|
Distortion product otoacoustic emissions (DPOAE) are increasingly used as an objective test for noninvasive hearing screening. When two pure tones with frequencies f1 and f2 are sent to the cochlea, the most prominent DPOAE is the cubic one produced at 2f1-f2, and this presentation will mainly emphasize its properties. DPOAEs are undoubtedly generated by cochlear nonlinearities. It is widely held that they arise from certain stages of sound processing by the outer hair cells (OHC) and that OHCs ensure normal cochlear sensitivity and tuning. Thus, DPOAEs should provide a privileged tool for monitoring the harmful effects of loud sound because OHCs are known to be one of the main targets of NIHL. Although DPOAEs provide the clinicians with a reliable screening limit of about 30 dB HL around f2, no reliable relationship has been found thus far between possible residual DPOAEs and either hearing loss or amount of impaired sensory cells. Furthermore, puzzling contradictory findings have been reported as to the presence of DPOAEs despite a large hearing loss (i.e. >30-40 dB) notably with high-level stimuli. These observations raise the following issues. What is the generation site of DPOAEs in a normal or pathological cochlea (OHCs, basilar membrane, place tuned to f2, 2f1-f2, places basal to f2...)? Is it necessary to account for interferences between several discrete sources, arising from different locations or different mechanisms and possibly exhibiting differential susceptibility to sensory cell damage? Do DPOAE changes depend on the nature of OHC pathology (NIHL, anoxia, ototoxic drugs, genetics...)? Once a source of DPOAE is characterized, is there any means of modelling the physiological process of its generation and deriving what might quantitatively relate DPOAE amount to sensory cell activity and thresholds? The goal of this presentation is to examine these issues, review the available data and propose a comparatively simple model. Keywords: distortion, otoacoustic emissions, acoustic overstimulation, anoxia, ototoxicity, outer-hair-cell function
How to cite this article:
Avan P, Bonfils P, Mom T. Correlations among distortion product otoacoustic emissions, thresholds and sensory cell impairments. Noise Health 2001;3:1-17 |
| Introduction | |  |
Exposure to loud sound can cause either temporary or permanent damage to the cochlea and particularly to its outer hair cells (OHC), one of the main targets of acoustic injury (e.g. Hamernik et al., 1989). It is widely held that thanks to their ability to perform bidirectional transduction (Weiss, 1982; Brownell et al., 1985; Dallos and Evans, 1995), OHCs are the key element of a mechanical feedback loop at the origin of the so-called "cochlear amplifier" (Gold, 1948; Davis, 1983). In short, the activity of OHCs enhances basilar membrane vibrations in a frequency-selective manner, thereby increasing both the sensitivity of the cochlea to low-level sounds and its tuning. Disruption of OHC function either by acoustic overexposure or by any other OHC-specific pathology leads to impaired auditory thresholds as well as worse frequency selectivity (Liberman and Dodds, 1984; Patuzzi et al., 1984; Hamernik et al., 1989; Davis et al., 1993). Auditory evoked potentials and behavioural methods have thus been widely used for years for assessing the harmful effects of sound exposure. However, these methods are either coarse (for example, only a few frequencies are monitored at octave intervals) or lengthy (especially when frequency selectivity has to be evaluated -Davis et al., 1989-). Since Kemp (1978) discovered otoacoustic emissions as sounds being reemitted in the external ear canal and representing a by-product of mechanical activity inside the cochlea, otoacoustic emissions have been proposed as a faster objective and non-invasive means of assessing noise-induced cochlear damage. Distortion product otoacoustic emissions (DPOAE; Kim et al., 1980) are particularly interesting in animal experiments for practical reasons, their level being much larger than that of other classes of otoacoustic emissions and their measurement taking only a few seconds, and more fundamentally, for their alleged frequency specificity.
Most of the recent research has been devoted to the cubic-difference DPOAE at 2f1 - f2 in response to a stimulation by two pure tones or primaries at f1 and f2, and most of the following presentation will be restricted to the properties of the 2f1 - f2 DPOAE. Obviously, its existence requires some nonlinear mechanism in the cochlea and appropriate mechanical feedback. Whenever the cochlear function is mechanically normal, DPOAEs are found not only in the external ear canal (Kim et al., 1980), but also in basilar membrane movements (Robles et al. 1991) and in scala vestibuli (Avan et al, 1998). By contrast, whenever the sensitivity and tuning of the cochlea have been altered in such a way that a hearing loss >30 dB is observed around f2, DPOAEs decrease or disappear at 2f1 - f2 (review in Probst et al., 1991). That OHCs should be healthy and motile for otoacoustic emissions to exist has been clearly acknowledged (Brownell, 1990). A possible contribution of cochlear inner hair cells to DPOAEs was suggested (Schrott et al., 1991) then definitely discarded when Trautwein et al. (1996) showed that carboplatin-induced absence of inner hair cells bore no influence on DPOAE levels.
The potential usefulness of DPOAEs to monitor OHC function has been determined more than a decade ago in relation to noise-induced damage (Zurek et al., 1982; Schmiedt, 1986; LonsburyMartin et al., 1987; Martin et al., 1987; review in Avan et al., 1996) or other types of OHC dysfunctions: Horner et al. (1985) studied various strains of mice with genetically impaired hearing system; Brown et al. (1989) exposed guinea pigs to repeated injections of an aminoglycoside antibiotics. Once the link between DPOAEs and OHC-dependent cochlear function was established, more ambitious protocols were tested in order to assess whether DPOAEs could serve as a tool for quantitative assessment of OHC function. For example, Hofstetter et al. (1997) have disclosed significant, simple relationships between DPOAE levels and percentage of OHC loss due to the ototoxic drug carboplatin. However, detailed analyses of the outcome of DPOAEs vs. other tests have led to contradictory or inconclusive reports suggesting no straightforward or general relation of DPOAE to OHC function (e.g. Canlon et al., 1993; Subramaniam et al., 1994). These contradictions are partly due to the complexity of DPOAEdetecting protocols involving more or less arbitrary choice of primary levels, ratio f2/f1, f2 steps, etc.
