Search

The von Bekesy Medal is presented to individuals, irrespective of nationality, age, or society affiliation, who have made outstanding contributions to the area of psychological or physiological acoustics as evidenced by publication of research results in professional journals or by other accomplishments in the fields.

Basilar membrane (BM) vibration was measured using a displacement measuring interferometer for single-tone and two-tone suppression (2TS) paradigms in the 6-9 kHz region of sensitive chinchilla cochleae that had gains near or better than 60 dB. Based on prior studies of basilar membrane vibration, three significant differences remain between BM and auditory nerve (AN) 2TS responses: (1) suppression thresholds in the tail of tuning curves were much higher in BM than the auditory nerve (AN); (2) rates of suppression were significantly higher in AN than BM; and (3) the amplitude of vibration with low-frequency suppressors was always greater than the single-tone displacement rendering it impossible to explain 2TS rate suppression in the AN. The first two differences are eliminated by the results of the present study while the third remains. Suppression amplitudes greater than 40 dB and rates of suppression larger than 2.5 dB/dB were found for low-frequency suppressors. A correlation between both the gain and nonlinearity of the cochlea and 2TS properties indicates that when sensitive cochleae are studied. The third difference between BM and AN behavior could be strictly a function of the high-pass filter characteristic of the inner hair cells.

The vibration of the basilar membrane in the 6-9 kHz region in the chinchilla cochlea has been studied using a displacement sensitive interferometer. Displacements of 0.7-1.4 nm at 0 dB sound pressure level have been obtained. At the characteristic frequency (CF), rate-of-growth (ROG) functions computed as the slope of input-output (IO) functions can be as low as 0.1 dB/dB. IO functions for frequenciesCF have ROGs near 0 dB/dB and can have notches characterized by both negative slopes and expansive ROGs, i.e.,1 dB/dB. For frequencies0.6*CF, ROGs1.2 dB/dB were found. Cochlear gain is shown to be greater than 60 dB in sensitive preparations with a single cochlea having nearly 80 dB gain. The compressive nature of the cochlea remains at all levels though it is masked at frequenciesCF when the amplitude of a compression wave exceeds that of the traveling wave. The compression wave produces the plateau region of the mechanical response at high intensities and has a nearly constant phase versus frequency function implying a high velocity. The summation of the traveling and compression waves explains the occurrence of the notches in both the IO and iso-intensity functions. Vibration of the osseous spiral limbus may alter the drive to inner hair cells.

Multicomponent stimuli consisting of two to seven tones were used to study suppression of basilar-membrane vibration at the 3-4-mm region of the chinchilla cochlea with a characteristic frequency between 6.5 and 8.5 kHz. Three-component stimuli were amplitude-modulated sinusoids (AM) with modulation depth varied between 0.25 and 2 and modulation frequency varied between 100 and 2000 Hz. For five-component stimuli of equal amplitude, frequency separation between adjacent components was the same as that used for AM stimuli. An additional manipulation was to position either the first, third, or fifth component at the characteristic frequency (CF). This allowed the study of the basilar-membrane response to off-CF stimuli. CF suppression was as high as 35 dB for two-tone combinations, while for equal-amplitude stimulus components CF suppression never exceeded 20 dB. This latter case occurred for both two-tone stimuli where the suppressor was below CF and for multitone stimuli with the third component=CF. Suppression was least for the AM stimuli, including when the three AM components were equal. Maximum suppression was both level- and frequency dependent, and occurred for component frequency separations of 500 to 600 Hz. Suppression decreased for multicomponent stimuli with component frequency spacing greater than 600 Hz. Mutual suppression occurred whenever stimulus components were within the compressive region of the basilar membrane.

The response of chinchilla basilar membrane in the basal region of the cochlea to multicomponent (1, 3, 5, 6, or 7) stimuli was studied using a laser interferometer. Three-component stimuli were amplitude-modulated signals with modulation depths that varied from 25% to 200% and the modulation frequency varied from 100 to 2000 Hz while the carrier frequency was set to the characteristic frequency of the region under study (approximately 6.3 to 9 kHz). Results indicate that, for certain modulation frequencies and depths, there is enhancement of the response. Responses to five equal-amplitude sine wave stimuli indicated the occurrence of nonlinear phenomena such as spectral edge enhancement, present when the frequency spacing was less than 200 Hz, and mutual suppression. For five-component stimuli, the first, third, or fifth component was placed at the characteristic frequency and the component frequency separation was varied over a 2-kHz range. Responses to seven component stimuli were similar to those of five-component stimuli. Six-component stimuli were generated by leaving out the center component of the seven-component stimuli. In the latter case, the center component was restored in the basilar-membrane response as a result of distortion-product generation in the nonlinear cochlea.

