlling Ca²⁺ entry with voltage clamp. The twitch elicited by repolarization of the cell varied with force applied to the bundle, going to zero where channels were all open or closed. The force dependence is quantitatively consistent with a model in which a Ca²⁺-bound channel requires ~3 pN more force to open, and rules out other models for the site of Ca²⁺ action.

Propagation of stimuli across the stereocilia within a hair bundle affects the gating of transduction channels. Tip links and lateral links are the two most probable candidates in providing the mechanical connection between the stereocilia. To distinguish between the two we measured the movement of individual stereocilia when pulling on the tallest stereocilium of a bundle. Hair cells were isolated and their hair bundles were displaced using a glass pipette attached to the kinocilium and driven by a piezoelectric bimorph. The stimuli were sinusoids with frequencies at 20 Hz and 700 Hz. The motion of the bundle was visualized using stroboscopic video microscopy and was quantified using cross correlation methods. Our data suggest that the bundle moves as a unit and that adjacent stereocilia bend at their bases and touch at their tips. We argue that this motion is consistent with the lateral links being involved in the propagation of stimuli across the bundle and with transduction channels being mechanically in parallel.

Mechanically activated transducer channels in cochlear outer hair cells (OHC's) transducer sound encoded mechanical signals into electrical signals. Entry of extracellular Ca²⁺ through these channels modulates transduction by reducing their open probability, a phenomenon called adaptation [1]. Analysis of the mechanical and electrical characteristics of the transducer channels in OHC's has shown that the transducer channel's open probability can be adequately described by a differentially activating two-state model [2], Also a direct relationship was demonstrated between the gating spring stiffness (K_s) and the accuracy (σ_min= 2kT/[K_s·D]= 5.4 nm, where D is the distance between the engaging positions of the closed and open conformational state) with which hair bundle position can be detected as a result of intrinsic channel stochastics. In combination with an assumed Ca²⁺-dependent gating spring stiffness [e.g. 3], we predict on the basis of the two-state model that at endolymphatic Ca²⁺ concentrations (~ 20 μM) an improved accuracy (σ_min ~ 3 nm) can be attained at the equilibrium position of the hair bundle.

Although experimental and theoretical information about intracellular concentration of Ca²⁺ in the stereocilia of lower vertebrates is available, there is only few information about mammalian systems. The aim of the present experiments was to investigate the origin of mechanically evoked Ca²⁺ signals in the hair bundle of outer hair cells (OHC).

The Namib Desert golden mole, Eremitalpa granti namibensis, is a nocturnal, surface-foraging mammal, possessing a massively hypertrophied malleus which presumably confers low-frequency, substrate-vibration sensitivity through inertial bone conduction. When foraging, E. g. namibensis typically moves between sand mounds topped with dune grass which contain most of the living biomass in the Namib Desert. We have observed that foraging trail segments between visited mounds appear remarkably straight, suggesting sensory-guided foraging behavior. Foraging trails are punctuated with characteristic sand disturbances in which the animal "head dips" under the sand. The function of this behavior is not known but it is thought that it may be used to obtain a seismic "fix" on the next mound to be visited. Geophone recordings on the mounds reveal spectral peaks centered at ca. 300 Hz ca. 15 dB greater in amplitude than those from the flats. Seismic playback experiments suggest that in the absence of olfactory cues, golden moles are able to locate food sources solely using vibrations generated by the wind blowing the dune grass on the mounds. Morever, the mallei of the golden moles in the genera Chrysochloris and Eremitalpa are massively hypertrophied. In fact, out of the 117 species for which data are available, these golden moles have the greatest ossicular mass relative to body size (Mason, 2001). Laser Doppler vibrometric measurements of the malleus head in response to seismic stimuli reveals peak sensitivity to frequencies below 300 Hz. Functionally, they appear to be low-frequency specialists, and it is likely that golden moles hear through substrate conduction (Supported by NIH Grant DC00222).