The level of primary tones has turned out to be one of the most important issues, and a confusing one. Whereas DPOAEs elicited by low-level primaries (i.e. <60 dB SPL) seem very sensitive to the cochlear status close to the place tuned to f2 (Brown and Kemp, 1984; Fahey and Allen, 1997; Avan et al., 1998), DPOAEs generated by higher-level primaries can persist with comparatively large amplitudes despite cochlear impairment. Such "high-level DPOAEs" have been shown to persist after exposure to loud sound (Puel et al., 1995), selective cochlear ischemia (Mom et al., 1997) or administration of furosemide (Mills and Rubel, 1994). Following Whitehead et al. (1992a, b), Mills et al. (1994) then Mills (1997) have proposed that DPOAEs result from the combination of two discrete sources termed "active" and "passive" in accordance with their alleged physiological properties. In this twosource model, "active" DPOAEs would be elicited by tones well below 60 dB SPL by vulnerable OHCs tuned to f2, whereas "passive" DPOAEs, generated basal to f2 by higher-level primaries, would reflect the much less vulnerable basal basilar-membrane macromechanics. Even though OHC impairment would affect the active contribution and eliminate low-level DPOAEs, high-level primaries might let the passive DPOAEs be detected and lead to false interpretations. As the highest primary level ensuring only "active" DPOAEs to be recorded is unknown, many researchers have been prompted to perform lengthy recordings of input-output plots, for example with the primary levels varying from 30 to 80 dB SPL in 5 dB steps at every tested frequency. The issue of primary levels also leads to crucially important questions as to the physiological origin of DPOAEs. If passive DPOAEs were to be detected despite damaged OHCs, this could mean that the ability to generate and emit acoustic distortion could be found in poorly tuned structures instead of being the exclusive property of OHCs. Thus, the idea for using DPOAEs as a quantitative tool for assessing OHC function would have to be discarded. Anyway, the meaning of residual DPOAEs and the location of their sources have to be ascertained before they can be relied upon as measurements of cochlear impairment.
This contribution will examine the properties of high- vs. low-level DPOAEs and compare the effects of temporary or permanent noise-induced hearing loss (NIHL) to those of other OHC pathologies due to ototoxic drugs, genetic diseases or ischemia. Experiments will be reported suggesting that high-level DPOAEs are indeed vulnerable and may well arise from the same mechanisms as low-level ones. Evidence that the main source of DPOAEs can be identified within the mechanical feedback loop of OHCs will be discussed and a model will be proposed to account for the properties of DPOAEs when OHC function is disrupted, as a function of the physiological cause of impairment.
High-level vs. low-level DPOAEs from a pathological cochlea.
Typical growth functions representing DPOAE levels against intensity of primary tones seem to be made of two different parts. Below 60 dB SPL, normal DPOAEs tend to increase rather slowly [Figure - 1] with a slope of about 1 dB per dB increase of the levels of the two primary tones, and can even tend to plateau or decrease around 60 dB SPL (thin continuous line in [Figure - 1]). Above 70 dB SPL, DPOAE growth becomes steeper. When the function of OHCs gets disrupted, an apparent differential vulnerability sometimes occurs in that the DPOAE growth function no longer exhibits two slopes, the high-level segment being little affected while the low-level one gets shifted downward and grows more steeply ([Figure - 1], thin continuous line before auditory fatigue, vs. dashed line after auditory fatigue). Brown et al. (1989) reported that two days after a treatment with gentamicin was started in guinea pigs, highlevel DPOAEs were unchanged while low-level DPOAEs had already undergone a sharp decrease. Norton et al. (1991) described a similar persistence of high-level DPOAEs in gerbils post mortem, as well as Fitzgerald et al. (1993) in guinea pigs after intracochlear perfusion of salicylate, Mills and Rubel (1994) in gerbils after furosemide injection, Puel et al. (1995) in guinea pigs with temporary threshold shift due to mild pure-tone exposure, Mom et al. (1997) in gerbil ears after complete selective cochlear ischemia. In contrast, no persistence of high-level DPOAEs was reported by Horner et al. (1985) in genetically defective ears up to 90 dB SPL stimulus level, Le Calvez et al. (1998) in mutant mice or Frolenkov et al. (1999) after roundwindow administration of an inhibitor of OHC electromotility. Lonsbury-Martin et al. (1993) pointed out that although sometimes more robust than the low-level segment, the high-level segment of the DPOAE input-output curve could be eliminated by combined factors known to affect OHCs, such as ethacrynic acid plus gentamicin.
Actually, most results from noise-exposed ears do not report such a large influence of stimulus level on DPOAEs responses to pathology. Zurek et al. (1982) found that DPOAEs were sensitive indicators of NIHL whatever the stimulus level between 30 and 90 dB SPL, Schmiedt (1986) confirmed this finding with 70 and 80 dB SPL stimuli. Most of the examples of input-output functions in Hamernik et al. (1996) depicted similar DPOAE level changes at all stimulus levels from 30 to 75 dB SPL. Clock Eddins et al. (1999) recently reported that low-level portions of DPOAE growth functions dropped down to noise floor, and that although high-level portions were retained, their decrease was systematic and almost as large for 80 dB SPL primary levels as for 60 dB SPL ones. [Figure - 1] represents two examples of DPOAE changes in gerbil ears after overexposure to a loud tone at 6 kHz (95 dB SPL, 15 min). The input-output plots were built at f2 = 10 kHz. In one case (thin lines), the temporary threshold shift was about 10 dB and the DPOAE was much more affected below 65 dB SPL than at higher levels. In the other case (bold lines), the temporary threshold shift was larger and the DPOAE level was decreased whatever the stimulus level.
All these examples lead to conclude that acoustic overexposure decreases DPOAE levels at all stimulus intensities, although the amount of decrease may be level-dependent. Even though it were not the case, due to signal-to-noise ratio considerations, DPOAEs could be said to disappear when low-level stimuli are used and to persist for higher level stimuli. Other types of OHC pathologies do not result in the same behaviour, so that it may be suggested that the exact site of cochlear pathology determines the way DPOAEs respond to this pathology, even though it may not influence much the attendant hearing loss. It is certainly important to take into consideration that overstimulation acts on the stereocilia bundles of OHCs and can induce prolonged closure of their mechanoelectrical transduction channels (Patuzzi et al., 1989), while ischemia and furosemide chiefly affect the endocochlear potential through stria vascularis impairment (e.g. Mills et al., 1993).
Is the vulnerability of high- vs. low-level DPOAEs really differential?
The most disturbing finding that prompted many researchers to oppose the use of high-level DPOAEs is their persistence in euthanized animals. It is well-known that any hint of cochlear activity on the basilar membrane vanishes within seconds after death (Johnstone et al., 1986). Post-mortem distortion should disappear simultaneously (Norton et al., 1991) if a straightforward relation were to exist between the amount of cochlear gain and the level of DPOAEs.