Basilar membrane (BM) responses to two types of broadband stimuli-clicks and Schroeder-phase complexes--were recorded at several sites at the base of the chinchilla cochlea. Recording sites (characteristic frequency, CF, in the range of 5.5-18 kHz) span the 1-4-mm basal region of the basilar membrane. BM responses to clicks consisted of undamped oscillations with instantaneous frequency that increased over time until it reached a value around CF. The time constant of this glide is CF dependent. Throughout the entire region under study, BM vibration exceeded umbo motion by up to 60 dB. Nonlinear properties of BM responses to clicks resemble those found in the more studied 8-10-kHz region. Amplitude spectra of Schroeder-phase complex stimuli, which consist of a series of sinusoidal components summed in negative (-SCHR) and positive Schroeder phase (+SCHR), are flat. The envelope of BM responses to +SCHR stimuli contains valleys, or dips, that are wider than those found in responses to the -SCHR stimuli. Hence, BM responses to the former stimuli are "peakier" than responses to the latter. Differences in response waveforms are less obvious in linear cochleae. Suppression of a near-CF tone by -SCHR stimuli was larger than that evoked by +SCHR stimuli.

Measurements from the 1-4-mm basal region of the chinchilla cochlea indicate the basilar membrane in the hook region (12-18 kHz) vibrates essentially as it does more apically, in the 5-9-kHz region. That is, a compressive nonlinearity in the region of the characteristic frequency, amplitude-dependent phase changes, and a gain relative to stapes motion that can attain nearly 10,000 at low levels. The displacement at threshold for auditory-nerve fibers in this region (20 dB SPL) was approximately 2 nm. Measurements were made at several locations in individual animals in the longitudinal and radial directions. The results indicate that there is little variability in the phase of motion radially and no indication of higher-order modes of vibration. The data from the longitudinal studies indicate that there is a shift in the location of the maximum with increasing stimulus levels toward the base. The cochlear amplifier extends over a 2-3-mm region around the location of the characteristic frequency.

Amplitude modulated (AM) signals have often been used as precisely defined partial analogs of speech sounds. This study considers the response to an AM complex with 200% sinusoidal modulation, that is, the amplitudes of the three AM components are equal. By varying the carrier frequency across the entire frequency range of unit response, it is shown that units in the cochlear nucleus of cat are relatively insensitive to variation in the carrier frequency, which is to say that population response to an AM signal at a fixed locus will be widespread. These stimuli and procedures result in the presentation of both harmonic and inharmonic complexes, and thus permit assessment of neural responses for the information needed to make spectral or time-domain pitch matches. It is shown that the reciprocals of the modes (favored intervals) in the interspike interval histogram reflect the first effect of pitch shift, which is defined psychophysically as a proportional shift in pitch to the change in carrier frequency. In particular, interspike intervals of units with a widespread spectral response provide a basis to explain phase and dominant component pitch behavior that early narrow-band pitch theories found problematical. The amplitude of phase locking to individual AM components varies systematically though there are some unexplained variations across the frequency-intensity plane that could be due to combination tones. The unit response to a quasifrequency modulated (QFM) stimulus shows that if pitch is based on interspike intervals, it would remain the smae as pitch for an AM signal. The magnitude of the synchrony response to QFM stimuli is less than to AM stimuli for the majority of cochlear nucleus units; however, there are exceptions.

