Auditory Nerve Responses Research Articles

Genotypic Differences in Behavioral, Physiological and Anatomical Expressions of Age-Related Hearing Loss in the Laboratory Mouse: Original Papers Travaux originaux

The auditory nerve isoelectric thresholds, in response to tone pigs ranging from 5 to 20 kHz, are similar in the young C57BL/6 and CBA/J mice, although the latter genotype has somewhat more sensitive responses from 30 to 80 kHz. But their behavioral audiograms, obtained by the classical conditioning technique of Ehret, are very discrepant. Even though the behavioral audiogram of the CBA/J mouse can be predicted by measurements from its auditory nerve, the behavioral thresholds of the young C57BL/6 mouse are approximately 40 dB less sensitive than its electrophysiological measurements. Cochlear hair cell loss was not evident in young mice of either genotype. As the CBA/J approaches the end of its predicted life span, its auditory anatomy, physiology and behavior are not significantly altered; but those rare individuals, who exceed their predicted life span by 40% finally develop hearing loss of a mixed nature. By contrast, the C57BL/6 mouse shows a relatively rapid decline of hearing as it ages. By 200 days of age, its auditory nerve responses are 30 dB less sensitive at 5 kHz, and 55 dB less sensitive at 30 kHz, than similar measures taken at adolescence. Over this same age span, its behavioral sensitivity has only declined by 15 and 25 dB at these two frequencies. Hair cell counts correlated poorly with both behavioral and electrophysiological auditory measures in the C57BL/6 mouse.

Investigation of the mastoid electrode contribution to the brain stem auditory evoked response.

The auditory nerve and bran stem responses recorded at the vertex and mastoids were closely examined in normal subjects using a larynx electrode as a neutral reference point. With monaural stimulation, significant activity was found at the vertex and both mastoids, and the wave-forms at the ipsi- and contralateral mastoids were consistently different. With binaural stimulation the small amplitude mastoid response were largely in phase with the vertex response. The implications of the findings are discussed.

A functional measure of brain activity: Brain stem transmission time

Surface-recorded auditory nerve and brain stem responses are being used routinely for diagnostic purposes in man. When interest is in auditory diagnosis, the electric response threshold is of primary importance. However, when used in neurological diagnosis, the wave form of the response is important. As a measure of one aspect of response wave form, this paper suggests the use of brain stem transmission time (BTT), defined as the time interval between the first earlobe-negative wave (response of the auditory nerve — the ‘input’ to the brain stem) and the earlobe-positive wave from the region of the inferior colliculus (the ‘output’ of the brain stem). The paper shows that BTT is longest in neonates, approaches adult values at the age of about 3 years, is relatively independent of click intensity, conductive hearing loss (middle ear lesion), click rate (except for high rates) and click frequency (filtered clicks). The finding that in a given age group, BTT is generally independent of most stimulus conditions, makes it a useful functional test of brain stem activity.

Hearing loss in relation to physical properties of experimentally induced middie-ear effusions

It is commonly assumed that effusions present in the middle ear for a short duration are thin and produce a mild hearing loss whereas effusions present for a longer duration become more viscous with time and produce a greater hearing loss. In order to test this hypothesis, we have used cats with the eustachian tube (ET) ligated on one side to produce middle-ear effusions. Auditory nerve responses to broad-band clicks were reported from the ear canal. Plots of N1 response amplitude and latency versus click SPL were constructed. Hearing loss was determined as the average displacement along the SPL axis of the latency plot, relative to a previously recorded control species. Viscosity of effusion was measured with a cone and plate viscometer. Specific gravity was measured using a 1 cc pycnometer. Middle-ear pressure was assessed tympanographically. After ET ligation all cats developed negative middle-ear pressure within 3 days and flat (Type B) tympanograms within 7 weeks. Hearing loss also develops early and becomes maximal at about 3 weeks. The data indicate there is not a significant difference between the amount of hearing loss associated with thin fluids and with glue. Viscosity is not clearly related to the duration of time the fluid was present in the middle ear. There is a significant positive correlation between the amount of hearing loss and the volume of fluid present in the middle ear. Thus, the amount of hearing loss appears to depend only on the volume of effusion and is not significantly dependent on other physical properties of the fluid.

