Bradford J. May

Pinna-based spectral cues for sound localization in cat

The directional dependence of the transfer function from free field plane waves to a point near the tympanic membrane (TM) was measured in anesthetized domestic cats. A probe tube microphone was placed approximately 3 mm from the TM from beneath the head in order to keep the pinna intact. Transfer functions were computed as the ratio of the spectrum of a click recorded near the TM to the spectrum of the click in freefield. We analyze the transfer functions in three frequency ranges: low frequencies (less than 5 kHz) where interaural level differences vary smoothly with azimuth; midfrequencies (5-18 kHz) where a prominent spectral notch is observed; and high frequencies (greater than 18 kHz) where the transfer functions vary greatly with source location. Because no two source directions produce the same transfer function, the spectrum of a broadband sound at the TM could serve as a sound localization cue for both elevation and azimuth. In particular, we show that source direction is uniquely determined, for source directions in front of the cat, from the frequencies of the midfrequency spectral notches in the two ears. The validity of the transfer functions as measures of the acoustic input to the auditory system is considered in terms of models of sound propagation in the ear canal.

Role of the dorsal cochlear nucleus in the sound localization behavior of cats

The role of the dorsal cochlear nucleus (DCN) in directional hearing was evaluated by measuring sound localization behaviors before and after cats received lesions of the dorsal and intermediate acoustic striae (DAS/IAS). These lesions are presumed to disrupt spectral processing in the DCN without affecting binaural time and level difference cues that exit the cochlear nucleus via the ventral acoustic stria. Prior to DAS/IAS lesions, cats made accurate head orientation responses toward sound sources in the frontal sound field. After a unilateral DAS/IAS lesion, subjects showed increased errors in the azimuth and elevation of their responses; in addition, the final orientation of head movements tended to be more variable. Largest deficits in response elevation were observed in the hemifield that was ipsilateral to the lesion. When a second lesion was placed in the opposite DAS/IAS, increased orientation errors were observed throughout the frontal field. Nonetheless, bilaterally lesioned cats showed normal discrimination of changes in sound source location when tested with a spatial acuity task. These findings support previous interpretations that the DCN contributes to sound orientation behavior, and further suggest that the identification of absolute sound source locations and the discrimination between spatial locations involve independent auditory processing mechanisms.

Amplification in the rehabilitation of unilateral deafness: speech in noise and directional hearing effects with bone-anchored hearing and contralateral routing of signal amplification.

Objective: Vibromechanical stimulation with a semi-implantable bone conductor (Entific BAHA device) overcomes some of the head-shadow effects in unilateral deafness. What specific rehabilitative benefits are observed when the functional ear exhibits normal hearing versus moderate sensorineural hearing loss (SNHL)? Design: The authors conducted a prospective trial of subjects with unilateral deafness in a tertiary care center. Patients: This study comprised adults with unilateral deafness (pure-tone average [PTA] >90 dB; Sp.D. <20%) and either normal monaural hearing (n = 18) or moderate SNHL (PTA = 25-50 dB: Sp.D. >75%) in the contralateral functional ear (n = 5). Interventions: Subjects were fit with contralateral routing of signal (CROS) devices for 1 month and tested before (mastoid) implantation, fitting, and testing with a bone-anchored hearing aid (BAHA). Outcome Measures: Outcome measures were: 1) subjective benefit; 2) source localization tests (Source Azimuth Identification in Noise Test [SAINT]); 3) speech discrimination in quiet and in noise assessed with Hearing In Noise Test (HINT) protocols. Results: There was consistent satisfaction with BAHA amplification and poor acceptance of CROS amplification. General directional hearing decreased with CROS use and was unchanged by BAHA and directional microphone aids. Relative to baseline and CROS, BAHA produced significantly better speech recognition in noise. Twenty-two of 23 subjects followed up in this study continue to use their BAHA device over an average follow-up period of 30.24 months (range, 51-12 months). Conclusion: BAHA amplification on the side of a deaf ear yields greater benefit in subjects with monaural hearing than does CROS amplification. Advantages likely related to averting the interference of speech signals delivered to the better ear, as occurs with conventional CROS amplification, while alleviating the negative head-shadow effects of unilateral deafness. The advantages of head-shadow reduction in enhancing speech recognition with noise in the hearing ear outweigh disadvantages inherent in head-shadow reduction that can occur by introducing noise from the deaf side. The level of hearing impairment correlates with incremental benefit provided by the BAHA. Patients with a moderate SNHL in the functioning ear perceived greater increments in benefit, especially in background noise, and demonstrated greater improvements in speech understanding with BAHA amplification.