In order to describe the properties of postischemic DPOAEs and investigate their origin, Mom et al. (1997, 1999) have designed a model of complete selective interruption of the blood supply to the cochlea in gerbils. They confirmed that several minutes after onset of ischemia, DPOAEs could remain present for a while well above noise floor when primary intensities exceeded 60 dB SPL, although the levels of residual DPOAEs were decreased relative to preischemic state. [Figure - 2] depicts an example of DPOAE level change during the first 5 min after onset of complete ischemia in a gerbil's ear, with 60 dB SPL primary intensities. Although the DPOAE reached noise floor after 50 s, it quickly rebounded and tended to a plateau around only 15 dB below initial level after two more min. In the meantime, DPOAEs elicited by primary intensities < 50 dB SPL irreversibly vanished after less than one min. Likewise, eight-nerve compound action potential thresholds were shifted by more than 60 dB within one min, in keeping with the idea that the cochlear activity had been quickly turned off. On average, the post-ischemic decrease in DPOAE levels was about -20 dB for 60 dB SPL primary intensities, -17 dB at 70 dB SPL and -10 dB at 80 dB SPL after 3 min, regardless of f2. For the next hour of ischemia, the subsequent decay of DPOAEs was very slow. Mom et al. (1997) also reported that, when the cochlear blood flow was let back to normal less than 5 to 6 min after ischemia onset, DPOAEs always recovered rapidly although with a non monotonic course back to their preischemic level. Full recoveries of DPOAEs, cochlear microphonics and eighthnerve compound action potential thresholds could even be observed for longer interruptions of cochlear blood flow, up to 8 min. This last finding clearly implies that none of the sensory structures in the organ of Corti were damaged by short-term ischemia, and that OHCs were not the primary target of ischemia-induced impairment. As the stria vascularis is known to be highly oxygen-dependent, it was most likely at the origin of a decrease in endocochlear potential (Konishi, 1979) sufficient to disable the cochlear activity.
In a further experiment, Mom et al. (1999a, 2000) exposed a series of gerbils to a moderately loud tone at frequency fL (fL = 6 kHz, 90-95 dB SPL, 15 to 30 min) so as to induce some auditory fatigue prior to ischemia. During the next minutes following tonal exposure, a drop in DPOAE levels was observed at several f2 frequencies as described previously see [Figure - 1]. As expected for auditory fatigue, postexposure DPgrams presented a notch relative to preexposure controls and, as usual, the DPOAE level decrease turned out to be maximum at frequencies f2 around half an octave above fL. Although partial recovery was observed during the first 15 min following the end of exposure, the DPOAE level decrease remained stable for the next hour, within 1 dB at all frequencies. The level and duration of pure tone exposure were such that a maximum DPOAE decrease of 15 to 30 dB was obtained around f2 = 10 - 13 kHz, whereas smaller decreases were present between 6 and 10 kHz and above 13 kHz. No DPOAE change was found at f2 < 6 kHz. A typical series of postexposure DPgrams is represented on [Figure - 3]A (dashed lines, while continuous lines represent the reference preexposure DPgrams). In this example, the DPOAE notch was hardly more than 10 dB deep and extended from f2 = 6.5 to 14.5 kHz. Beyond these limits, DPOAEs were insensitive to the tonal exposure so that postexposure plots were overlaid with preexposure ones. Notice that DPOAE levels grew faster with stimulus intensity than before exposure, especially around f2 = 10 kHz. About 30 min after the end of tonal exposure, i.e. once the DPOAE levels were stabilised for a while, the cochlear blood flow was suddenly interrupted and a series of DPgrams was collected every minute (primary levels: 60 and 70 dB SPL).
The lower frequency range (f2 < 6 kHz) corresponded to unaltered DPOAEs after exposure to the loud tone thus the effect of ischemia upon DPOAEs was expected to be identical to that happening in unexposed ears. Indeed, DPOAE level changes after 15 min of complete ischemia [Figure - 3]B were very similar to what [Figure - 2] has already shown. The DPOAE remained clearly present around 20 dB SPL for 70 dB SPL primaries, and although the level decrease was larger for 60 dB SPL primaries, DPOAEs were above noise floor at all frequencies. The increase of DPOAE levels for a 10-dB rise in primary levels reached about 30 dB [Figure - 3]B.
At frequencies f2 within the two thin vertical lines of [Figure - 3]B (6 and 13.5 kHz), even the high-level DPOAEs were seen to get close to the noise floor. These limits for f2 coincide with the DPOAE notch produced by auditory fatigue, within 1 kHz [Figure - 3], 6.5 and 14.5 kHz). As the postexposure DPOAE decrease hardly exceeded 10 dB, it is too small to account for the large level difference of ischemic DPOAEs above vs. below 6 kHz. Another example of the effects on a DPgram of tonal overexposure followed by ischemia is depicted on [Figure - 4]. This time, 70 dB SPL primary intensities were used and auditory fatigue affected all recorded frequencies above 7 kHz. The behaviour of all fatigued ears submitted to ischemia followed the example of [Figure - 3],[Figure - 4]. Ischemic DPgrams plummeted to within 10 dB of noise floor in a frequency interval whose boundaries coincided, within less than 1 kHz in all ears, with those of the notch imprinted by auditory fatigue (Mom et al., 2000). Such results have several consequences. In the first place, they confirm the remark of Lonsbury-Martin et al. (1993) that high-level DPOAEs were most likely of cochlear origin, and presumably of OHC origin, since combined insults made them vanish according to the same time course as the one required for OHCs to degenerate. Each of the methods used by Mom et al. (2000) to impair the cochlea induced only a moderate decrease of high-level DPOAEs, whereas their combination turned out to be very effective. Since the auditory fatigue Mom et al. (2000) exerted to obliterate ischemic high-level DPOAEs was not likely to modify anything but the function of OHCs, these cells would be good candidates as the source of DPOAEs whatever the primary levels.
Furthermore, Mom et al's data hint at the location of the source of high-level DPOAEs. The only difference between DPOAEs below and above fL, frequency of the fatiguing tone, relates to the OHCs having been exposed to mild overstimulation. The places concerned by tonal overstimulation are well-documented: they follow the 'half-octave shift' rule (Cody and Johnstone, 1981; Mc Fadden, 1983). Only these places, within less than 1 kHz, do not behave as usual once cochlear ischemia is produced. The absence of high-level DPOAEs has thus to be due to fatigued OHCs being disabled in some respect. Since the sharp decrease of high-level DPOAEs fell in the f2-range [6, 13.5 kHz] in the example of [Figure - 3], while the fatigued OHCs were tuned to [6.5, 14.5 kHz], it is thus possible to infer that when present, these DPOAEs were produced as usual, in the range tuned to f2.
