Auditory midbrain responses parallel spectral integration phenomena

Hierarchical differences in the encoding of sound and choice in the subcortical auditory system

Detection of sounds is a fundamental function of the auditory system. Although studies of auditory cortex have gained substantial insight into detection performance using behaving animals, previous subcortical studies have mostly taken place under anesthesia, in passively listening animals, or have not measured performance at threshold. These limitations preclude direct comparisons between neuronal responses and behavior. To address this, we simultaneously measured auditory detection performance and single-unit activity in the inferior colliculus (IC) and cochlear nucleus (CN) in macaques. The spontaneous activity and response variability of CN neurons were higher than those observed for IC neurons. Signal detection theoretic methods revealed that the magnitude of responses of IC neurons provided more reliable estimates of psychometric threshold and slope compared with the responses of single CN neurons. However, pooling small populations of CN neurons provided reliable estimates of psychometric threshold and slope, suggesting sufficient information in CN population activity. Trial-by-trial correlations between spike count and behavioral response emerged 50-75 ms after sound onset for most IC neurons, but for few neurons in the CN. These results highlight hierarchical differences between neurometric-psychometric correlations in CN and IC and have important implications for how subcortical information could be decoded.NEW & NOTEWORTHY The cerebral cortex is widely recognized to play a role in sensory processing and decision-making. Accounts of the neural basis of auditory perception and its dysfunction are based on this idea. However, significantly less attention has been paid to midbrain and brainstem structures in this regard. Here, we find that subcortical auditory neurons represent stimulus information sufficient for detection and predict behavioral choice on a trial-by-trial basis.

Wide sensory filters underlie performance in memory-based discrimination and generalization

The way animals respond to a stimulus depends largely on an internal comparison between the current sensation and the memory of previous stimuli and outcomes. We know little about the accuracy with which the physical properties of the stimuli influence this type of memory-based discriminative decisions. Research has focused largely on discriminations between stimuli presented in quick succession, where animals can make relative inferences (same or different; higher or lower) from trial to trial. In the current study we used a memory-based task to explore how the stimulus’ physical properties, in this case tone frequency, affect auditory discrimination and generalization in mice. Mice performed ad libitum while living in groups in their home quarters. We found that the frequency distance between safe and conditioned sounds had a constraining effect on discrimination. As the safe-to-conditioned distance decreased across groups, performance deteriorated rapidly, even for frequency differences significantly larger than reported discrimination thresholds. Generalization width was influenced both by the physical distance and the previous experience of the mice, and was not accompanied by a decrease in sensory acuity. In conclusion, memory-based discriminations along a single stimulus dimension are inherently hard, reflecting a high overlap between the memory traces of the relevant stimuli. Memory-based discriminations rely therefore on wide sensory filters.

Adaptation and spectral enhancement at auditory temporal perceptual boundaries - Measurements via temporal precision of auditory brainstem responses

In human and animal auditory perception the perceived quality of sound streams changes depending on the duration of inter-sound intervals (ISIs). Here, we studied whether adaptation and the precision of temporal coding in the auditory periphery reproduce general perceptual boundaries in the time domain near 20, 100, and 400 ms ISIs, the physiological origin of which are unknown. In four experiments, we recorded auditory brainstem responses with five wave peaks (P1 –P5) in response to acoustic models of communication calls of house mice, who perceived these calls with the mentioned boundaries. The newly introduced measure of average standard deviations of wave latencies of individual animals indicate the waves’ temporal precision (latency jitter) mostly in the range of 30–100 μs, very similar to latency jitter of single neurons. Adaptation effects of response latencies and latency jitter were measured for ISIs of 10–1000 ms. Adaptation decreased with increasing ISI duration following exponential or linear (on a logarithmic scale) functions in the range of up to about 200 ms ISIs. Adaptation effects were specific for each processing level in the auditory system. The perceptual boundaries near 20–30 and 100 ms ISIs were reflected in significant adaptation of latencies together with increases of latency jitter at P2-P5 for ISIs < ~30 ms and at P5 for ISIs < ~100 ms, respectively. Adaptation effects occurred when frequencies in a sound stream were within the same critical band. Ongoing low-frequency components/formants in a sound enhanced (decrease of latencies) coding of high-frequency components/formants when the frequencies concerned different critical bands. The results are discussed in the context of coding multi-harmonic sounds and stop-consonants-vowel pairs in the auditory brainstem. Furthermore, latency data at P1 (cochlea level) offer a reasonable value for the base-to-apex cochlear travel time in the mouse (0.342 ms) that has not been determined experimentally.