DPOAE amplitude variations with frequency can be due to interference between place and wave fixed components. When these components are separated other structure remains on a scale of one octave for wave-fixed DPOAE and on a scale of approx 1/5 octave or 400Hz at 3kHz in place-fixed DPOAE. Quasi-periodic peaks and valleys occur in both 2f₁-f₂ and 2f₂-f₁ place-fixed emissions at specific DPOAE frequencies irrespective of the ratio of f2/f1 and hence irrespective of the stimulus configuration on the basilar membrane. We present data and statistics on this structure from 12 human subjects and discuss its origin. Various hypotheses for the structure are discussed and assessed against the data including; a second DPOAE place fixed source, basal reflection standing waves, periodicity in cochlear refection and coherent reflection filtering. The experimental evidence best supports a coherent reflection filtering origin.

To help elucidate how otoacoustic emissions (OAEs) originate at and propagate from certain cochlear sites to the middle ear, we extend a previous theoretical investigation [1]. Using simplified 1D and 2D linear cochlea models, we try to explain the rationale that stands behind our key assumption, namely the instantaneous hydrodynamic coupling between stapes and basilar membrane (BM) and among BM portions. We argue that, because of the mathematical peculiarities of the undamped cochlea model and of doubtful implications about group velocity in the damped case, a physical interpretation of the WKB method is far from being clear and conclusive. By contrast, a faithful representation of fluid coupling in terms of Green's functions reveals an instantaneous longitudinal interaction between BM segments, which can be characterized as a sort of space delayed damping. To exemplify the range of our view, a frequency–domain simulation of the mechanisms underlying the generation of distortion products in a nonlinear hydrodynamic cochlea model is reported.

Otoacoustic emissions provide unambiguous evidence that the cochlea supports energy propagation both towards, and away from, the stapes. It is generally accepted that energy propagation away from the stapes involves a traveling wave mechanism. The mechanism by which energy propagates back to the stapes remains controversial. Examination of the mode by which energy propagates back to the stapes has been done by interpreting otoacoustic emission delay times and comparing these delay times with basilar membrane measured cochlear delay times. However, cochlear delay times inferred from basilar membrane measurements represent a measurement from a spatially confined region on the basilar membrane that is fixed in position whereas otoacoustic emissions generally represent measurement from a more spatially distributed region of the cochlea. Further complicating matters, otoacoustic emissions appear to have a complex origin that confounds estimates of cochlear delay based on the phase derivative. Here, signal onset delay is reported for otoacoustic emissions arising from different mechanisms (SFOAE versus DPOAE), matched for stimulus frequency. Comparison of signal onset delay for the two emissions argues for a bi-directional traveling wave mechanism i.e., energy propagates back to the stapes as a traveling wave rather than an acoustic compression wave.

By comparing the range of emission frequencies with that of neural characteristic frequencies of the amphibian and basilar papillae, the emission generation site may be inferred. Spontaneous otoacoustic emissions in the amphibian ear seem to originate from the amphibian papilla. In contrast, distortion product otoacoustic emission are presumably generated by both the amphibian and the basilar papillae. Distortion products from the amphibian papilla are very sensitive to ischemia; distortion products from the basilar papilla are less sensitive. These results suggest that the basilar papilla may not include an active amplifier. In support of this hypothesis, we show that distortion products from the basilar papilla show only a weak temperature dependence. These emissions are possibly independent of metabolic rate. The basilar papilla in frogs may be the only passive vertebrate hearing organ. In contrast, emissions from the amphibian papilla are clearly temperature dependent, consistent with active auditory processing.

The complex deafness locus DFNB1 contains GJB2, the gene encoding connexin 26 (Cx26) and GJB6, encoding Cx30, the two most abundant connexins in the inner ear (Petit et al., 2001). These connexins may form heteromeric/heterotypic channels in the gap junctions that interconnect cochlear supporting cells. By showing that a specific defect of Cx26 affects metabolic coupling mediated by IP₃ we have recently offered a mechanistic explanation for the pathogenesis of deafness due to connexin mutations [1]. Abnormal or impaired connexin function has been linked to several other diseases, including skin disease, peripheral neuropathies, and cataracts [7], thus our data may have a more general impact. Gap junction blockade impairs the spreading of Ca²⁺ waves and the formation of a functional (glial-like) syncytium in cochlear supporting cells. Wave propagation necessitates also a regenerative mechanism mediated by P2Y receptors [6] and this may constitute a fundamental mechanism by which supporting cells co-ordinate their responses following activation of sensory hair cells by sound.