This partly addresses the first question of this presentation, i.e. "What is the generation site of DPOAEs in a normal or pathological cochlea (OHCs, basilar membrane, place tuned to f2, 2f1-f2, places basal to f2...)"?: we propose that whatever the stimulus level used to elicit DPOAEs, up to 70 dB SPL at least and provided no instrumental distortion is produced by the recording equipment, high- as well as low-level DPOAEs arise from the same generation site(s). It does not mean, to answer the second question of the abstract, that it is not necessary to account for interferences between several discrete sources, arising from different locations or different mechanisms and possibly exhibiting differential susceptibility to sensory cell damage. Actually, several convincing proofs have been recently provided to support the multi-source concept, although with a meaning totally different from the previously discussed "active" vs. "passive" distinction.
Several sources for DPOAEs
The primary site for DPOAE generation must be the place on the basilar membrane where maximum interaction occurs between the vibrations produced by the two primary tones, accordingly, it is widely thought to be at or near the place tuned to f2 (e.g. Fahey and Allen, 1997). Direct confirmation has been obtained from normal cochleas by Avan et al. (1998) who checked that the phase of acoustic pressure at 2f1-f2 inside scala vestibuli is minimum at the place tuned to f2 as long as the stimulus level does not exceed 70 dB SPL. It indicates that the emission is generated at f2, then exhibits a progressive phase lag along with its intracochlear propagation. Damaged cochleas also provide clear evidence of the role of the place tuned to f2: Martin et al. (1987) then Puel et al. (1995) showed that, in animals exposed to auditory fatigue, the map of DPOAE decrease tends to coincide with that of audiometric alterations when DPOAEs are plotted against f2. The idea that DPOAEs are a "frequencyspecific" analyser of cochlear mechanics relies upon these considerations.
Nevertheless, several secondary DPOAE sources have been discovered. It has been admitted for years that once emitted, otoacoustic emissions behave as regular external sounds while they propagate along the cochlear scalae. The DPOAE at 2f1 - f2 is generated at a place tuned to the higher frequency f2, thus the corresponding acoustic pressure is expected to propagate toward to the more apical place tuned to 2f1-f2. After Goldstein and Kiang (1968) reported that auditory-nerve fibres with characteristic frequency 2f1 - f2 respond to pairs of primary tones at f1 and f2, Robles et al. (1991) directly showed that the basilar membrane vibrates at 2f1 - f2 as a result of DPOAE propagation from f2 toward its characteristic place. In turn, when it is healthy, the place tuned to 2f1 - f2 can emit an acoustic contribution to the overall emission at 2f1 - f2, that interferes with the signal directly coming from the place tuned to f2. Interferences manifest themselves as a fine structure in the DPOAE patterns, due to partial cancellation of the two contributions for every combination of frequencies producing a half-cycle phase difference between the two sources (Brown et al., 1996; Talmadge et al., 1999; Mauermann et al., 1999). Fine structure is particularly evident when the two sources have similar amplitudes, as happens for large f2/f1 ratios and high-level primaries (Fahey and Allen, 1997; Talmadge et al., 1999). The relevance of fine-structure analysis in humans has been pinpointed by Engdahl and Kemp (1996) in the case of acoustic injury. Furthermore, places basal to f2 probably get important for DPOAE generation when high-level primaries are used, partly because nonlinear interactions between f1 and f2 may occur over a broader length of basal basilar membrane. Suppression data by Martin et al. (1987, 1998) and direct intracochlear pressure measurements of DPOAEs elicited by 80 dB SPL stimuli (Avan et al., 1998) support this hypothesis.
In theory the existence of secondary sources may confuse the issue of frequency specificity, that is of the frequency-to-place dependence of DPOAE levels on OHC function. However, the most common pathologies apparently lead to simpler situations. In human ears with different profiles of hearing loss, Mauermann et al. (1999) recently reported that DPOAE generation mainly depended on the cochlear status at the place tuned to the primary tones, in other words on the ability of the primary source of DPOAEs to work properly. If restricted to the places tuned to 2f1 - f2, cochlear impairment only flattened the DPOAE fine structure without influencing much the DPOAE level. This observation was backed up by a theoretical model of intracochlear sound propagation in active or less active cochleas.
As a provisional conclusion, it seems conservative to state that the primary source of DPOAE is near the place tuned to f2 and is predominant in many experimental situations. Secondary sources undoubtedly exist elsewhere in the cochlea, one of them with characteristic frequency 2f1 - f2, and the cochlear status at this place influences the fine structure of DPgrams.
The contributions from hypothetical basal sources should be small enough to be negligible as long as stimulus levels do not exceed 70-80 dB SPL.
Cochlear feedback loop and generation of distortion
The data presented in sections 1 and 2 clearly suggest that DPOAE changes depend on the nature of cochlear pathology, since their inputoutput characteristics exhibit different sensitivities to noise-induced hearing loss, anoxia, ototoxic drugs or genetic impairment. As we asserted that the main source entailed in DPOAE generation is the same in every pathological model as in a healthy cochlea, we have to address the next issue, that is: is there any means of modelling the physiological process of DPOAE generation so as to derive some quantitative relationship between amount of distortion and sensory-cell activity around f2?
As shown by Mom et al. (2000), auditory fatigue proves to be an effective means to abolish the residual high-level DPOAEs, the existence of which had been considered paradoxical (Norton et al., 1991; Mills et al., 1994). Mild exposure to loud sound (90-95 dB SPL for a few min in Mom's experiment) can only affect OHCs, because effects of loud sound on inner hair cells or synapses via excitotoxic processes have been reported only for much louder and longer exposures. Overexposure probably acts on OHCs by decreasing the ability of their mechanoelectrical transduction channels to let potassium ions in when stereocilia are deflected (Patuzzi, 1998). This is all the more relevant to the present discussion since there is strong evidence that the mechanoelectrical stage of OHC function is nonlinear and generates distortion. The gating processes can be described by a second-order Boltzmann-shaped transfer characteristics (Patuzzi et al., 1989; Kros et al., 1992; Jaramillo et al., 1993). The Boltzmann function is asymmetrical and it is easy to show that with a mechanical input consisting of two sinusoids at frequencies f1 and f2, the electrical output is going to contain odd- and even-order distortion products in addition to primaries at f1 and f2, and particularly a prominent component at 2f1 - f2. Furthermore, shifts of the operating point of OHCs along the transfer function are likely to occur naturally, causing modulations and sometimes cancellations of the distortion products (Frank and Koessl, 1996; Lukashkin and Russell, 1999). Obviously, the existence of such electrical distortion products do not suffice to warrant that DPOAEs will come out of the cochlea: Jaramillo et al. (1993) observed them from bullfrog saccular cells but they definitely do not emit DPOAEs. The whole OHC feedback loop has to be taken into account for acoustic distortion to be emitted (Brownell, 1990).
