Additive noise can enhance temporal coding in a computational model of analogue cochlear implant stimulation.

Conventional analogue multichannel cochlear implants are unlikely to convey formant information by the fine time structure of evoked discharges. Theoretically, however, the addition of noise to the channel outputs could enhance the representation of formants by time coding. In this study, the potential benefit of noise in analogue coding schemes was investigated using a computer model of cochlear implant stimulation. The cochlear nerve was modelled by the Frankenhauser-Huxley equations. For all five vowels investigated, the optimal addition of noise to the first channel of the simulated implant (200-671 Hz) caused enhancement of the first formant representation (as seen in amplitude spectra of the simulated discharges). For vowels with a low-frequency second formant, clear enhancement of the second formant resulted from the optimal addition of noise to the third channel (1200-2116 Hz). On the basis of the present computational study, additive noise would be expected to enhance the coding of temporal information by the discharges of a single nerve fiber.

Preferential and non-preferential transmission of formant information by an analogue cochlear implant using noise: the role of the nerve threshold.

Previous experiments have shown that, in principle, the addition of noise to any vowel coded by an analogue multichannel cochlear implant can enhance the representation of formant information by the temporal pattern of evoked nerve discharges. The optimal addition of noise to some vowel stimuli caused a largely uniform transmission of all input harmonics, including those related to a formant. But for other vowel stimuli, the optimal addition of noise caused preferential transmission of the harmonic closest to a formant compared with other input harmonics. Such preferential transmission may be useful to a cochlear implantee for formant estimation, but the basis of this transmission is unknown. In the present study, the nature of this preferential transmission was investigated with a set of parallel discriminators (or level-crossing detectors) to determine whether the inherent threshold of a nerve fiber was the main cause of the effect. An explicit threshold was found to account for some but not all of the previously observed preferential transmission. Furthermore, many discriminators were required to obtain preferential transmission. Therefore, preferential transmission of a formant-related harmonic may be best achieved by pre-processing a stimulus and using methods associated with stochastic resonance.

Enhancement of vowel coding for cochlear implants by addition of noise.

Profoundly deaf people, who gain no benefit from conventional hearing aids, can receive speech cues by direct electrical stimulation of the cochlear nerve. This is achieved by an electronic device, a cochlear implant, which is surgically inserted into the ear. Here we show physiological results from the isolated sciatic nerve of the toad Xenopus laevis, used to predict the response of the human cochlear nerve to vowels coded by a cochlear implant. These results suggest that standard analogue cochlear implants do not evoke the patterns of neural excitation that are normally associated with acoustic stimulation. Adding noise to the stimulus, however, enhanced distinguishing features of the vowel encoded by the fine time structure of neural discharges. On the basis of these results, and those concerning stochastic resonance, we advocate a cochlear implant coding strategy in which noise is deliberately added to cochlear implant signals.

Varieties of inhibition in the processing and control of processing in the mammalian cochlear nucleus.

Seven-barrel micropipettes were used to apply drugs microiontophoretically to single units in the dorsal cochlear nucleus (DCN) in chloralose-anaesthetised guinea-pigs. While both agonists and antagonists of putative neurotransmitters in the cochlear nucleus have been investigated in these experiments, the main thrust has been to explore the influence of specific antagonists on cells' spectral and temporal properties, thus elucidating the effects of naturally occurring inhibitory transmitters. At least five types of inhibition appear to be pharmacologically/physiologically separable: (1) Stimulus-evoked tonic "lateral/sideband" inhibition: glycinergic; (blocked by strychnine); responsible for the lateral inhibition of dorsal cochlear nucleus (DCN) type III and IV cells. Strychnine has its predominant effect on sustained (lateral) inhibition compared with the more transient forms of inhibition. Subtraction of receptive field maps enables us to visualise the extent of the inhibitory receptive field. It extends virtually throughout the unit's response field for both these classes but is generally, especially in type IV cells, maximal at the characteristic frequency (CF). This type of inhibition will primarily be responsible for enhancing spectral contrasts in the way that, in the visual system, surround inhibition enhances visual contrast. Furthermore, lateral inhibitory sidebands can "bias" the "working point" of a cell's response so that the dynamic range of effective stimuli and response can be extended. (2) "Background" tonic inhibition: GABAAergic; (blocked by bicuculline). Blocking this inhibition generally results in an increase in the background (i.e., spontaneous) activity. This inhibition is probably responsible for adjusting excitatory-inhibitory contrasts in both spectral and temporal domains. (3) Stimulus-related off-inhibition appears to be neither glycinergic nor GABAAergic. Blocking these receptors actually enhances off-inhibition. Nicotinic cholinergic blockers may have a small effect on off-inhibition, but so far we have not been able to block it entirely. This off-inhibition is important for enhancing temporal contrast. This inhibition must, therefore, be mediated by other transmitters, yet undetermined, or by a local feedback circuit or, less likely, be a membrane-based after-effect of stimulation. (4) Pre-synaptic inhibition, mediated by GABAB receptors presumed to act on primary afferent terminals, thus controlling afferent input to DCN principal cells. (5) Short-latency contralateral inhibition, mediated by glycine.