Transiently evoked otoacoustic emissions - TEOAEs (tones and broadband stimuli) - were analyzed by means of an adaptive approximations method based on the Matching Pursuit (MP) algorithm. This method is an iterative, nonlinear procedure which decomposes a signal into a sum of known waveforms of well defined frequencies, latencies, time-spans and amplitudes. It provides high resolution energy distributions in the time-frequency space. The MP method allows for direct and accurate determination of component latency, since that is one of the parameters returned by the procedure. It was found that the parameters of components identified by the MP method - amplitude, latency and time-span - are affected in case of hearing disturbance caused by noise. The MP method made it possible to identify the resonant modes that appear for a given subject, with the same frequencies and latencies for different stimulation frequencies. The same resonant modes were also identified in responses to click stimuli. The method of adaptive approximations opens new possibilities in the field of investigation of hearing mechanisms and offers new tools for diagnosis of hearing disturbances.

The effects of changes in primary level on DPOAE are evaluated using frequency-modulated primaries (log frequency sweeps), which maintain a constant frequency ratio. We use 8s/octave sweeps to evaluate the DPOAE fine structure, and 2s/octave sweeps to evaluate the generator component alone. Using this procedure we have obtained data over a wide range of levels in one session, permitting evaluation of changes in the relative level of components with level. The fine structure spacing and phase of DPOAE at higher primary levels are consistent with the development of a nonlinear reflection component from the distortion product region as hypothesized in Talmadge et al., (2000).

While studied extensively since their discovery by Kemp in the late seventies, the cellular basis of the phenomenon of otoacoustic emission remains unknown. Data from experiments in humans, chinchillas and Mongolian gerbils was used to test the hypothesis that otoacoustic emissions originate in the hair cell transduction apparatus. Specifically, a double Boltzmann model of the transducer predicts that emissions generated by a single tone (stimulus frequency otoacoustic emissions - SFOAE) should be measurable at stimulus levels 20 or more dB below neural threshold, but sufficient to modulate the activity of enough transduction channels to produce a macroscopically observable result. On the other hand, for a fixed low-level probe tone that evokes SFOAE, it should only be possible to demonstrate the presence of emission by using a suppressor tone large enough to drive the transducer into its nonlinear range, approximately where the suppressor level reaches neural threshold. This result should be independent of suppressor frequency. Both predictions were confirmed experimentally in all three species. The threshold suppressor level was consistently near the threshold of the compound neural response monitored with an extracochlear electrode, even for suppressors more than an octave higher than the frequency of a low-level (30 dB SPL) probe tone. Cochlear microphonic responses were always detected at the lowest levels demonstrating SFOAE. The hair cell transducer appears to be the site of interaction between the probe and suppressor tones for all suppressor frequencies, consistent with a single suppression mechanism. Nonlinear interactions demonstrated in SFOAE and CM between widely separated tones do not appear to have a correlate in the basilar membrane, suggesting that, at least under some conditions, pressure waves can be initiated directly from forces produced by the hair bundle.

Shera [1] proposed that pressure effects on the middle ear provide a model for distinguishing between a point-source and a global standing-wave model of SOAE generation. A point source is supposed to be insensitive to changes in the boundary conditions for oscillation, whereas a standing wave would be influenced. Changing middle-ear pressure in humans alters both frequency and amplitude of SOAE, supporting Shera's assumption that mammalian SOAE originate through global standing waves.

Lizards are highly reliable generators of SOAE, but their hearing organ differs from that of mammals in size, structure and micromechanics. Thus they provide a good system in which to continue to examine ideas about the generation of spontaneous emissions. In lizards, both negative and positive pressure changes were produced in the ear canal by adding or withdrawing air. Increases in pressure led to no or only small changes in frequency and amplitude, whereas pressure drops led to a fall or rise in SOAE frequency of up to several percent and to amplitude loss. These changes were observed over much smaller pressure ranges than those necessary in humans. The question is discussed as to whether such data permit a clear distinction of the nature of the emission source.

