Cortical sensory neurons are commonly characterized using the receptive field, the linear dependence of their response on the stimulus. In primary auditory cortex neurons can be characterized by their spectrotemporal receptive fields, the spectral and temporal features of a sound that linearly drive a neuron. However, receptive fields do not capture the fact that the response of a cortical neuron results from the complex nonlinear network in which it is embedded. By fitting a nonlinear feedforward network model (a network receptive field) to cortical responses to natural sounds, we reveal that primary auditory cortical neurons are sensitive over a substantially larger spectrotemporal domain than is seen in their standard spectrotemporal receptive fields. Furthermore, the network receptive field, a parsimonious network consisting of 1–7 sub-receptive fields that interact nonlinearly, consistently better predicts neural responses to auditory stimuli than the standard receptive fields. The network receptive field reveals separate excitatory and inhibitory sub-fields with different nonlinear properties, and interaction of the sub-fields gives rise to important operations such as gain control and conjunctive feature detection. The conjunctive effects, where neurons respond only if several specific features are present together, enable increased selectivity for particular complex spectrotemporal structures, and may constitute an important stage in sound recognition. In conclusion, we demonstrate that fitting auditory cortical neural responses with feedforward network models expands on simple linear receptive field models in a manner that yields substantially improved predictive power and reveals key nonlinear aspects of cortical processing, while remaining easy to interpret in a physiological context.

The discrimination of interaural phase differences (IPDs) requires accurate binaural temporal processing and has been used as a measure of sensitivity to temporal envelope and temporal fine structure (TFS). Previous studies found that TFS-IPD discrimination declined with age and with sensorineural hearing loss (SNHL), but age and SNHL have often been confounded. The aim of this study was to determine the independent contributions of age and SNHL to TFS and envelope IPD discrimination by using a sample of adults with a wide range of ages and SNHL. A two-interval, two-alternative forced-choice procedure was used to measure IPD discrimination thresholds for 20-Hz amplitude-modulated tones with carrier frequencies of 250 or 500 Hz when the IPD was in either the stimulus envelope or TFS. There were positive correlations between absolute thresholds and TFS-IPD thresholds, but not envelope-IPD thresholds, when age was accounted for. This supports the idea that SNHL affects TFS processing independently to age. Age was positively correlated with envelope-IPD thresholds at both carrier frequencies and TFS-IPD thresholds at 500 Hz, when absolute thresholds were accounted for. These results suggest that age negatively affects the binaural processing of envelope and TFS at some frequencies independently of SNHL.

Coding for auditory space in the nucleus of the brachium of the inferior colliculus in the ferret. J. Neurophysiol. 78: 2717-2731, 1997. The nucleus of the brachium of the inferior colliculus (BIN) projects topographically to the deeper layers of the superior colliculus (SC), which contain a two-dimensional map of auditory space. In this study, we have used broadband stimuli presented in the free field to investigate how auditory space is represented in the BIN of the ferret. Response latencies and temporal firing patterns were comparable with those in the SC, and both properties showed some variation with stimulus location. We obtained spatial response profiles at two sound levels (5-15 and 25-35 dB above unit threshold). A large proportion of azimuth profiles (41% in the suprathreshold condition, 80% in the near-threshold condition) presented a single peak, indicating that they were tuned to single regions in space. For some of these units, the preferred speaker position varied considerably with sound level. The remaining units showed predominantly either broad "hemifield" or spatially ambiguous "bilobed" response profiles. At suprathreshold sound levels, the preferred azimuths of the tuned cells were ordered topographically along the rostrocaudal axis of the BIN, although this representation is considerably more scattered than that in the SC. In contrast to the SC, we observed no systematic variation in the distribution of near-threshold best azimuths, which were instead concentrated around the interaural axis in the contralateral hemifield. The azimuth tuning of individual units in the BIN was generally broader at both sound levels than that in the SC. Many units also were tuned for the elevation of the sound source (48% for supra-, 77% for near-threshold stimulation), but there was no evidence for topographic order in the distribution of preferred elevations within the BIN. These results suggest that the BIN sends inputs to the SC that are already selective for sound azimuth and elevation and that show some degree of topographic order for sound azimuth. These inputs then presumably are sharpened and their topography refined by a mechanism that is likely to involve convergence of other inputs and activity-dependent fine tuning of terminal connections, to result in a precise two-dimensional map of auditory space in the SC.

The capacity of the auditory system to extract spatial information relies principally on the detection and interpretation of binaural cues, i.e., differences in the time of arrival or level of the sound between the two ears. In this review, we consider the effects of unilateral or asymmetric hearing loss on spatial hearing, with a focus on the adaptive changes in the brain that may help to compensate for an imbalance in input between the ears. Unilateral hearing loss during development weakens the brain's representation of the deprived ear, and this may outlast the restoration of function in that ear and therefore impair performance on tasks such as sound localization and spatial release from masking that rely on binaural processing. However, loss of hearing in one ear also triggers a reweighting of the cues used for sound localization, resulting in increased dependence on the spectral cues provided by the other ear for localization in azimuth, as well as adjustments in binaural sensitivity that help to offset the imbalance in inputs between the two ears. These adaptive strategies enable the developing auditory system to compensate to a large degree for asymmetric hearing loss, thereby maintaining accurate sound localization. They can also be leveraged by training following hearing loss in adulthood. Although further research is needed to determine whether this plasticity can generalize to more realistic listening conditions and to other tasks, such as spatial unmasking, the capacity of the auditory system to undergo these adaptive changes has important implications for rehabilitation strategies in the hearing impaired.

To examine the influence of acoustic experience on the development of the mammalian auditory brain stem, darkly pigmented ferrets were reared with a plug inserted in the right outer ear. The plugs were first inserted on postnatal day 23-34 and produced a variable, frequency-dependent attenuation of up to 60 dB. Between 3-15 months after the ear plug was begun, animals were prepared for physiological recording and injection of wheat germ agglutinin-HRP (WGA-HRP) in the left inferior colliculus (IC). The plug was removed and the condition of the right ear was assessed by pure-tone stimulation and recordings from neurons in the left IC. Neural audiograms for each animal showed a residual deficit in most cases. Following 24-60 hr survival, the animals were perfused and the right ear was examined. Brain-stem sections were reacted with tetramethylbenzidine. Outer and/or middle ear pathology was present in over half of the animals. However, the cochleas appeared to be normal and the spiral ganglion cells were normal by several quantitative criteria: number, area, and nucleolar eccentricity. The volume of each division of the cochlear nuclei (CN) and the areas of individual neurons in the anteroventral CN were the same on the right and left sides. The number of CN neurons retrogradely labeled from the left IC injection of WGA-HRP was found to be significantly increased in the left CN, relative to normal animals, when expressed as a ratio of the number labeled in the right CN. We conclude that the residual hearing loss in the previously plugged ears was predominantly or exclusively conductive. Neonatal, unilateral conductive hearing loss in the ferret does not lead to degeneration of the CN on the side of the loss, but it does lead to at least one rearrangement of auditory brain-stem connectivity. We suggest that the extent to which the brain stem is modified by early auditory deprivation is dependent on the type, degree, and symmetry of the hearing loss.