Spontaneous activity in the inferior colliculus of CBA/J mice after manipulations that induce tinnitus.

Several physiological studies have linked experimentally induced tinnitus to increases in the spontaneous activity of auditory neurons. These results have led to the proposal of hyperactivity models of tinnitus in which elevated neural activity in the absence of auditory stimulation is perceived as phantom sound. Such models are appealing in their simplicity but remain controversial because a generalized elevation of spontaneous rates may not be observed after treatments that induce tinnitus in humans and experimental animals. Our study addressed these issues by characterizing the effects of common methods of tinnitus induction on spontaneous activity in the central nucleus of the inferior colliculus (ICC). The ICC is an interesting structure in tinnitus research because its diverse inputs include putative generator sites in the dorsal cochlear nucleus, as well as brainstem sources that appear to remain normal after tinnitus induction. Groups of CBA/J mice were subjected to one of three induction methods: bilateral or unilateral sound exposure, and acute salicylate intoxication. Relative to normal baselines, bilaterally exposed mice showed increases in the spontaneous rates of neurons with tuning near the exposure frequency. When the sample was separated into physiologically defined response classes, exposure effects were strongest among neurons with broad excitatory bandwidths. By contrast, salicylate decreased the spontaneous rates of low-frequency neurons with transient sound-evoked activity. Our results suggest that the disordered processes of hearing that give rise to tinnitus do not involve a pervasive elevation of spontaneous activity or a single mode of induction.

Synaptic alterations at inner hair cells precede spiral ganglion cell loss in aging C57BL/6J mice

Hearing deficits have often been associated with loss of or damage to receptor hair cells and/or degeneration of spiral ganglion cells. There are, however, some physiological abnormalities that are not reliably attributed to loss of these cells. The afferent synapse between radial fibers of spiral ganglion neurons and inner hair cells (IHCs) emerges as another site that could be involved in transmission abnormalities. We tested the hypothesis that the structure of these afferent terminals would differ between young animals and older animals with significant hearing loss. Afferent endings and their synapses were examined by transmission electron microscopy at approximately 45% distance from the basal end of the cochlea in 2-3 month-old and 8-12 month-old C57BL/6J mice. The number of terminals in older animals was reduced by half compared to younger animals. In contrast, there was no difference in the density of SGCs between the age groups. Older animals featured enlarged terminals and mitochondria and enlarged postsynaptic densities and presynaptic bodies. These morphological changes may be a combination of pathologic, adaptive and compensatory responses to sensory dysfunction. Improved knowledge of these processes is necessary to understand the role of afferent connectivity in dysfunction of the aging cochlea.

Auditory Processing of Spectral Cues for Sound Localization in the Inferior Colliculus

The head-related transfer function (HRTF) of the cat adds directionally dependent energy minima to the amplitude spectrum of complex sounds. These spectral notches are a principal cue for the localization of sound source elevation. Physiological evidence suggests that the dorsal cochlear nucleus (DCN) plays a critical role in the brainstem processing of this directional feature. Type O units in the central nucleus of the inferior colliculus (ICC) are a primary target of ascending DCN projections and, therefore, may represent midbrain specializations for the auditory processing of spectral cues for sound localization. Behavioral studies confirm a loss of sound orientation accuracy when DCN projections to the inferior colliculus are surgically lesioned. This study used simple analogs of HRTF notches to characterize single-unit response patterns in the ICC of decerebrate cats that may contribute to the directional sensitivity of the brains spectral processing pathways. Manipulations of notch frequency and bandwidth demonstrated frequency-specific excitatory responses that have the capacity to encode HRTF-based cues for sound source location. These response patterns were limited to type O units in the ICC and have not been observed for the projection neurons of the DCN. The unique spectral integration properties of type O units suggest that DCN influences are transformed into a more selective representation of sound source location by a local convergence of wideband excitatory and frequency-tuned inhibitory inputs.

Sound orientation behavior in cats. II. Mid-frequency spectral cues for sound localization

The cats head-related transfer function creates a directionally dependent mid-frequency notch in the amplitude spectrum of a broadband sound as the stimulus propagates to the tympanic membrane [Rice et al., Hear. Res. 58, 132-152 (1992)]. Our previous behavioral studies [May and Huang, J. Acoust. Soc. Am. 100, 1059-1069 (1996)] have indicated that the cats sound-evoked orientation responses are well directed to the azimuth and elevation of broadband noise bursts in the frontal sound field, where pinna-based spectral notches are prominent and change systematically with sound direction. In the present study, the importance of mid-frequency directional cues in the cats sound localization behavior was further evaluated by manipulating the frequency and bandwidth of orientation stimuli. The accurate pattern of orientation behavior seen previously with bursts of broadband noise was relatively unaffected when stimulus bandwidth was decreased to mid-frequency bandpass noise of 5-18 kHz. In contrast, poorly directed head orientation responses were observed in tests with high-pass noise (> 18 kHz) and mid-frequency pure tones. When tested with narrow bands of mid-frequency noise, cats oriented toward the spatial location where HRTF-filtering properties most closely matched the stimulus spectrum. These results suggest that important sound localization cues are derived from mid-frequency spectral features of the cats HRTF.