The non-uniform, but highly precise patterns of hair-cell orientation in the basilar papilla of birds makes this organ an attractive model for studying the developmental mechanisms that determine hair-cell polarity. We show here that, consistent with earlier observations on the chicken, the final maturation of the hair-cell orientation pattern in the basilar papilla of the barn owl, an altricial bird, occurs late in development, after the onset of hearing. This raises the question whether hair-cell polarity is entirely governed by internal signals or whether normal stimulation and function of hair cells might be necessary for final adjustment.

Transient evoked otoacoustic emissions (TEOAE) are simulated by a cochlear model with embedded outer hair cells (OHC). TEOAE were produced due to the nonuniform behavior of the OHC gain. Normal audiograms are related with TEOAE. Abnormal audiograms are obtained with increasing nonuniformity and TEOAE abolishment.

We measured click-evoked and stimulus-frequency otoacoustic emission input-output transfer functions (T_CE and T_SF) over a broad range of stimulus intensities in humans. T_CE and T_SF are similar in overall magnitude, spectral structure, and phase at all intensities studied. The strong similarity between T_CE and T_SF supports the hypothesis [6,4] that human CEOAEs and SFOAEs are generated by the same mechanism.

Models of otoacoustic emission (OAE) generation mechanisms often attribute important features of OAEs to waves traveling along the cochlear partition. Since the lizard basilar papilla manifests no obvious analog of the mammalian traveling wave, detailed characterization of lizard OAEs offers an important opportunity to test and extend our knowledge of emission mechanisms. We report otoacoustic measurements (DPOAEs and SFOAEs) in the ears of adult leopard geckos (Eublepharis macularius) and humans. We compare and contrast the properties of gecko and human OAEs and discuss their implications for mechanisms of OAE generation.

In the years 1970-1980 it gradually became known that the live cochlea shows, for high frequencies and low stimulation levels, a mechanical response that in its peak is as frequency-selective as primary auditory nerve fibers, and demonstrates a comparable degree of nonlinearity. It proved impossible to simulate these properties, in particular the selectivity, with a "normal", i.e., "passive" cochlear model. Accordingly, "active" models of cochlear mechanics were developed. In most of these the basilar membrane (in fact, the organ of Corti) is assumed to be capable of augmenting (amplifying) the power of the cochlear wave. This must occur "locally", i.e., over a restricted region of the length of the basilar membrane (BM), while in other regions the generated power is dissipated, thus rendering the model stable. The power-amplifying elements are generally believed to be outer hair cells (OHCs). Nonlinear transduction in these cells can explain many, possibly all, nonlinearities found in BM responses. With the "inverse solution" method as developed by this author, local activity can be quantitatively determined and analyzed from experimentally obtained BM response data – a most fruitful interaction between experiment and theory. Nonlinearity can be studied as well with this method.

Cochlear activity can be viewed from many angles, from many perspectives. A number of these are described in this paper. More questions can be formulated and only few of them can be addressed in this paper. A most fundamental question remains: How does the "local" character of activity arise? Several theories have been put forward. One involves a secondary resonance in the organ of Corti. It will be suggested that this leads to a conflict with data on impulse responses. Spatial integration (including both feed-forward and feed-backward) provides another possibility, but this type of model does not seem universal. In summary, only few of our perspectives in cochlear activity have yet reached their horizons.

The coupling of the mechanical and the electrical response of the cochlea is well known. Acoustical stimulation gives rise to the cochlear microphonic while electrical stimulation of the cochlea elicits both basilar membrane motion and otoacoustic emissions. Disruption of the resting electrical environment, through efferent stimulation, artificial injection of current, and a variety of other means is known to affect hearing sensitivity and the mechanical response of the cochlea to stimulus. The key missing element in most models is the explicit coupling of the electrical domain to the mechanical degrees of freedom. By modeling this coupling, predictions of both mechanical forces and transducer currents may be made enabling comparisons and analysis of electro-physiological experiments. A mechanical–electrical–acoustic model of the cochlea is presented whose key components are micro–electro–mechanical coupling of the cochlear structures, a two-duct acoustic model with structural–acoustic coupling at the basilar membrane (BM), and a global electrical circuit to model conductances in the different scalae. Predictions of the cochlear microphonic, other cochlear potentials, BM velocity and otoacoustic emissions in response to pure acoustic input and bipolar electrical stimulation are presented. Model simulations show that including three dimensional fluid effects improves BM response characteristics. A tectorial membrane shear mode resonance is shown to provide amplification to the BM at the characteristic frequency of a location. This project is funded by NIH NIDCD R01 - 04084.