One of the first simple models of cochlear feedback allowing the effects of usual pathologies to be quantitatively predicted has been proposed by Patuzzi et al (1989). It is now well accepted that the function of the cochlear loop entails a sequence of events [Figure - 5], the initial one being the shear displacement of stereocilia due to the action of differential acoustic pressure across the basilar membrane. An electrochemical gradient normally exists at the apex of OHCs, mainly due to the difference between the positive endocochlear potential (EP, close to +100 mV, generated in stria vascularis) and the negative dc potential of OHCs (VOHC, around -40 to -70 mV). When stereocilia get deflected away from the modiolus, thereby opening their gating channels, potassium ions flow from the endolymph to inside the OHCs, driven by (EP - VOHC). This K+ current generates an alternative receptor potential. As a result of the unique property of OHC membranes to exhibit electromotility (Brownell et al., 1985), length changes or force generation occur at the frequency of the ac membrane potential. It opens two possibilities. In the first place, mechanical activity of OHCs on the basilar membrane can modify the stimulus. If the feedback occurs with an appropriate phase, the overall stimulus is enhanced and the sensitivity of the cochlea gets improved. As the phase of the fed back signal is frequency-dependent, so is the cochlear amplification thus the tuning is also greatly improved by the active process. The larger the intensity of the external stimulus, the smaller the relative contribution of OHCs, hence the gain due to feedback decreases at higher levels and virtually vanishes above 80 dB SPL (Ruggero, 1992; Withnell and Yates, 1998). This why the input-ouput characteristics of the basilar membrane exhibits a compressive behaviour in the intermediate range of levels, with a growth rate of about 0.3 dB / dB increase of the external stimulus level. The second consequence of the existence of mechanical feedback in the cochlea is that it transforms the aforementioned electrical distortion products into acoustic ones, in other words DPOAEs. Notice that the exact mechanism by which otoacoustic emissions eventually appear in scala vestibuli (Avan et al., 1998) is unknown. Although emissions are probably but epiphenomena of sound-processing mechanisms, the loop model supports the idea that DPOAEs provide the experimenter with indirect clues of several important stages, and helps understand why DPOAEs, auditory thresholds and tuning appear to be so tightly correlated: whenever the feedback loop is interrupted because one of its stages is pathologically disrupted, DPOAEs disappear, auditory thresholds increase and frequency selectivity gets poorer at the same time.
It is less straightforward to predict how DPOAEs are going to change when the efficiency of the feedback loop is only decreased, as it probably happens in furosemide and ischemia experiments. In both cases, an electrochemical gradient definitely persists at the apex of OHCs. The endocochlear potential drops to 0 mV shortly after furosemide injection when a high dose is used (Sewell, 1984), and to about -40 mV in an anoxic cochlea (Konishi, 1979; Kakigi and Takeda, 1998), thus a residual EP - VOHC of about 50 and 10% of its normal value persists in such experiments. Besides, mechanoelectrical transduction should not be directly affected and OHC electromotility is said to be extremely robust, to the point that even when OHC are placed in an ATP- and calcium-free medium and after their cytoplasm has been digested by proteolytic enzymes, electromechanical transduction occurs if an external electric field is provided (Huang and Santos-Sacchi, 1994). Thus DPOAEs can still be generated at the same site and by the same mechanism as healthy ones, although with a decreased level. By contrast, the gain of a feedback loop is extremely sensitive to the amount of feedback as confirmed by the computerized model of the cochlear bidirectional feedback designed by Preyer and Gummer (1996). Thus it is not surprising that after furosemide injection, the basilar membrane behaves as if nearly no gain were available: Ruggero and Rich (1991) reported that the growth function of the basilar membrane lost its compression at intermediate sound levels and got linearised with a slope close to 1 dB / dB increase of sound level. This also implies a decrease of cochlear sensitivity, by as many dB as previously provided by the gain of the healthy cochlear feedback loop (Sewell, 1984, for eightnerve-fibre tuning curves). It is important to notice that according to the model and as confirmed by experimental data, DPOAEs can still be present provided the stimulus intensities are large enough to offset the loss of efficiency of the feedback loop. Conversely, noise-induced hearing loss or auditory fatigue directly affect the sensitivity of mechanoelectrical transduction channels (Patuzzi, 1998), i.e. the actual source of distortion, so that one expects DPOAEs to vanish whatever the stimulus level. To summarize, any insult to some stage of the cochlear feedback loop, whatever its degree or target, is expected to seriously impair the gain of the loop and, as a result, the sensitivity of the cochlea and its compressive behaviour. However, as long as the nonlinear stage of transduction is preserved, DPOAEs can still exist. To abolish them, it is necessary either to specifically damage the nonlinear mechanoelectrical channels, or to block another mandatory stage of the cochlear loop then the blockage must be complete. This is what Frolenkov et al. (1998) achieved with an inhibitor of OHC electromotility.
That DPOAE growth functions present an unusually steep slope (3 dB / dB increase of stimulus intensities) is often considered as a signature of "passive", possibly instrumental DPOAEs (Norton et al., 1991) and instrumental DPOAEs do present this property. Actually, it is a general law that when the intensities of a pair of tonal stimuli increase by 1 dB at the input of a nonlinear device, the cubic 2f1 - f2 distortion coming out of the device must exhibit a 3-dB increase (2 dB for quadratic distortion, 5 dB for the 3f1 - 2f2 combination, etc). For algebraic reasons this holds true whatever the origin of distortion, physiological or not. What matters is that in a healthy cochlea as already stated, the actual input to the nonlinear mechanoelectrical channels of OHC stereocilia, following the basilar membrane, grows at a rate of about 0.3 dB per dB increase of the external stimuli at f1 and f2 around the place tuned to f1 and f2. This is basically why normal 2f1 - f2 DPOAEs grow at about 0.9 - 1 dB / dB sound pressure level in the ear canal for primary intensities < 70-80 dB SPL. At higher intensities, or at all intensities in a cochlea with impaired feedback and linearised basilar-membrane growth function, the input to OHC stereocilia grows at a rate of 1 dB / dB and the rate of increase of DPOAEs remains three times steeper. We suggest that this is what was observed in all the cases we mentioned of persistent high-level DPOAEs, and that there is no reason to consider them as spurious ones.