A composite model of the auditory periphery, based upon a unique analysis technique for deriving filter response characteristics from cat auditory-nerve fibers, is presented. The model is distinctive in its ability to capture a significant broadening of auditory-nerve fiber frequency selectivity as a function of increasing sound-pressure level within a computationally tractable time-invariant structure. The output of the model shows the tonotopic distribution of synchrony activity of single fibers in response to the steady-state vowel [e] presented over a 40-dB range of sound-pressure levels and is compared with the population-response data of Young and Sachs (1979). The model, while limited by its time invariance, accurately captures most of the place-synchrony response patterns reported by the Johns Hopkins group. In both the physiology and in the model, auditory-nerve fibers spanning a broad tonotopic range synchronize to the first formant (F1), with the proportion of units phase-locked to F1 increasing appreciably at moderate to high sound-pressure levels. A smaller proportion of fibers maintain phase locking to the second and third formants across the same intensity range. At sound-pressure levels of 60 dB and above, the vast majority of fibers with characteristic frequencies greater than 3 kHz synchronize to F1 (512 Hz), rather than to frequencies in the most sensitive portion of their response range. On the basis of these response patterns it is suggested that neural synchrony is the dominant auditory-nerve representation of formant information under "normal" listening conditions in which speech signals occur across a wide range of intensities and against a background of unpredictable and frequently intense acoustic interference.

Distortion product otoacoustic emissions (DPOAEs) and basilar membrane (BM) vibration were measured simultaneously in the 6-9 kHz region of chinchilla cochleae. BM-Input-Output functions in a two-tone paradigm behaved similarly to DPOAEs for the 2f1-f2 component, nonmonotonic growth with the intensity of the lower frequency primary and a notch in the functions around 60 dB SPL. Ripples in frequency functions occur in both BM and OAE curves as a function of the distortion frequency. Optimum f2/f1 ratios for DPOAE generation are near 1.2. The slope of phase curves indicates that for low f2f1(1.1) the emission source is the place location while for f2f11.1 the relative constancy of the phase function suggests that the place is the nonlinear region of f2, i.e., the wave location. Magnitudes of the DPOAEs increase rapidly above 60 dB SPL suggesting a different source or mechanism at high levels. This is supported by the observation that the high level DPOAE and BM-DP responses remain for a considerable period postmortem.

Differences between cochlear mechanical tuning curves and those of auditory nerve fibers (ANFs) exist. In particular, mechanical transfer functions exhibit a high‐frequency plateau; ANFs frequency threshold curves (FTCs) do not. ANF‐FTCs may have a low‐frequency slope due to a velocity forcing function operating on inner hair cells at low frequencies. Neither basilar membrane velocity nor displacement adequately explain the entire ANF tuning curve. A displacement sensitive interferometer was used to study basilar membrane and spiral limbus mechanics in the 6‐kHz region of the chinchilla cochlea. The spiral limbus vibrates at the same phase as the basilar membrane nearly up to the location’s characteristic frequency. In the plateau region, the limbus appears to vibrate 0 to 20 dB less than the basilar membrane. The basilar membrane/limbus amplitude transfer function has a low‐frequency slope of ∼3 dB/oct at low frequencies and is ∼10 dB lower than the basilar membrane amplitude at 1 kHz. It appears that spiral limbus vibration may contribute to the excitation of the cilia of the inner hair cells. [Work supported by NIDCD grant R01 DC001910.]

Professor Joe Zwislocki is a rare individual who has a strong background in mathematics and hydrodynamics that he applied to the physiological characterization of cochlear function. Realizing the preeminent need for more accurate measurements of cochlear mechanical properties, he undertook the study of properties of the tectorial membrane (TM). His measurements are unique in that they remain the only in vivo measurements of stiffness of the TM. In 1979, he introduced longitudinal effects into cochlear modeling that today has been shown to result in realistic cochlear mechanical transfer curves. A difference between the filter properties seen in auditory nerve fibers and basilar membrane (BM) mechanics led to a leading theory of separate resonances for organ of Corti and TM. He developed a model of how this would affect the shear motion between the TM and reticular lamina. Additional studies were undertaken that substituted recordings in the cells of Hensen for the very difficult direct observation of BM mechanics that epitomize the ingenuity and resourcefulness of Joe Zwislocki.