Behavioural evidence for recovery of auditory function in guinea pigs following kanamycin administration.

Deterioration followed by recovery of behavioural absolute threshold and frequency selectivity has been observed in guinea pigs following kanamycin administration of 200 mg/kg body weight daily for 16 days. Deterioration in function consistently follows a high-to-low frequency pattern and recovery generally occurs at the lowest of the high (8-32 kHz) frequencies affected. The degree of recovery is related to the magnitude of the threshold elevation; where large (40-45 dB) elevations occur initially, the process appears to be partial since threshold recovers only to within 5-12 dB of pre-administration levels. In instances where smaller threshold elevations (5-20 dB) take place initially, recovery can sometimes be complete. However, when threshold elevations of over 50 dB occur, no recovery is apparent. Recovery is relatively slow, taking place over periods of up to 100 days post-kanamycin administration. Hair cell counts have established that the threshold elevation which remains in instances of partial recovery is not related to a reduction in hair cell numbers at the light microscope level.

Auditory processing of complex sounds: an overview.

The past 30 years has seen a remarkable development in our understanding of how the auditory system--particularly the peripheral system--processes complex sounds. Perhaps the most significant has been our understanding of the mechanisms underlying auditory frequency selectivity and their importance for normal and impaired auditory processing. Physiologically vulnerable cochlear filtering can account for many aspects of our normal and impaired psychophysical frequency selectivity with important consequences for the perception of complex sounds. For normal hearing, remarkable mechanisms in the organ of Corti, involving enhancement of mechanical tuning (in mammals probably by feedback of electro-mechanically generated energy from the hair cells), produce exquisite tuning, reflected in the tuning properties of cochlear nerve fibres. Recent comparisons of physiological (cochlear nerve) and psychophysical frequency selectivity in the same species indicate that the ear's overall frequency selectivity can be accounted for by this cochlear filtering, at least in bandwidth terms. Because this cochlear filtering is physiologically vulnerable, it deteriorates in deleterious conditions of the cochlea--hypoxia, disease, drugs, noise overexposure, mechanical disturbance--and is reflected in impaired psychophysical frequency selectivity. This is a fundamental feature of sensorineural hearing loss of cochlear origin, and is of diagnostic value. This cochlear filtering, particularly as reflected in the temporal patterns of cochlear fibres to complex sounds, is remarkably robust over a wide range of stimulus levels. Furthermore, cochlear filtering properties are a prime determinant of the 'place' and 'time' coding of frequency at the cochlear nerve level, both of which appear to be involved in pitch perception. The problem of how the place and time coding of complex sounds is effected over the ear's remarkably wide dynamic range is briefly addressed. In the auditory brainstem, particularly the dorsal cochlear nucleus, are inhibitory mechanisms responsible for enhancing the spectral and temporal contrasts in complex sounds. These mechanisms are now being dissected neuropharmacologically. At the cortical level, mechanisms are evident that are capable of abstracting biologically relevant features of complex sounds. Fundamental studies of how the auditory system encodes and processes complex sounds are vital to promising recent applications in the diagnosis and rehabilitation of the hearing impaired.

Simultaneous dorsal dislocation of interphalangeal joints in a finger.