Auditory Nerve and Brainstem Responses in the Newborn and in Preschool Children

The practicability of recording auditory nerve and brainstem responses during early childhood is demonstrated. This paper reports our experience with two different electrode locations: ear-canal/ear-lobe and ear-lobe/vertex. The test is performed with a light portable stimulus-generating and response-recording system that can easily be brought into wards and nurseries. In most instances, the presence of responses can be assessed using ear-lobe/vertex recordings. There is seldom any need for sedation or general anaesthesia. Testing personnel can be limited to one trained technician. If responses are not detected, ear-canal/ear-lobe recording may be used to clarify the results. General anaesthesia may then be necessary and the presence of a physician will be required for the placement of the canal electrode.

Hair cell condition and auditory nerve response in normal and noise-damaged cochleas.

Histological and physiological data are presented from cats born and raised in a low-noise environment and from cats with long-standing, noise-induced threshold shifts. Even after a 1 1/2-year survival from acoustic trauma, there were threshold elevations of roughly 40 dB in the response of single auditory-nerve fibers which could not be correlated with significant loss of hair cells. An attempt was made to correlate these threshold differences with differences in the condition of the sensory cells as seen in a light-microscopic examination of epon-embedded surface preparations. Of all the histological features evaluated, the orderlines of the stereocilia, on both inner and outer hair cells, showed the closest correlation with single-unit thresholds. In the final analysis, most of threshold shift in the noise-exposed ears could be accounted for by loss or damage to sensory cells clearly visible under the light microscope.

Auditory responses in cats produced by pulsed ultrasound

Auditory nerve responses and cochlear microphonics are produced in cats by pulsed 5‐MHz ultrasonic energy from a transducer placed against the dura mater. The pulses must be relatively intense (∼30 W/cm2) to produce a response, but can be sufficiently brief (less than 70 μs) that the brain tissue is not observably heated. The cats apparently respond to radiation pressure transients accompanying the absorption of the ultrasound in the brain tissue. Both the amplitude and latency of the N1 neural responses to the ultrasound can be matched to those produced by relatively weak tone pips or clicks from an external source. The cochlear microphonic (CM) produced by a pulse shows a prominent ringing at 5–10 kHz in different cats; the amplitude of the N1 response exhibits a broad maximum, for constant‐amplitude pulses, at pulse widths of 20–60 μs. This variation of N1 response amplitude with pulse width is similar to that of a high‐pass filter with a cutoff frequency at the dominant frequency of the CM.

Auditory responses in cats produced by pulsed ultrasound.

Auditory-nerve responses and cochlear microphonics are produced in cats by pulsed 5-MHz ultrasonic energy from a transducer placed against the dura mater. The pulses must be relatively intense (approximately 30 W/cm2) to produce a response, but can be sufficiently brief (less than 70 microsecond) that the brain tissue is not observably heated. The cats apparently respond to radiation pressure transients accompanying the absorption of the ultrasound in the brain tissue. Both the amplitude and latency of the N1 neural responses to the ultrasound can be matched to those produced by relatively weak tone pips or clicks from an external source. The cochlear microphonic (CM) produced by a pulse shows a prominent ringing at 5-10 kHz in different cats; the amplitued of the N1 response exhibits broad maximum, for constant amplitude pulses, a pulse widths of 20-60 microsecond. This variation of N1 response amplitude with pulse width is similar to that of a high-pass filter with a cutoff frequency at the dominent frequency of the CM, which is tentatively identified with a ringing frequency of the skull.