Toward a quantitative assessment of the noise-damaged cochlea with DPOAEs?
For the loop model of [Figure - 5] to be useful for quantitative predictions of basilar membrane responses in normal and pathological conditions, several data would be required: the Boltzmann transfer function of mechanoelectrical transduction channels may be derived using the method proposed by Patuzzi and Moleirinho, 1998; EP can be measured; a description is needed of how the mechanical output produced by electromotile hair cells acts on the basilar membrane and adds up to the external stimulus. Using such a model, Patuzzi et al. (1989) have derived several predictions of pathological effects, and Preyer and Gummer (1996) have related gain changes to linearisation of basilar membrane growth function. Complete predictions of DPOAEs are more involved. Obviously, being the source of distortion, the nonlinear mechanoelectrical transfer function is crucially important. This leads to expect the number of functioning mechanoelectrical transduction channels to be the prime factor determining the amount of DPOAEs. Indeed, Hofstetter et al. (1997) have reported significant relationships between DPOAE change and OHC loss after carboplatin intoxication. A similar correlation is expected for NIHL inasmuch as it directly acts on the nonlinear devices. This would enable an experimenter to monitor the number of impaired OHCs with noninvasive measurements in the outer ear canal.
Unfortunately, more factors will have to be taken into account, as suggested by Hofstetter et al. (1997) who pointed out the inaccurate predictions of their simple assumption that each OHC make an equal contribution to DPOAE amplitude. Among other factors, the position of the operating point OP [Figure - 5] along the Boltzmann transfer function has to be known accurately for DPOAE amplitudes to be computed, because OP strongly influences the amount of even- as well as odd-order distortion products. This operating point probably moves whenever the cochlear function is disrupted: for example, as discussed by Patuzzi and Moleirinho (1998), any change in EP might cause a change in OHC polarization and length and displace the cochlear structures, ultimately bending OHC stereocilia and displacing their resting position. Osmotic imbalance might also accompany functional impairment and modify the cell size and its resting position. The coupling processes that generate otoacoustic emissions in scala vestibuli or on the basilar membrane after (or perhaps before?) electromechanical transduction has taken place are largely unknown, and so is the relation of generated force to membrane potential. Finally, although it seems natural to guess that pathology should systematically decrease DPOAE levels, it may not be true for two reasons. In the first place, as pathology ultimately decreases the gain of the cochlear loop [Figure - 5], it should be kept in mind that the attendant linearisation of basilar-membrane growth function can make up for the loss of input to the OHCs when stimulus levels exceed 70-80 dB. Second, OP displacements may well alter DPOAEs in a counterintuitive manner, so that they may exhibit nonmonotonic growth functions and paradoxical "adaptive" level increases as observed by Frank and Koessl (1996) and predicted by Lukashkin and Russell (1999).
In conclusion, the origin of cubic DPOAEs is increasingly well understood. The cochlear outer hair cells are the sole nonlinear device entailed in the generation of combination tones, most likely through the operation of their mechanoelectrical transduction channels. This property makes it potentially interesting to use DPOAEs in order to monitor the effects of loud sound on the cochlea. Although certain categories of pathological events seem prone to generating a counterintuitive behaviour of DPOAEs, with persistent high-level emissions despite a large hearing loss, NIHL has a simpler effect on DPOAEs and most reports describe that DPOAE levels vary monotonically with the percentage of impaired OHCs, in a frequency-specific manner and regardless of the protocol implemented in the recording equipment. However, due to the way the cochlea processes sound with interdependent steps, further work is needed before DPOAEs can be reliably used for quantitatively assessing OHC function. Nevertheless, to date such a goal seems sensible and not unattainable.[71]
| References | | |
| 1. | Avan P., Magnan P., Smurzynski J., Probst R., Dancer A. (1998) Direct evidence of cubic difference tone propagation by intracochlear acoustic pressure measurements in the guinea-pig. Eur.J.Neurosci. 10: 17641770.  |
| 2. | Avan P., Bonfils P., Loth D. (1996) Effects of acoustic overstimulation on distortion-product and transientevoked otoacoustic emissions. In Scientific basis of noiseinduced hearing loss. Axelsson A. et al, eds. Thieme, New-York, pp 65-81. |
| 3. | Brown A.M., Harris F.P., Beveridge H.A. (1996) Two sources of acoustic distortion products from the human cochlea. J.Acoust.Soc.Am. 100: 3260-3267. |
| 4. | Brown A.M., Kemp D.T. (1984) Suppressibility of the 2f1f2 stimulated otoacoustic emissions in gerbil and man. Hear.Res.13, 29-37. |
| 5. | Brown A.M., McDowell B., Forge A. (1989) Acoustic distortion products can be used to monitor the effects of chronic gentamicin treatment. Hear.Res 42: 143-156. |
| 6. | Brownell W.E. (1990) Outer hair cell electromotility and otoacoustic emissions. Ear Hear. 11: 82-92. |
| 7. | Brownell W.E., Bader C.R., Bertrand D., de Ribaupierre Y. (1985) Evoked mechanical responses of isolated cochlear outer hair cells. Science 227: 194-196. |
| 8. | Canlon B., Marklund K., Borg E. (1993) Measures of auditory brain-stem responses, distortion product otoacoustic emissions, hair cell loss and forward masked tuning curves in the waltzing guinea pig. J.Acoust.Soc.Am. 94: 3232-3243. |