Surface gravity waves in a spiral channel can be used as an analogue for cochlear macromechanics (Manoussaki et al. [1]). We found that in a vertical-walled channel with uniform cross section, as a low-frequency wave propagates inward from larger to smaller spiral radii, the wave amplitude near the outside wall grows while the amplitude near the inside wall decreases. This relative amplitude change induces a radial tilt of the free surface, the magnitude of which increases in inverse proportion to the spiral radius. The tilt, which can be interpreted in terms of energy redistribution, can be explained by a "whispering gallery effect," can develop dynamically on the cochlear partition and, by contributing to the bending of apical stereocilia (Cai et al. [2]), can augment low-frequency hearing by as much as 20 dB. We therefore hypothesized that cochlear spiral radii ratios (largest/smallest) could account for interspecies differences in low-frequency hearing. Preliminary analyses of spiral parameters obtained from published data on mammalian species, as well as from histological sections and 3D CT scans of the cochleae of low-frequency baleen whales and high-frequency dolphins and porpoises, tend to support our hypothesis: species with good low-frequency hearing have larger spiral ratios than do those with poor low-frequency hearing.

The capability for the accurate and efficient computation of the three-dimensional elastic features of the organ of Corti has been available for some time. Our recent work has been on the inclusion of the viscous fluid. Novel measurements from various laboratories provide the opportunity to refocus on the elastic properties. The current detailed model for the organ of Corti is reasonably consistent with these diverse measurements in guinea pig. The individual rows of IHC cilia with tip links and the Hensen stripe are included. The results for low frequency show a phase of tip link tension similar to auditory nerve measurements. For high amplitudes for the guinea pig base, because of the near contact with the Hensen stripe, the excitation changes polarity, similar to the peak-splitting neural behavior sometimes observed.

Recent experimental evidence has demonstrated that somatic outer hair cell (OHC) motility is important for amplification in the mammalian cochlea [1,2]. However, under the 'somatic electromotility' theory, the transmembrane potential that is responsible for driving the somatic OHC force is subject to low-pass filtering by the electrical RC time constant of the OHC membrane [3], Numerous mechanisms have been proposed to compensate for the attenuation of the membrane potential by the low membrane time constant at high frequencies [3,4-10]. We present a micromechanical model derived from an engineering-based analysis of cochlear mechanics and experimental data. Our model does not require novel compensatory mechanisms and demonstrates that adequate OHC gain with negative feedback significantly extends closed-loop system bandwidth and increases resonant gain. The OHC gain-bandwidth product, not just bandwidth, determines if high-frequency amplification is possible. Thus, fast cochlear amplification is possible with slow OHCs simply due to in situ feedback dynamics, though our model does not preclude other compensatory mechanisms.

The hydrodynamic box model of the cochlea is revisited for the purpose of studying in detail the approximate scaling law that governs the tonotopic arrangement of its frequency–domain solutions. The law differs significantly from that derived by Sondhi in 1978, commonly known as "approximate shift-invariance", which suffers from an inaccuracy in the representation of the hydrodynamic coupling. Despite the absence of a similar scaling law in real mammalian cochleas, the results here presented may be significant in the perspective that a covariance law of a more general type should hold for real cochleas. To support this possibility, an argument related to the problem of cochlear amplifier-gain stabilization is advanced.

No abstract received.

A major goal in cochlear modeling is to account for the functional mechanism of the cochlear amplifier (CA). Although numerous hypotheses have been presented, many are based on addons to the fundamental one-dimensional model, which assumes only the basilar membrane between two fluid-filled channels, scala vestibuli and scala tympani. Another class of models assumes more than two wave propagation channels (modes), and we call them multicompartmental models, a concept that originated with de Boer [1, 2, and 3]. Using a multicompartmental formulation, we put forward the hypothesis that the CA function is due to a combination of forces on the reticular lamina and the basilar membrane coming from both local hair cells and from a pressure wave that propagates in the fluid-filled spaces between the reticular lamina and the basilar membrane. A generic version of the model has been used to match data from various species. An improved model with parameters based on anatomical data from the gerbil can better match physiological data from the gerbil. A more advanced model that separates arcuate and pectinate regions of the basilar membrane shows the phase angle of the response of the arcuate region to low-frequency probe tones reverses at about midway down the cochlea. Overall, the models provide an explanation of how the CA might work.