Simultaneous proximal and distal interphalangeal joint dislocations in a single finger are rare. Even though the joints above and below are routinely evaluated, one of the joint dislocations can be missed. When phalangeal dislocations are suspected, special attention should be given to obtaining sufficient roentgenographic views both before and after reduction. The joints should be thoroughly examined in an attempt to distinguish capsular contraction from soft-tissue interposition. Failure to recognize initial simultaneous dislocations may result in capsular contraction and require surgical intervention.

Early detection of hearing damage in young listeners resulting from exposure to amplified music.

Sixty young subjects in the 15-23 age range, selected on the basis of a questionnaire on the degree of exposure to amplified music, underwent 'high resolution' (4 min per octave) sweep-frequency Békésy tracking audiometry, and measurement of auditory frequency resolution at 4 kHz using the psychophysical comb-filtered noise masking technique. The more exposed groups had 10-15% wider bandwidths than the least exposed, and this difference achieved significance when subjects having audiometric notches were excluded, or when a subset of exposed subjects was taken in whom subjective evidence was obtained of Temporary Threshold Shift or post-exposure tinnitus. While the most exposed groups did not show significantly greater averaged thresholds, there was in the older age group, a significantly increased prevalence of notches in the audiograms in the 3.5-6 kHz range. We conclude that exposure to amplified music can be harmful, the earliest sign being decrease in frequency resolution, and that early elevation of thresholds is better detected by high-resolution Békésy tracking (extending over the range 2-8 kHz) than by conventional fixed-frequency audiometry.

The effects of intracochlear cyanide and tetrodotoxin on the properties of single cochlear nerve fibres in the cat.

1. Tuning properties and spontaneous discharge rate of single cochlear fibres in the anaesthetized cat were determined under conditions where millimolar concentrations of KCN were instilled into the scala tympani. 2. Short-term effects on the tuning properties were obtained, in which the threshold of the low threshold sharply tuned tip segment of the frequency-threshold ('tuning') curve (f.t.c.) was elevated by up to 40 db, without changes in the threshold of the low frequency 'tail' segment of the f.t.c., or necessarily changes in the spontaneous and maximally evoked activity. These changes were accompanied by a shift of the characteristic frequency tip segment towards lower frequencies. All these effects could be reversed. 3. The long-term effects of repeated KCN instillations produced irreversible changes similar to the short-term effects. 4. These changes correlated well with depression of the amplitude of the gross cochlear action potential but not with the cochlear microphonic potential, both recorded at the round window. 5. Instillations of tetrodotoxin (TTX) rapidly reduced and blocked the cochlear fibre discharges without effects on their tuning, in contrast to the effects of KCN.

The effects of intracochlear and systemic furosemide on the properties of single cochlear nerve fibres in the cat.

1. Tuning properties and spontaneous discharge rate of single cochlear fibres in the anaesthetized cat were determined during short- and long-term poisoning of the cochlea by locally and systemically applied furosemide.2. With intra-arterial administration of furosemide, short-term reversible elevation occurred of the low threshold sharply tuned ;tip' segment of the frequency threshold (;tuning') curve (f.t.c.) by up to 40 db, without substantial changes in the threshold of the low frequency ;tail' segment of the f.t.c. These changes could occur in part without changes in the spontaneous activity and entirely without changes in the maximal evoked activity. These effects were observed in all fibres examined, the characteristic frequencies of which ranged from 3.5 to 31 kHz.3. Intracochlear administration of furosemide in 0.9 mM concentrations produced similar changes, but these were not reversible.4. The changes correlated with the depression of the amplitude of the gross cochlear action potential. The cochlear microphonic potential, however, was either unchanged, or only slightly reduced.5. In long-term furosemide poisoning of the cochlea, fibres with anomalous response properties were found alongside fibres having normal tuning. The former exhibited either reduced excitability of the low threshold tip segment, or a tip segment attenuated in both excitability and threshold.6. It is concluded that the selective effects of furosemide on the tip segment of cochlear fibre f.t.c.s offer further evidence for a physiologically vulnerable ;second filter' in the cochlea. The selective influence of the furosemide on the low threshold tip segment provides support for the hypothesis that the normal f.t.c. is generated by two largely independent processes: one vulnerable, low threshold and sharply tuned, and the other less vulnerable, but high threshold and more broadly tuned.7. The findings, obtained with an agent known to produce reversible impairment of hearing in man, provide direct physiological evidence in support of the hypothesis that in sensorineural hearing loss of cochlear origin the frequency selectivity of cochlear nerve fibres is impaired.