| 9. | Clock Eddins A., Zuskov M., Salvi R.J. (1999) Changes in distortion product otoacoustic emissions during prolonged noise exposure. Hear.Res. 127: 119-128. |
| 10. | Cody A.R., Johnstone B.M. (1981) Acoustic trauma: single neuron basis for the 'half octave shift'. J. Acoust. Soc. Am. 70: 707-711. |
| 11. | Dallos P., Evans B.N. (1995) High-frequency motility of outer hair cells and the cochlear amplifier. Science 267: 2006-2009. |
| 12. | Davis H. (1983) An active process in cochlear mechanics. Hear. Res. 9, 79-90. |
| 13. | Davis R.I., Ahroon W.A., Hamernik R.P. (1989) The relation among hearing loss, sensory cell loss, and tuning characteristics in the chinchilla. Hear.Res. 41: 1-14. |
| 14. | Davis R.I., Hamernik R.P., Ahroon W.A. (1993) Frequency selectivity in noise-damaged cochleas. Audiol. 32: 110-131. |
| 15. | Engdahl B., Kemp D.T. (1996) The effect of noise exposure on the details of distortion product otoacoustic emissions in humans. J.Acoust.Soc.Am. 99, 1573-1587. |
| 16. | Fahey P.F., Allen J.B. (1997) Measurement of distortion product phase in the ear canal of the cat. J.Acoust.Soc.Am. 102: 2880-2891. |
| 17. | Fitzgerald J.J., Robertson D., Johnstone B.M. (1993) Effects of intra-cochlear perfusion of salicylates on cochlear microphonic and other auditory responses in the guinea pig. Hear.Res. 67, 147-156. |
| 18. | Frank G., Koessl M. (1996) The acoustic two-tone distortions 2f1-f2 and f2-f1 and their possible relation to changes in the operating point of the cochlear amplifier. Hear.Res. 98: 104-115. |
| 19. | Frolenkov G.I., Belyantseva I.A., Kurc M., Mastroianni M.A., Kachar B. (1998) Cochlear outer hair cell electromotility can provide force for both low and high intensity distortion product otoacoustic emissions. Hear.Res. 126: 67-74. |
| 20. | Gold T. (1948) Hearing II. The physical basis for the action of the cochlea. Proc.Roy.Soc. B (Edinburgh) 135: 492-498. |
| 21. | Goldstein J.L., Kiang N.Y.S. (1968) Neural correlates of the aural combination tone 2f1-f2. Proc.IEEE 56: 981-992. |
| 22. | Hamernik R.P., Ahroon W.A., Lei S.F. (1996) The cubic distortion product otoacoustic emissions from the normal and noise-damaged chinchilla cochlea. J.Acoust.Soc.Am. 100: 1003-1012. |
| 23. | Hamernik R.P., Patterson J.H., Turrentine G.A., Ahroon W.A. (1989) The quantitative relation between sensory cell loss and hearing thresholds. Hear.Res. 38: 199-212. |
| 24. | Hofstetter P., Ding D., Powers N., Salvi R.J. (1997) Quantitative relationship of carboplatin dose to magnitude of inner and outer hair cell loss and the reduction in distortion product otoacoustic emission amplitude in chinchillas. Hear.Res. 112: 199-215. |
| 25. | Horner K.C., Lenoir M., Bock G.R. (1985) Distortion product otoacoustic emissions in hearing-impaired mutant mice. J.Acoust.Soc.Am. 78: 1603-1611. |
| 26. | Huang G., Santos-Sacchi J. (1994) Motility voltage sensor of the outer hair cell resides within the lateral plasma membrane. Proc.Natl. Acad.Sci. USA 91: 12268-12272. |
| 27. | Jaramillo F., Markin V.S., Hudspeth A.J. (1993) Auditory illusions and the single hair cell. Nature 364: 527-529. |
| 28. | Johnstone B.M., Patuzzi R., Yates G.K. (1986) Basilar membrane measurements and the traveling wave. Hear.Res.22:147-153. |
| 29. | Kakigi A., Takeda T. (1998) Effect of artificial endolymph injection into the cochlear duct on the endocochlear potential. Hear.Res. 116: 113-118. |
| 30. | Kemp D.T. (1978) Stimulated acoustic emissions from within the human auditory system. J.Acoust.Soc.Am 64: 1386-1391. |
| 31. | Kim D.O., Molnar C.E., Matthews J.W. (1980) Cochlear mechanics: nonlinear behavior in two-tone responses as reflected in cochlear nerve-fiber responses and in ear-canal sound pressure. J.Acoust.Soc.Am. 67: 1704-1721. |
| 32. | Konishi T. (1979) Some observations on negative endocochlear potential during anoxia. Acta Otolaryngol (Stockh.) 87: 506-516. |
| 33. | Kros C.J., Rusch A., Richardson G.P. (1992) Mechanoelectrical transducer currents in hair cells of the cultured neonatal mouse cochlea. Proc.Roy.Soc.Lond.B Biol.Sci. 249: 185-193. |
| 34. | Le Calvez S., Avan P., Gilain L., Romand R. (1998) CD1 hearing-impaired mice. I: Distortion product otoacoustic emission levels, cochlear function and morphology. Hear.Res. 120: 37-50. |
| 35. | Liberman M.C., Dodds L.W. (1984) single-neuron labelling and chronic cochlear pathology. III. Stereocilia damage and alterations of threshold tuning curves. Hear. Res. 16: 55-74. |
| 36. | Lonsbury-Martin B.L., Martin G.K., Probst R., Coats A.C. (1987) Acoustic distortion products in rabbit ear canal. I. Basic features and physiological vulnerability. Hear.Res.28: 173-189. |
| 37. | Lonsbury-Martin B.L., Whitehead M.L., Martin G.K. (1993) Distortion-product otoacoustic emissions in normal and impaired ears: insight into generation processes. Prog.Brain Res. 97: 77-90. |
| 38. | Lukashkin A.N., Russell I.J. (1999) Analysis of the f2 - f1 and 2f1 - f2 distortion components generated by the hair cell mechanoelectrical transducer: Dependence on the amplitudes of the primaries and feedback gain. J.Acoust.Soc.Am. 106: 2661-2668. |
| 39. | Martin G.K., Lonsbury-Martin B.L., Probst R., Scheinin S.A., Coats A.C. (1987) Acoustic distortion products in rabbit ear canal. II. Sites of origin revealed by suppression contours and pure-tone exposures. Hear.Res. 28: 191-208. |
| 40. | Martin G.K., Jassir D., Stagner B.B., Whitehead M.L., Lonsbury-Martin B.L. (1998) Locus of generation for the 2f1-f2 vs 2f2-f1 distortion product otoacoustic emissions in normal-hearing humans revealed by suppression tuning, onset latencies and amplitude correlations. J.Acoust.Soc.Am. 103: 1957-1971. |