One of the most challenging dilemmas for theories of speech perception is the lack of invariance between acoustic signal and perception. Due to physical constraints upon articulators, there is a good deal of context‐dependency both in speech production and in the resulting acoustic speech signal. Consequently, the acoustic pattern most closely related to a given speech sound varies dramatically depending on context. Yet, by some means, the perceptual system perceives these unique acoustic events as linguistically equivalent. This phenomenon is observed experimentally as perceptual context effects whereby adjacent speech can modulate perceived identity of a given speech sound. Recent perceptual results suggest that this phenomenon may be governed by general auditory mechanisms [Holt et al., 1996; Lotto et al., 1997; Lotto and Kluender, 1998; Holt, 1999] rather than speech‐specific processes. In the present study, we sought to explore how such context‐dependencies might be encoded. We recorded responses of ventral cochlear nucleus (VCN) neurons of anesthetized chinchillas to nonspeech stimulus targets with adjacent context stimuli that varied in spectral content. Results demonstrate context‐dependent effects of frequency and intensity.

Ever since Ohm and Seebeck, hearing theory has pondered whether “place” or “time” is the dominant means by which acoustic information is encoded in the ear. Over the past decade it has become increasingly clear that, at the level of the auditory nerve, both place and timing mechanisms participate in the representation of complex sounds. As a consequence, attention is now focused on the cochlear nucleus (CN), to which the auditory nerve projects. Within this nuclear complex occurs a remarkable degree of physiological diversity of potential significance to speech coding. The present survey reviews the temporal response properties of the major CN cell types to voiced CV syllables and the potential role played by each in the encoding of speech. All of the major response classes—primarylike, onset, chopper, onset‐chopper, and pauser‐buildup—are capable of synchronizing to the fundamental frequency or first formant. However, the range of acoustic conditions over which they do so, as well as the precision and up...

This paper describes part of the anatomy and physiology of the cochlear nucleus and some of the recent techniques developed to study the structure and function of the cochlear nucleus. The cochlear nucleus is the first region in the brain stem that receives information from the auditory nerve. The use of intracellular markers has made possible a new detail of description that should increase significantly our understanding of how the cochlear nucleus organizes the neural information transmitted to it from the auditory nerve.

A set of experiments was conducted using the Mössbauer effect to determine the vibratory characteristics of the basilar membrane, Reissner's membrane, the malleus, incus, and oval window in squirrel monkey. A few measurements were also made in guinea pig in the basal cochlear region. The nonlinear vibration properties of the basilar membrane are described in detail for the midfrequency region in the squirrel monkey. Only in this region have nonlinear effects been observed. A comparison of mechanical and neural data indicates good qualitative agreement.

Experimental phase data from the Mossbauer‐effect measurements of basilar‐membrane vibration are given. The phase lag increased with intensity for frequencies less than best frequency, but was a decreasing function of intensity above best frequency. A series of linear minimum‐phase representations of the basilar membrane’s motion were also calculated; they show similar trends.

The Mossbauer technique has been applied to the measurement of vibration of the basilar membrane in the squirrel monkey cochlea. Both steady‐state and transient responses have been recorded in the 7–8‐kHz locus of the cochlea. The steady‐state response indicates that the basilar membrane vibrates nonlinearly for frequencies of stimulation near or greater than the characteristic frequency. The nonlinearity can be observed at the lowest levels of stimulation, 70–80 dB SPL, for which measurements could be made. The nonlinearity extends to lower frequencies and the basilar membrane transfer function tends to broaden as SPL is increased. Rapid postmortem changes occur in the cochlea: (1) the amplitude of the transfer ratio (basilar membrane/malleus) decreases 10–15 dB over a period of several hours with a downward shift of 1.5–3 kHz in the characteristic frequency of the basilar membrane at a given location; (2) the low‐frequency slope of the transfer ratio settles to 6 dB/octave by 6 h after death; (3) the slope of the phase of the transfer function increases the characteristic frequency decreases; (4) the basilar membrane vibrates linearly after death. The transient response was studied using acoustic clicks of approximately 150‐μ duration, presented in sequences of 100 000 to 400 000. The transient response has an early component which has a fast decay and a second component which has an extremely slow rate of decay and which displays nonlinear behavior. Preliminary results for the vibration of Reissner's membrane in the 13‐kHz region of the cochlea are also reported.