Intensity coding in the auditory periphery of the cat: responses of cochlear nerve and cochlear nucleus neurons to signals in the presence of bandstop masking noise.

The dynamic range over which fine intensity discrimination is possible has been reported to be largely unaffected by limitation of the spread of neuronal activity to neighbouring frequency regions by bandstop noise masking. We have therefore examined the responses of cochlear nerve and nucleus neurons to tone and noise signals in the presence of a bandstop masking noise designed to be comparable to that employed in the psychophysical experiments. Under these conditions, the vast majority of cochlear nerve fibres were saturated by sound levels at which some 50% of our sample of cochlear nucleus neurons still responded to signal level differences. The extended dynamic ranges of these cochlear nucleus neurons was shown to be a result of activation, by the masking noise, of the lateral inhibitory side-bands 'biassing' the neuron's discharge. A small proportion of cochlear fibres, having low spontaneous discharge rates and showing strong two-tone suppression effects, demonstrated analogous but not so pronounced effects. It is unclear in what form information on the level of stimuli under these conditions is transmitted by the majority of apparently saturated cochlear nerve fibres, but several possible mechanisms are discussed.

Ototoxic effects of salicylates on the responses of single cochlear nerve fibres and on cochlear potentials.

Anaesthetized cats were given 400 mg/kg sodium salicylate i.v. producing blood levels in excess of 300-400 mg/kg. Within 10 min of injection, thresholds of fibers had risen by values ranging from 13 to 21 dB. The elevation in thresholds progressed rapidly over the subsequent 5 or so hours, reaching a plateau in about 10 h. The Q10 dB values for tuning of the cochlear fibres decreased by a factor of 3-4 on average over the same period. Likewise, the dynamic range of response was significantly reduced. These effects on the cochlear fibres were reflected in the elevation of the gross cochlear action potential thresholds. In contrast to the findings with other types of cochlear pathology, the mean discharge rate of the subpopulation of cochlear fibres having discharge rates above 20 sp/s was significantly increased by an average of 10-20 sp/s. There was a tendency for this increase to be more marked for fibres with higher characteristic frequencies and to be inversely related to threshold. In addition, 63% of fibres had anomalous temporal patterns of spontaneous activity. In view of the relevance of these data on the ototoxic effects of salicylates for our understanding of tinnitus, the effects of direct electrical current stimulation via the round window have been studied. Positive currents up to at least 600 microA suppressed the spontaneous and evoked activity of all cochlear fibres studied in the normal cochlea. These findings are consistent with the effects of such current stimulation in patients with tinnitus of peripheral origin and support the hypothesis that the neural correlate of such tinnitus is hyperactivity at the cochlear nerve level.

Reverse correlation study of cochlear filtering in normal and pathological guinea pig ears.

The filtering properties of single cochlear fibres have been determined in normal and kanamycin-treated guinea pigs using the reverse correlation technique. This method allows investigation of filtering over a wide dynamic range. For normal guinea pig fibres, the near threshold filter functions obtained with this method correspond to the tone derived frequency threshold curves ((FTCs). The 10 dB bandwidth of the filter functions increased monotonically with increasing noise levels above threshold. Thus with noise levels at approximately 50 dB above threshold, the 10 dB bandwidth had increased by a factor of 1.3--3. The changes in 3 dB bandwidth with increasing levels were, for some fibres, different from those of the 10 dB bandwidths. For the pathological fibres, the derived filter functions corresponded to their tone determined FTCs, and were therefore comparatively broadly tuned. Their tuning (Q10dB) approximated to those of normal fibres when the latter were measured 60 dB or more above threshold (i.e., at similar levels of stimulus), and did not increase further with increase in level. The findings in the normal guinea pig are consistent with those obtained by others in rodents, but are not consistent with those from the cat, where normal filtering is more robust to high levels of stimulus noise.

Location-specific components of the gross cochlear action potential: an assessment of the validity of the high-pass masking technique by cochlear nerve fibre recording in the cat.