| 41. | Mauermann M., Uppenkamp S., van Hengel P.W., Kollmeier B. (1999) Evidence for the distortion product frequency place as a source of distiortion product otoacoustic emission (DPOAE) fine structure in humans. II. Fine structure for different shapes of cochlear hearing loss. J.Acoust.Soc.Am. 106: 3484-3491. |
| 42. | McFadden D. (1983) Intense sounds may alter the mechanical properties of the cochlear partition. J.Acoust.Soc.Am.74: 447-455. |
| 43. | Mills D.M. (1997) Interpretation of distortion product otoacoustic emissions measurements: I. two stimulus tones. J.Acoust.Soc.Am. 102: 413-429. |
| 44. | Mills D.M., Norton S.J., Rubel E.W. (1993) Vulnerability and adaptation of distortion product otoacoustic emissions to endocochlear potential variation. J.Acoust.Soc.Am. 94: 2108-2122. |
| 45. | Mills D.M., Norton S.J., Rubel E.W. (1994) Development of active and passive mechanics in the mammalian cochlea. Auditory Neurosci. 1: 77-99. |
| 46. | Mills D.M., Rubel E.W. (1994) Variation of distortion product otoacoustic emissions with furosemide injection. Hear.Res. 77: 183-199. |
| 47. | Mom T., Avan P., Romand R., Gilain L. (1997) Monitoring of functional changes after transient ischemia in gerbil cochlea. Brain Res. 751: 20-30. |
| 48. | Mom T., Avan P., Bonfils P., Gilain L. (1999) A model of cochlear function assessment during reversible ischemia in the Mongolian gerbil. Brain Res.Prot. 4: 249-257. |
| 49. | Mom T., Gilain L., Avan P. (2000) On the origin of distortion in damaged cochleas. In Abstract of the 23rd ARO Midwinter Meeting . Popelka G., ed. ARO, Mt Royal, abstr. 236 pp.66-67. |
| 50. | Norton S.J., Bargones J.Y., Rubel E.W. (1991) Development of otoacoustic emissions in gerbil: evidence for micromechanical changes underlying development of the place code. Hear.Res. 51: 73-91. |
| 51. | Patuzzi R.B. (1998) A four state kinetic model of the temporary threshold shift after loud sound based on inactivation of hair cell transduction channels. Hear.Res. 125: 39-70. |
| 52. | Patuzzi R.B., Johnstone B.M., Sellick P.M. (1984) The alteration of the vibration of the basilar membrane produced by loud sound. Hear.Res. 13: 99-100. |
| 53. | Patuzzi R.B., Yates G.K., Johnstone B.M. (1989) Outer hair cell receptor and sensorineural hearing loss. Hear.Res. 42: 47-72. |
| 54. | Patuzzi R.B., Moleirinho A. (1998) Automatic monitoring of mechano-electrical transduction in the guinea pig cochlea. Hear.Res. 125: 1-16. |
| 55. | Preyer S., Gummer A.W. (1996) Nonlinearity of mechanoelectrical transduction of outer hair cells as the source of nonlinear basilar-membrane motion and loudness recruitment. Audiol. Neurootol. 1: 3-11. |
| 56. | Probst R., Lonsbury-Martin B.L., Martin G.K. (1991) A review of otoacoustic emissions. J.Acoust.Soc.Am. 89: 2027-2067. |
| 57. | Puel J.L., Durrieu J.P., Rebillard G., Vidal D., Assie R., Uziel A. (1995) Comparison between auditory brainstem responses and distortion products otoacoustic emissions after temporary threshold shift in guinea pig. Acta Acustica 3: 75-82. |
| 58. | Robles L., Ruggero M.A., Rich N.C. (1991) Two-tone distortion in the basilar membrane of the cochlea. Nature 349: 413-414. |
| 59. | Ruggero M.A. (1992) Responses to sound of the basilar membrane of the mammalian cochlea. Curr.Opin.Neurobiol. 2: 449-456. |
| 60. | Ruggero M.A., Rich N.C. (1991) Furosemide alters organ of Corti mechanics: evidence for feedback of outer hair cells on the basilar membrane. J.Neurosci. 11: 1057-1067. |
| 61. | Schmiedt R.A. (1986) Acoustic distortion in the ear canal. I. Cubic difference tones: Effects of acute noise injury. J.Acoust.Soc.Am. 79: 1481-1490. |
| 62. | Schrott A.L., Puel J.L., Rebillard G. (1991) Cochlear origin of 2f1-f2 distortion products assessed by using two types of mutant mice. Hear.Res. 52: 245-254. |
| 63. | Sewell W.F. (1984) The effects of furosemide on the endocochlear potential and auditory-nerve fiber tuning in cats. Hear.Res. 14: 305-314. |
| 64. | Subramaniam M., Salvi R.J., Spongr V.P., Henderson D., Powers N.L. (1994) Changes in distortion product otoacoustic emissions and outer hair cells following interrupted noise exposures. Hear.Res. 74: 204-216. |
| 65. | Talmadge C.L., Long G.R., Tubis A., Dhar S. (1999) Experimental confirmation of the two-source interference model for the fine structure of distortion product otoacoustic emissions. J.Acoust.Soc.Am. 105: 275-292. |
| 66. | Trautwein P., Hofstetter P., Wang J., Salvi R.J., Nostrant A. (1996) Selective inner hair cell loss does not alter distortion product otoacoustic emissions. Hear.Res. 96: 71-82. |
| 67. | Weiss T.F. (1982) Bidirectional transduction in vertebrate hair cells: a mechanism for coupling mechanical and electrical processes. Hear.Res. 7: 353-360. |
| 68. | Whitehead M.L., Lonsbury-Martin B.L., Martin G.K. (1992a) Evidence for two discrete sources of 2f1-f2 distortion-product otoacoustic emissions in the rabbit: I. Differential dependence on stimulus parameters. J.Acoust.Soc.Am. 92: 1587-1607. |
| 69. | Whitehead M.L., Lonsbury-Martin B.L., Martin G.K. (1992b) Evidence for two discrete sources of 2f1-f2 distortion-product otoacoustic emissions in the rabbit: II. Differential physiological vulnerability. J.Acoust.Soc.Am. 92: 2662-2682. |
| 70. | Withnell R.H., Yates G.K. (1998) Onset of basilar membrane non-linearity reflected in cubic distortion tone input-output functions. Hear.Res. 123: 87-96. |
| 71. | Zurek P.M., Clark W.W., Kim D.O. (1982) The behavior of acoustic distortion products in the ear canals of chinchillas with normal or damaged ears. J.Acoust.Soc.Am. 72: 774 780. |
Correspondence Address:
Paul Avan
Laboratory of Sensory Biophysics, School of Medicine, Clermont-Ferrand
France
 Source of Support: None, Conflict of Interest: None  | Check |
PMID: 12678937 
[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5] |
Search Pubmed for