Effects of bilateral olivocochlear lesions on vowel formant discrimination in cats

Operant conditioning procedures were used to measure the effects of bilateral olivocochlear lesions on the cats discrimination thresholds for changes in the second formant frequency (deltaF2) of the vowel /epsilon/. Three cats were tested with the formant discrimination task under quiet conditions and in the presence of continuous broadband noise at signal-to-noise ratios (S/Ns) of 23, 13, and 3 dB. In quiet, vowel levels of 50 and 70 dB produced average deltaF2s of 42 and 47 Hz, respectively, and these thresholds did not change significantly in low levels of background noise (S/Ns = 23 and 13 dB). Average deltaF2s increased to 94 and 97 Hz for vowel levels of 50 and 70 dB in the loudest level of background noise (S/N = 3 dB). Average deltaF2 thresholds in quiet and in lower noise levels were only slightly affected when the olivocochlear bundle was lesioned by making bilateral cuts into the floor of the IVth ventricle. In contrast, post-lesion deltaF2 thresholds in the highest noise level were significantly larger than pre-lesion values; the most severely affected subject showed post-lesion discrimination thresholds well over 200 Hz for both 50 and 70 dB vowels. These results suggest that olivocochlear feedback may enhance speech processing in high levels of ambient noise.

Spectral cues for sound localization in cats: Effects of frequency domain on minimum audible angles in the median and horizontal planes

Rice et al. [Hear. Res. 58, 132-152 (1992)] classified directional properties of the cats head-related transfer function (HRTF) into three frequency domains. Low frequencies (< 5 kHz) display a broad azimuth-sensitive spectral peak that establishes interaural level differences, mid frequencies (5-18 kHz) are marked with a single deep spectral notch that changes in frequency as a function of both azimuth and elevation, and high frequencies (18-50 kHz) exhibit a complex pattern of peaks and notches that shows extensive but less systematic changes with sound location. Spectral cues conveyed by the mid frequencies of broadband sounds are important in tasks that require cats to identify the actual location of acoustic stimuli [Huang and May, J. Acoust. Soc. Am. (in press)]. The present study investigates how directional cues conveyed by the mid- and high-frequency spectrum of the HRTF influence the cats ability to discriminate between sound locations. Thresholds for spatial acuity were measured as minimum audible angles (MAAs) [Mills, J. Acoust. Soc. Am. 30, 237-246 (1958)] at positive azimuths in the interaural horizontal plane and at positive and negative elevations in the median vertical plane. The frequency domain of the noise burst had little effect on MAAs in the horizontal plane, but removal of high-frequency spectral information significantly increased thresholds at positive and negative elevations in the median plane. These results suggest that cats are sensitive to directional properties of the HRTF at frequencies above 18 kHz and may use this information to detect small changes in sound source elevation.

Behavioral investigation of some possible effects of the central olivocochlear pathways in transgenic mice.

This study investigated the auditory behaviors of transgenic mice with deletions of alpha9 nicotinic acetylcholine receptor subunits. In the normal mammalian cochlea, the mechanical properties of outer hair cells are modified by the release of acetylcholine from olivocochlear efferent terminals. Electrophysiological correlates of this efferent feedback have not been demonstrated in alpha9 knockout mice, presumably because they are mediated by alpha9 receptors. Previous studies have associated lesions of olivocochlear pathways with hearing impairments in background noise. The prediction that alpha9 knockout mice would show similar deficits was tested by collecting psychophysical thresholds for tone detection and intensity discrimination from knockout mice, within-strain control subjects, and CBA/CaJ mice. Comparable performance was observed for the subject groups in quiet and in continuous background noise. The preservation of auditory function in alpha9 knockout mice suggests that central efferent pathways work in combination with the peripheral olivocochlear system to enhance hearing in noise, and may compensate for profound manipulations of peripheral feedback in highly routine testing procedures. An intriguing possibility is that these central mechanisms include the brainstem collaterals of olivocochlear neurons since their post-synaptic targets do not express alpha9 receptors and therefore are likely to maintain their effects in alpha9 knockout mice.