The auditory high-pass masking technique has been used in attempts to define the origin, along the cochlear partition, of the gross cochlear action potential (CAP) and the gross brain stem potential. Theoretically, the high-pass masking paradigm should be frequency and location specific at the cochlear level, and some indirect evidence does point to this specificity. However, this hypothesis has not yet been directly substantiated. In the present experiment, click-evoked cochlear nerve activity was recorded simultaneously from the round window and from single fibres of the cochlear nerve, with and without high-pass maskers spaced in octaves from 0.5 to 16 kHz, at three intensities, in the anaesthetized cat. The "derived' CAPs were computed and compared with the mapping of single cochlear fibre responses under the same conditions. With one main exception, the conclusions drawn on the origin of the frequency components of the "derived' potentials were found to be valid in the normal cat. The exception concerned fibres with characteristic frequencies below 1-2 kHz, where the substantial spread towards the high frequencies of their frequency threshold curves, and the effects of lateral suppression or of other "remote masking' phenomena rendered the high-pass masking less location specific. From these results and certain assumptions, we would predict the high-pass masking technique to be valid in electrophysiological investigations in normal humans for frequencies down to 0.5-1 kHz.

Animal models of tinnitus.

There are few physiological data available on the origin and nature of tinnitus. It is not even known whether tinnitus associated with cochlear pathology is a manifestation of increased or decreased activity in the cochlear nerve. In previous investigations of cochlear pathology, the spontaneous neural activity has generally been found to be depressed. In the present experiments, an animal model has been established by the administration of sodium salicylate in doses producing blood concentrations that evoke tinnitus in humans. Under these conditions, changes occur in cochlear nerve-fibre thresholds and tuning, similar to those obtained in other types of cochlear pathology. However, under salicylate, the distribution of spontaneous discharge shifts significantly to higher rates than normal. These changes are accompanied in some, but not all, fibres by changes in the temporal patterns of discharge suggestive of excitation. In the second animal model studied, a normal guinea-pig that had a naturally occurring continuous tonal emission, analogous to that recently recorded in human "physiological" tinnitus, was investigated in detail. The emitted signal was recorded in the ear-canal acoustic pressure and in the round-window potential. Several lines of evidence point to the signal as being cochlear in origin, including: its resistance to muscular paralysis and section of the stapedius muscle; the effects of changes in middle-ear pressure; its reversible elimination by hypoxia; and its suppression by tones of higher frequency.

Cochlear fibre rate--intensity functions: no evidence for basilar membrane nonlinearities.

The presence of inflexions in the slopes of the rate--intensity functions of single cochlear fibres has been cited as evidence for nonlinearities, in the motion of the basilar membrane, or the type described by Rhode. We have carried out a detailed study of these inflexions in the anaesthetized cat. The inflexions were found to be related more to the threshold of the individual fibre than to occur at a common sound pressure level, for fibres of similar and differing characteristic frequency. Likewise, contrary to the earlier evidence, no strong inverse relationship was found to exist between the magnitude of the slope of the rate-intensity function and the threshold of a fibre. What relationship did exist depended upon the presence of fibres having low spontaneous discharge rate activity. The data therefore do not suport the concept of a common input (i.e. basilar membrane) nonlinearity; rather, they suggest that the shape and slope of the rate--intensity functions are more related to the properties of individual neural channels.

Relationship between the dynamic range of cochlear nerve fibres and their spontaneous activity.

The dynamic ranges of cochlear nerve fibres in cats were determined automatically and were related to the fibres' rates of spontaneous activity, in both pooled data and data from individual cochlear nerves. The dynamic range represents the range of levels of a tone at the characteristic frequency of the fibre evoking mean discharge rates between spontaneous and saturated activity. In common with the findings of other investigators, the distribution of spontaneous discharge rates was bimodal. The total population could be divided into two sub-populations with spontaneous discharge rates above and below 15 spikes/s, respectively. The mean dynamic range of fibres having spontaneous discharge rates in excess of 15 spikes/s, was 41 dB (+/- 0.65 S.E.); that for fibres with rates below 15 spikes/s was 50 dB (+/- 1.2 S.E.). While the distributions of dynamic ranges of the two populations overlapped, they were significantly different, and dynamic ranges in excess of 60 dB were only found in substantial numbers (23%) in the population having low spontaneous discharge rates. Some of these were not saturated at the highest stimulus levels used.