Auditory nerve and brain-stem evoked responses in normal, autistic, minimal brain dysfunction and psychomotor retarded children

In several forms of abnormal behavior in infants and children, the cause is either uncertain or an unspecified organic brain lesion is assumed to be present. Evoked averaged auditory nerve and brain-stem responses to click stimuli were recorded from groups of normal, autistic, minimal brain dysfunction (MBD) and psychomotor retarded children in order to search for electrophysiological evidence for a brain lesion. Most of the control children were awake and most of the experimental children were in sedated sleep but records were also made from some children while awake and again in the same children when asleep and no change was seen in the response traces. In several of the autistic children there were no electrophysiological responses (indicating a profound hearing loss). In several of the psychomotor retarded children the responses from the region of the inferior colliculus were absent. In the other children in the experimental group, auditory nerve and brain-stem responses were recorded with normal response thresholds but abnormal in other ways: the latency of the auditory nerve response was longer than in the normal children. Also, brain-stem transmission time, measured as the time interval from the negative peak of the auditory nerve response to the positive trough of the response from the inferior colliculus, was shortest in the control group and longest in the MBD group. These results thus present electrophysiological evidence for the existence of an organic brain lesion in these children, at least in the brain-stem regions concerned with auditory function. Together with evidence from other studies, there is support for the conclusion that most of the abnormal behavior seen in these patients is due to a diffuse brain lesion.

Properties of auditory nerve responses in absence of outer hair cells.

1. Recordings were made from chinchilla auditory nerve fibers after portions of the cochlear outer hair cell (OHC) population were destroyed with the antibiotic kanamycin. In most cases the inner hair cell (IHC) population was completely preserved as determined by phase-contrast microscopy. We presume that the remaining IHCs are functionally normal, and thus that recordings obtained from fibers originating from the lesioned cochlear segment reflect IHC behavior. 2. Behavioral thresholds were measured for all animals both before and after the production of the cochlear lesion. The audiograms and the histological evaluation of the ears were the basis for assessing whether a particular fiber originated in a normal, pathological (shifted threshold; IHC only), or border region. These criteria also identified the animals that sustained IHC damage together with the destruction of part of the OHC population. Only the data obtained from those fibers which probably originated from the OHC-free segment of the cochlea are considered in detail. 3. Fibers whose characteristic frequency (CF) identified them as belonging to the normal (audiometrically and histologically) region, were found to be normal in all respects. 4. Fibers from the border region (where the audiogram has a steep slope between normal and hearing-loss regions probably corresponding to the segment where OHC loss progresses from less than 10% to more than 90%) had very complex response patterns. Their frequency threshold curves (FTC) showed great variability. In general, the closer the fiber was to the fully developed lesion, the more abnormal its FTC became. 5. Those units that were concluded to have originated from the OHC-free part of the cochlea could be divided into three categories on the basis of the shape of their FTCs. A small fraction had very broad tuning (9%). The majority (53%) had approximately normal tail segment, normal bandwidth of the tip segment, and highly elevated threshold at CF. A group of fibers (38%) could not be assigned a CF. Probably the FTC of most of these latter fibers are similar to those of the previous group, but the sharply tuned short tip segment was either missed or was not reachable on account of its extremely high threshold level. 6. Such indexes of fiber response as latency, spontaneous rate, and time pattern (PST histograms) were not affected by the loss of OHCs. 7. On the basis of the data and of the assumptions made it was suggested that outer hair cells provide a frequency-dependent sensitizing influence to the inner hair cells. The frequency dependence could best be expressed as a flat-topped band pass characteristic.

Auditory-nerve response from cats raised in a low-noise chamber.

A litter of four cats, born and raised in a soundproofed chamber, was studied in an attempt to determine which, if any, features of the auditory-nerve response from routinely available cats might be due to the chronic effects of noise exposure. Two features of routine-normal response were especially suspect in this regard: (1) a "notch" in the distribution of single-unit thresholds centered at characteristic frequencies (CF's) near 3 kHz and (2) a compression of the distribution of rates of spontaneous discharge for units with CF above 10 kHz. A third feature of response in routine animals was the presence of a small number (roughly 10%) of units with virtually no spontaneous discharge and very high thresholds, sometimes 80 dB less sensitive than high-spontaneous units of similar CF. In the data from chamber-raised animals, the high-spontaneous units showed exceptionally low thresholds at all CF regions, however, there were signs of the midfrequency notch in the threshold distribution of at least two of these animals. The compression of the spontaneous rate distribution was not seen in any of the three most sensitive animals. The data suggest that there is a significant amount of "normal pathology" in the high-CF units from routine animals. Low-spontaneous, high-threshold units were present in all four chamber-raised ears with the same characteristics as in routine animals (exceptionally narrow tuning curves and exceptionally low maximum discharge rates) and at roughly the same percentage of the unit sample. A class of units with medium spontaneous rates and intermediate thresholds could also be identified. The possible significance of a classification of auditory-nerve units according to spontaneous rate is discussed.