Recent measurements obtained with a variety of techniques have extended von Békésy's observations of cochlear-partition mechanics including the traveling-wave pattern and the frequency-place relation. Whereas the cochlear fiber has been shown to be sharper than expected, on the basis of Békésy's results, it apparently is not sharp enough to account for the frequency selectivity observed in auditory nerve fibers. The main remaining question is whether the vibration of the basilar membrane is linear or not. Steady-state nonlinearities that disappear rapidly upon death, transient nonlinear response, and two-tone suppression have been observed in the mid-frequency range in one animal species. Whether the linearity-nonlinearity controversy arises from a species difference or a frequency-place difference remains to be resolved.

Measurements of the transient response of the basilar membrane were conducted using the Mossbauer effect on 33 squirrel monkeys using an experimental preparation identical to that of Rhode (1971). The stimuli were acoustic clicks 150 μsec in duration repeated 100 000–400 000 times. The amplitude of the click was varied and the responses of the malleus and of the basilar membrane at a point in the basal turn were measured. The basilar membrane’s click response is oscillatory, with a period near that of the characteristic frequency. The first few response peaks behave almost linearly with stimulus intensity, while the later peaks exhibit a pronounced nonlinearity. This behavior is shown to be consistent with the nonlinearity reported using steady‐state measurement methods (Rhode, 1971). The transient response observed in some of the preparations was very lightly damped; however, a wide range in the damping of the responses was found in the different animals. A progressive increase in the rate of decay of th...

The amplitude and the phase of vibration of the basilar membrane and the bony limbus of the cochlea were measured in living squirrel monkeys using the Mossbauer technique. In the middle ear, the vibration of the malleus (and occasionally the incus) was measured. The Mossbauer technique makes possible the measurement of very small velocities, e.g., 0.2 mm/sec. This sensitivity permits measurement of the motion of the malleus at sound‐pressure levels (SPLs) of 90 to 110 dB and measurement of the motion of the basilar membrane at 70 to 120 dB SPL, depending on the frequency. The basilar membrane vibrates nonlinearly for frequencies which produce the largest deflections at the spot on the basilar membrane under observation. The ratio of the displacement of the basilar membrane to that of the malleus was observed to have the following characteristics: (1) As the frequency is increased from a low value, its amplitude increases at 6 dB/oct until just below the maximum ratio where the slope increases to about 24 dB/oct; (2) the maximum ratio was about 24 dB for the SPLs used; (3) for frequencies above that producing the maximum ratio, the drop‐off rate was approximately 100 dB/oct; (4) the amplitude ratio did not drop off indefinitely but tended to level off; (5) the motion of the basilar membrane differs from the motion of the malleus by 90° at very low frequencies; (6) for frequencies below that producing the maximum ratio, the phase differences between the motion of the basilar membrane and that of the malleus is a linear function of frequency; (7) near the frequency corresponding to the maximum ratio, the phase difference decreases at a faster rate; and (8) the phase difference approaches a constant value (7π 8π or 9π) for high frequencies. Anatomical constraints allowed only a small portion of the basal turn to be studied (6.5–7.5 kHz produced maximum deflection of the basilar membrane in this region).

A quantitative model of the relative motion between the tectorial membrane and reticular lamina of the mammalian cochlea is presented. The model is proposed for the stapes side of the region of maximum deflection during sinusoidal excitation. Based on familiar concepts, the model assumes that basilar membrane deflection causes a sliding motion between the tectorial membrane and the reticular lamina. Equations for the motion between opposing points on these two structures are derived in terms of various cochlear dimensions and of the basilar‐membrane displacement. Average values of these dimensions, obtained from eight cat ears, were used to achieve numerical answers.

Measurements were performed on 25 squirrel monkeys, using an experimental preparation identical to the one used by Rhode to measure frequency response [J. Acoust. Soc. Amer. 49, 1218–1231 (1971)]. The stimuli used were acoustic clicks, about 150 μsec in duration, presented in sequences of 100 000 to 400 000. Clicks with different amplitudes were used, and the responses of the basilar membrane at a point in the basal turn and of the umbo were measured. The results indicate that the click response of the basilar membrane at the points measured is a lightly damped oscillation with a natural frequency between 6 and 8 kHz. There is an initial part of the response that has a faster decay and a later part that has an extremely slow rate of decay. This later part of the response displays a nonlinear behavior for changes in amplitude of the stimulus. Even when the general features of the vibratory pattern were stable for a period of hours, there was usually a marked decrease with time in the number of cycles which could be observed. [Supported by NIH Grants.]