We have recently proposed that the cochlear amplifier is a fluid pump driven by outer hair cell (OHC) somatic motility [1,2]. According to this hypothesis, the OHCs pump fluid into the tunnel of Corti (TOC) creating a second type of traveling wave that we call the organ of Corti (OC) wave. It is the OC-wave and not the classical traveling wave that is amplified by the OHCs according to the fluid-pump hypothesis. The question remains, however, how does the motion of the OC-wave get coupled to the inner hair cell (IHC) stereocilia? We hypothesize that the organ of Corti pressure distends the tissue in the IHC region leading to deflection of the IHC hair bundle. This hypothesis is supported by the observation that low-frequency IHC receptor potentials can be quite distorted and that the onset of distortion correlates with saturation of the OHC receptor current. We present a model based on the fluid pump hypothesis that replicates many features of the experimentally observed distortion in the IHC receptor potential.

The ear relies on nonlinear amplification to enhance its sensitivity and frequency selectivity. In the bullfrog's sacculus, a hair cell can mobilize active oscillatory movements of its hair bundle to amplify its response to faint stimuli. Hair-bundle oscillations can result from an interplay between a region of negative stiffness in the bundle's force-displacement relation and the Ca²⁺-regulated activity of molecular motors. Within the framework of this simple model, we calculate a state diagram which describes the possible dynamical states of the hair bundle in the absence of fluctuations. Taking different sources of fluctuations into account, we find conditions that yield response functions and spontaneous noisy movements of the hair bundle in quantitative agreement with experiments. We show that fluctuations restrict the bundle's sensitivity and frequency selectivity but find that a hair bundle studied experimentally operates near an optimum of mechanosensitivity in our state diagram.

Waves propagating along the basilar membrane are amplified by an active nonlinear process. The general aspects of the active amplification of periodic signals can be discussed in the framework of critical oscillators. Here, we show how the concepts of a traveling wave and of critical oscillators can be combined to describe the main features of nonlinear wave propagation, energy flow and reflections in the cochlea.

In the fruit fly Drosophila melanogaster, hearing is based on dedicated mechanosensory neurons transducing vibrations of the distal part of the antenna. Examination of this receiver's vibrations in wild-type flies and mechanosensory mutants had shown that the auditory mechanosensory neurons are motile and give rise to key characteristics that define the cochlear amplifier of vertebrates, including nonlinear compression and self-sustained oscillations, the mechanical equivalent of spontaneous otoacoustic emissions. Violations of the equipartition theorem now have confirmed that the neurons exhibit power gain, lifting the fluctuations of the receiver above thermal noise. By opposing damping, this neural energy contribution boosts the sensitivity and frequency-selectivity of the fly's antennal ear.

The geometry and physical properties of the subtectorial space are well-suited to propagation of symmetric Lloyd–Redwood waves or "squirting" waves [1]. Here I analytically model squirting wave interaction in an array of 62 outer hair cells in 3 rows and find that it can lead to a sharply tuned standing-wave resonance – essentially a narrow-band cochlear amplifier.

We use our recently developed hybrid analytical/finite-element approach to investigate the significance of curvature in cochlear micromechanics. The new computational model includes the detailed cellular structures within the organ of Corti, the interactions between the cochlear partition and fluid, as well as the coiling effects of the cochlear geometry. Gorverning equations are formulated in a curvilinear coordinate system. We develop an iterative algorithm to solve a fluid-solid interaction eigenvalue problem. We find that the cochlear curvature greatly increases the apical shear gain of the cochlear partition, which is a measure of the bending efficiency of the outer hair cell stereocilia.

Incorporating two nonlinear negative damping elements into a simple fourth-order system accounts for many aspects of the basilar membrane (BM) response to both clicks and tones over a wide range of sound levels.