Responses of Auditory Nerve Fibres to Noise Stimuli Show Cochlear Nonlinearities

The frequency selectivity of the peripheral auditory frequency analyser was determined on the basis of recordings from single auditory nerve fibres in the rat, using pseudorandom noise as stimulus. The impulse response function was arrived at by cross-correlating period histograms of the recorded activity with the noise. Frequency transfer functions were computed by Fourier transformation of the derived impulse response function. The impulse response function was observed to become shorter with increasing stimulus intensity. A concomitant widening of the frequency transfer function and a downward shift in the centre frequency were seen. Since the method used is insensitive to nonlinearities, the derived impulse and frequency respone functions are assumed to be valid approximations of the basilar membrane properties. The results, then, support the hypothesis that the basilar membrane is a nonlinear frequency analyser whose selectivity decreases as sound intensity increases.

Effects of Eustachian Tube Ligation on Auditory Nerve Responses to Clicks

Auditory nerve responses to condensation and rarefaction clicks were recorded from the external ear canal of cats using a closed acoustic system. Repeated control recordings from both ears formed a baseline for each of four animals used in this study. After a baseline had been established, the Eustachian tube on one side was ligated and serial recordings of N1 responses were performed for up to 140 days postligation. By comparing the shift which occurred in the N1 latency-vs-click level plots after ligation, the equivalent hearing loss was determined. In all cases where the Eustachian tube was successfully ligated, the loss was progressive for the first 20 days, then usually showed some transitory improvement. The loss stabilized after 60 days, varying from 15 to 40 dB in different animals. In addition to N1 recordings, serial tympanograms were also measured. These indicated negative middle ear pressure in the first two days postligation and the presence of middle ear fluid by one week postligation.

Noninvasive Chronic Recording of Auditory Nerve Potentials

A system is described which allows repeated, noninvasive recording of auditory nerve responses in the cat to transient acoustic stimuli using a closed acoustic system. N1 responses to clicks are recorded from a stainless steel ring electrode at the end of a hollow earbar, the tapered end of which is made of insulating plastic. Acoustic stimuli are generated by a dynamic earphone coupled to the earbar. A calibrated probe microphone is also imcorporated into the earbar to measure sound pressure near the tympanic membrane. This allows better stimulus control than is available with free-field systems. To facilitate insertion of the earbar, meatoplasties were performed on all animals. Responses recorded with this system in anesthetized cats are described and compared with those recorded at the round window. Good repeatability of measurements is described for an animal population of 20 domestic cats over a period of several months. For some of these animals, response amplitude varied from one session to another, but response latency, especially for condensation clicks, was consistent. By comparing statistics of multiple measurement of both N1 amplitude and latency for rarefaction and condensation clicks, it is concluded that the N1 latency vs click-level function for condensation clicks provides the most reliable measure of the cat's auditory nerve function.

Correlations between psychophysical magnitude estimates and simultaneously obtained auditory nerve, brain stem and cortical responses to click stimuli in man

Responses from the auditory nerve, brain stem auditory nuclei and cortex, as well as subjective responses to click stimuli at 10 intensities, were recorded simultaneously in the same human subjects. For various measures of the responses, the power-law exponents of their intensity functions were calculated, along with their statistical significances. The electrophysiological and psycho-physical functions were compared for similarity. On average, the exponents of the intensity functions of amplitudes of the auditory nerve and earlier brain stem responses were highly significant, showing similarity across subjects and similarity with the exponents of the subjective estimates. However, a closer examination proved this similarity to be superficial, since magnitude estimates showed an appreciable intersubject and intersession variability while the auditory nerve and brain stem responses were approximately constant. All other electric response measures either had exponents which were not significant or showed even poorer correlation with the subjective response. It is proposed that the type of electrical activity recorded in this study may not be the proper set of neural parameters which give rise to the loudness estimate.