Spectral summation and facilitation in on- and off-responses for optimized representation of communication calls in mouse inferior colliculus

Selectivity for processing of species-specific vocalizations and communication sounds has often been associated with the auditory cortex. The midbrain inferior colliculus, however, is the first center in the auditory pathways of mammals integrating acoustic information processed in separate nuclei and channels in the brainstem and, therefore, could significantly contribute to enhance the perception of species' communication sounds. Here, we used natural wriggling calls of mouse pups, which communicate need for maternal care to adult females, and further 15 synthesized sounds to test the hypothesis that neurons in the central nucleus of the inferior colliculus of adult females optimize their response rates for reproduction of the three main harmonics (formants) of wriggling calls. The results confirmed the hypothesis showing that average response rates, as recorded extracellularly from single units, were highest and spectral facilitation most effective for both onset and offset responses to the call and call models with three resolved frequencies according to critical bands in perception. In addition, the general on- and/or off-response enhancement in almost half the investigated 122 neurons favors not only perception of single calls but also of vocalization rhythm. In summary, our study provides strong evidence that critical-band resolved frequency components within a communication sound increase the probability of its perception by boosting the signal-to-noise ratio of neural response rates within the inferior colliculus for at least 20% (our criterion for facilitation). These mechanisms, including enhancement of rhythm coding, are generally favorable to processing of other animal and human vocalizations, including formants of speech sounds.

Frequency-specific adaptation and its underlying circuit model in the auditory midbrain

Receptive fields of sensory neurons are considered to be dynamic and depend on the stimulus history. In the auditory system, evidence of dynamic frequency-receptive fields has been found following stimulus-specific adaptation (SSA). However, the underlying mechanism and circuitry of SSA have not been fully elucidated. Here, we studied how frequency-receptive fields of neurons in rat inferior colliculus (IC) changed when exposed to a biased tone sequence. Pure tone with one specific frequency (adaptor) was presented markedly more often than others. The adapted tuning was compared with the original tuning measured with an unbiased sequence. We found inhomogeneous changes in frequency tuning in IC, exhibiting a center-surround pattern with respect to the neuron's best frequency. Central adaptors elicited strong suppressive and repulsive changes while flank adaptors induced facilitative and attractive changes. Moreover, we proposed a two-layer model of the underlying network, which not only reproduced the adaptive changes in the receptive fields but also predicted novelty responses to oddball sequences. These results suggest that frequency-specific adaptation in auditory midbrain can be accounted for by an adapted frequency channel and its lateral spreading of adaptation, which shed light on the organization of the underlying circuitry.

Rodent auditory perception: Critical band limitations and plasticity

What do animals hear? While it remains challenging to adequately assess sensory perception in animal models, it is important to determine perceptual abilities in model systems to understand how physiological processes and plasticity relate to perception, learning, and cognition. Here we discuss hearing in rodents, reviewing previous and recent behavioral experiments querying acoustic perception in rats and mice, and examining the relation between behavioral data and electrophysiological recordings from the central auditory system. We focus on measurements of critical bands, which are psychoacoustic phenomena that seem to have a neural basis in the functional organization of the cochlea and the inferior colliculus. We then discuss how behavioral training, brain stimulation, and neuropathology impact auditory processing and perception.

Representation of loudness in the auditory cortex

Changes in stimulus intensity are reflected in changes in the fundamental perceptual attribute of loudness. Stimulus intensity changes also profoundly impact the evoked neural responses throughout the auditory system. A fundamental question is how measurements of neural activity, from the single-neuron level to mass-activity metrics such as functional magnetic resonance imaging or magnetoencephalography, reflect the physical properties of stimulus intensity as opposed to perceived loudness. In this chapter we discuss findings from psychophysics and animal neurophysiology as well as human brain activity measurements to clarify our current understanding of the neural mechanisms that contribute to the perceptual correlate of stimulus intensity.

Stimulus-specific adaptation and deviance detection in the auditory system: experiments and models

Stimulus-specific adaptation (SSA) is the reduction in the response to a common stimulus that does not generalize, or only partially generalizes, to other, rare stimuli. SSA has been proposed to be a correlate of 'deviance detection', an important computational task of sensory systems. SSA is ubiquitous in the auditory system: It is found both in cortex and in subcortical stations, and it has been demonstrated in many mammalian species as well as in birds. A number of models have been suggested in the literature to account for SSA in the auditory domain. In this review, the experimental literature is critically examined in relationship to these models. While current models can all account for auditory SSA to some degree, none is fully compatible with the available findings.