The mammalian cochlea performs a remarkable signal processing function that maps the frequency of the incoming signal into different spatial locations along its length. A hydro-mechanical model was built to mimic the biological cochlea. Companion computational models of the mechanical devices were also implemented. The measured results from the artificial cochlea (ACochlea) demonstrate the cochlear-like features. Experimental data were compared with simulation results. The simulation results from both artificial basilar membrane (ABM) and ACochlear computational models exhibit the same trends as their experimental counterpart. Using the model as an analytic tool, the behavior of the devices was investigated. We determined that the ABM is under high tension, which degrades the frequency response of both ABM and ACochlea.

We use a 1-D discrete time domain linear model using the Fettweiss [1] wave-digital filter approach. The model, comprised of 350 segments, is used to gain insight into observed characteristics of a database of 12000 human click-evoked otoacoustic emissions (CEOAE). Its properties are explored by considering various percentages of total OHC activity and time constants of the passive and active mechanics and adaptation effect. The experiments conducted are 1) Variation of time constants to see how CEOAE latency is affected, 2) Influence of aging upon emission strength, 3) Influence of decline in OHC function upon the tonotopic map and the effect of bounding values of stiffness change; the effects of 4) small punctate "lesions" of down-graded OHC activity, 5) large regions of OHC loss and 6) The effect of varying the curvature or 'warp' of the frequency-place map [2].

We tested the hypothesis that hearing loss in superior semicircular canal dehiscence (SCD) syndrome is due to a "third cochlear window". SCD in human temporal bones produced a fluid motion in the dehiscence, a reduction in round window velocity comparable to the low-frequency hearing loss seen in SCD patients, and an increase in stapes velocity; all these results support the "third window" hypothesis. A functionally- and anatomically-based model predicts the temporal bone results.

A hydromechanical multicompartment cochlear model employing outer hair cell (OHC) force generation creates a slow traveling pressure wave inside the organ of Corti, which is principally responsible for enhanced response of the basilar membrane. This model has been shown generally to mimic physiological data using physiologically realistic parameters. NIH supported this work.

A 3D finite-element model of the gerbil cochlea was built, based on a simplified cochlear partition and a viscous fluid field, with fully coupled fluid-structure interaction. A first model, designed to match experimental measurements of point stiffness, generated traveling waves with phase accumulation in line with experimental results but failed to recreate the expected passive frequency-place map of gerbil. Cochlear input impedance was also higher than previously calculated values based on measurementsin Gerbil. In a second model, the stiffness map used as a parameter set was altered to bring the frequency-place map of the model in line with the known function. However, this change further increased cochlear input impedance suggesting that models of this type are unable to account for the full range of experimental measures of the passive gerbil cochlea.

The passive behavior of the human cochlea is simulated with a "box model" design. A micromachined composite material of polyimide and aluminum represents the cochlear partition. Two fluid channels were macro machined from plastic and filled with saline. The physical model demonstrated several important cochlea features: traveling waves, tuning, frequency to place tonotopic organization, and roll off at the characteristic place. Calculations using the WKB asymptotic approximation confirm the measured responses and improvements are seen in the calculation using quasi-static stiffness measurement data.

A life-size hydrodynamical cochlear model is demonstrated. The structure is fully micromachined and suitable for batch fabrication. Laser Doppler Velocimetry (LDV) measurements show cochlear-like traveling fluid structure waves with a phase lag of approximately 3 cycles (6π radians) and displacement magnitude of 0.2-0.5 nm/Pa at the location of maximum response. The device responds in the 10-70 kHz band.

This paper proposes a novel model for central auditory processing, a network of nonlinear oscillators. The properties of such networks are common to a family of physiological models that includes active cochlear models and oscillatory neural networks. Auditory perception can be modeled based on the generic properties of such physiological mechanisms, providing a bridge between physiology and psychoacoustics.

Quick Questions.

Stereocilia and Tip Links.

Somatic Motility of Outer Hair Cells.

Waves in the Cochlea.

Fluid Flow in the Cochlea.

Traveling Waves in the Cochlea.

Are Traveling Waves in the Cochlea Going in Both Directions?

Author Index.