Usher's syndrome: electrophysiological tests of the visual and auditory systems.

The symptoms of Usher's syndrome - congenital hearing impairment and tapetoretinal dystrophy - are difficult to detect in young children. The electroretinogram (ERG), visual evoked potential (VEP), auditory nerve and brain-stem responses as well as the cochlear microphonic potentials were used to evaluate the defects of the visual and auditory systems in 20 patients, 3-45 years of age. This study demonstrates the usefulness, reliability and convenience of the electrophysiological tests for early diagnosis and functional evaluation of Usher's syndrome.

Single auditory nerve responses to acoustic and electric stimuli

The characteristics of single auditory nerve fiber responses to low frequency acoustic stimuli and sinusoidal electric current injected in the scala media were investigated in guinea pigs. A glass microelectrode in the scala media of the second turn was used to measure cochlear microphonics or to inject sinusoidal electrical currents up to 8μA p-p. At 40 Hz, the phase lag between the maximum of the neural response and the positive peak of the voltage change inside the scala media was −3±65° for acoustic stimuli and 127±47° for electrical stimuli. Thresholds were as low as 0.5μA p-p for electrical stimuli and the degree of phase-locking increased proportionally (slope=0.94) with stimulus amplitude. These data, together with the finding that electrical stimuli modulate neural responses to acoustic stimulation, suggest that these electric stimuli act upon hair cells rather than directly upon the nerve fibers. [Research supported by NIH Grant NS 08193 and the Deafness Research Foundation.]

Auditory nerve and brain stem responses to sound stimuli at various frequencies.

The responses of the auditory nerve and brain stem auditory nuclei have been recorded with ear lobe and scalp electrodes in response to unfiltered click stimuli which excite mainly high-frequency nerve fibers. In order to overcome this limitation, normal subjects and patients with hearing loss were stimulated with third-octave filtered clicks with center frequencies of 250, 500 and 1 000 Hz. Analysis of the responses suggests that these lower-frequency stimuli excite lower-frequency cochlear nerve fibers. It is concluded that low-frequency filtered clicks can be used to complement unfiltered clicks in order to obtain a more complete objective audiogram.

Model study of nonlinear response to paired clicks

T. J. Goblick, Jr. and R. R. Pfeiffer [J. Acoust. Soc. Am. 46, 924–938 (1969)] found various nonlinear effects in auditory-nerve responses of cats to paired clicks. We have been able to reproduce these effects in a nonlinear model of the basilar membrane [M. R. Schroeder, Proc. IEEE 63 (9), 1332–1350 (1975); J. L. Hall, J. Acoust. Soc. Am.56, 1818–1828 (1974)] in greater detail than has been possible in previous studies [D. O. Kim, C. E. Molnar, and R. R. Pfeiffer, J. Acoust. Soc. Am. 54, 1517–1524 (1973)]. To investigate the origin of these nonlinear effects, selected portions of the membrane model were made linear. In this manner it was found that the effects arise in two regions of the membrane model. The strongest contribution occurs just basal to the place of observation (characteristic frequency f=fc). A weaker contribution occurs just basal to the place with characteristic frequency f=2fc.

Human auditory nerve responses: a comparison of three commonly used recording sites.

The computer-averaged N1 response from humans to click stimuli was compared at three recording sites -- dermal and subdermal electrodes in the ear canal, and a transtympanic electrode on the promontory. A total of 26 subjects, some of whom were measured at all three sites, participated in 47 experiments. The promontory electrodes were consistently more sensitive and yielded cleaner N1 waveforms than did the ear canal sites. In considering the clinical use of N1 averaging, the advantages of the transtympanic method outweigh the risks. While a few persons yield reasonable good ear canal potentials, the reliablitity of this method in general application is inadequate and unpredictable.