Axon guidance in the auditory system: multiple functions of Eph receptors

The neural pathways of the auditory system underlie our ability to detect sounds and to transform amplitude and frequency information into rich and meaningful perception. While it shares some organizational features with other sensory systems, the auditory system has some unique functions that impose special demands on precision in circuit assembly. In particular, the cochlear epithelium creates a frequency map rather than a space map, and specialized pathways extract information on interaural time and intensity differences to permit sound source localization. The assembly of auditory circuitry requires the coordinated function of multiple molecular cues. Eph receptors and their ephrin ligands constitute a large family of axon guidance molecules with developmentally regulated expression throughout the auditory system. Functional studies of Eph/ephrin signaling have revealed important roles at multiple levels of the auditory pathway, from the cochlea to the auditory cortex. These proteins provide graded cues used in establishing tonotopically ordered connections between auditory areas, as well as discrete cues that enable axons to form connections with appropriate postsynaptic partners within a target area. Throughout the auditory system, Eph proteins help to establish patterning in neural pathways during early development. This early targeting, which is further refined with neuronal activity, establishes the precision needed for auditory perception.

Frequency tuning in the behaving mouse: different bandwidths for discrimination and generalization

When faced with sensory stimuli, an organism may be required to detect very small differences in a physical parameter (discrimination), while in other situations it may have to generalize over many possible values of the same physical parameter. This decision may be based both on learned information and on sensory aspects of perception. In the present study we describe frequency processing in the behaving mouse using both discrimination and generalization as two key aspects of behaviour. We used a novel naturalistic behavioural apparatus designed for mice, the Audiobox, and paradigm contingencies that were identical for both auditory discrimination and generalization, the latter measured using latent inhibition. Mice learned to discriminate between frequencies that were an octave apart in a single trial. They showed significant discrimination between tone frequencies that were as close as 4–7%, and had d' of about 1 for ΔF of around 10%. In contrast, pre-exposure frequencies that were half an octave or less below the conditioned tone elicited latent inhibition, showing a generalization bandwidth of at least half an octave. Thus, in the same apparatus and using the same general memory paradigm, mice showed generalization gradients that were considerably wider than their discrimination threshold, indicating that environmental requirements and previous experience can determine whether the same two frequencies will be considered same or different. Remarkably, generalization gradients paralleled the typical bandwidths established in the auditory periphery and midbrain, suggesting that frequencies may be considered similar when falling within the same critical band.

Spectral coding in auditory midbrain neurons

The data of frequency properties of single neurons in the auditory midbrain center, central nucleus of the inferior colliculus and morphologic aspects of its functional ordering are reviewed. On the basis of reconstruction of single units frequency receptive fields and morphophysiologic mapping of their location within the frequency-band lamina of the central nucleus the model of spectral coding by auditory midbrain neurons is developed. The main structural basement of spectral coding in the auditory midbrain is the tonotopic organization of its central nucleus as well as the order in its morphological structure expressed in alternation of layers of disc-shaped neurons and neuropil. The main neural mechanism of spectral coding is a critical bands mechanism. The critical bandwidth is controlled by inhibitory disc-shaped neurons which constitute an essential part of the population of the central nucleus disc-shaped cells.

Temporal coding of intensity of NaCl and HCl in the nucleus of the solitary tract of the rat

Sensory neurons are generally tuned to a subset of stimulus qualities within their sensory domain and manifest this tuning by the relative size of their responses to stimuli of equal intensity. However, response size alone cannot unambiguously signal stimulus quality, since response size also depends on stimulus intensity. Thus a common problem faced by sensory systems is that response size (e.g., spike count) confounds stimulus quality and intensity. Here, using the gustatory system as a model, we asked whether temporal firing characteristics could disambiguate these axes. To address this question, we recorded taste responses of single neurons in the nucleus of the solitary tract (NTS, the first central gustatory relay) in anesthetized rats to a range of concentrations of NaCl and HCl and their binary mixtures. To assess the contribution of the temporal characteristics of the response to discrimination among tastants, a family of metrics that quantifies the similarity of two spike trains in terms of spike count and spike timing was used. Results showed that the spike count produced by different taste qualities and different concentrations overlapped in most cells, implying that information conveyed by spike count is imprecise. Multidimensional scaling analysis of taste responses using similarity of temporal characteristics showed that different taste qualities, intensities, and mixtures formed distinct clusters in this "temporal coding" taste space and were arranged in a logical order. Thus the temporal structure of taste responses in single cells in the NTS can simultaneously convey information about both taste quality and intensity.

A discontinuous tonotopic organization in the inferior colliculus of the rat

Audible frequencies of sound are encoded in a continuous manner along the length of the cochlea, and frequency is transmitted to the brain as a representation of place on the basilar membrane. The resulting tonotopic map has been assumed to be a continuous smooth progression from low to high frequency throughout the central auditory system. Here, physiological and anatomical data show that best frequency is represented in a discontinuous manner in the inferior colliculus, the major auditory structure of the midbrain. Multiunit maps demonstrate a distinct stepwise organization in the order of best frequency progression. Furthermore, independent data from single neurons show that best frequencies at octave intervals of approximately one-third are more prevalent than others. These data suggest that, in the inferior colliculus, there is a defined space of tissue devoted to a given frequency, and input within this frequency band may be pooled for higher-level processing.

Some investigations into non-passive listening

Our knowledge of the function of the auditory nervous system is based upon a wealth of data obtained, for the most part, in anaesthetised animals. More recently, it has been generally acknowledged that factors such as attention profoundly modulate the activity of sensory systems and this can take place at many levels of processing. Imaging studies, in particular, have revealed the greater activation of auditory areas and areas outside of sensory processing areas when attending to a stimulus. We present here a brief review of the consequences of such non-passive listening and go on to describe some of the experiments we are conducting to investigate them. In imaging studies, using fMRI, we can demonstrate the activation of attention networks that are non-specific to the sensory modality as well as greater and different activation of the areas of the supra-temporal plane that includes primary and secondary auditory areas. The profuse descending connections of the auditory system seem likely to be part of the mechanisms subserving attention to sound. These are generally thought to be largely inactivated by anaesthesia. However, we have been able to demonstrate that even in an anaesthetised preparation, removing the descending control from the cortex leads to quite profound changes in the temporal patterns of activation by sounds in thalamus and inferior colliculus. Some of these effects seem to be specific to the ear of stimulation and affect interaural processing. To bridge these observations we are developing an awake behaving preparation involving freely moving animals in which it will be possible to investigate the effects of consciousness (by contrasting awake and anaesthetized), passive and active listening.

Frequency response areas of mouse inferior colliculus neurons: II. Critical bands

Critical bands are perceptual filters that detect and separate spectral peaks in complex sounds. Here, we show that the main properties of psychophysically defined critical bands, as measured in narrow-band noise masking tests (species-specific frequency dependence and intensity independence of the bandwidths), are present in single neurons of the mouse's central nucleus of the inferior colliculus. Bandwidths of critical bands amount to, on average, 3/8-1/3 octave related to the neurons' characteristic frequencies. They are not determined by the shapes of the neurons' excitatory receptive fields. The results support the view that frequency resolution in the auditory system is shaped to its perceptual level in the main nucleus of the auditory midbrain.

Spectral resolution of monkey primary auditory cortex (A1) revealed with two-noise masking

An important function of the auditory nervous system is to analyze the frequency content of environmental sounds. The neural structures involved in determining psychophysical frequency resolution remain unclear. Using a two-noise masking paradigm, the present study investigates the spectral resolution of neural populations in primary auditory cortex (A1) of awake macaques and the degree to which it matches psychophysical frequency resolution. Neural ensemble responses (auditory evoked potentials, multiunit activity, and current source density) evoked by a pulsed 60-dB SPL pure-tone signal fixed at the best frequency (BF) of the recorded neural populations were examined as a function of the frequency separation (DeltaF) between the tone and two symmetrically flanking continuous 80-dB SPL, 50-Hz-wide bands of noise. DeltaFs ranged from 0 to 50% of the BF, encompassing the range typically examined in psychoacoustic experiments. Responses to the signal were minimal for DeltaF = 0% and progressively increased with DeltaF, reaching a maximum at DeltaF = 50%. Rounded exponential functions, used to model auditory filter shapes in psychoacoustic studies of frequency resolution, provided excellent fits to neural masking functions. Goodness-of-fit was greatest for response components in lamina 4 and lower lamina 3 and least for components recorded in more superficial cortical laminae. Physiological equivalent rectangular bandwidths (ERBs) increased with BF, measuring nearly 15% of the BF. These findings parallel results of psychoacoustic studies in both monkeys and humans, and thus indicate that a representation of perceptual frequency resolution is available at the level of A1.

Adaptive stimulus optimization for auditory cortical neurons

Despite the extensive physiological work performed on auditory cortex, our understanding of the basic functional properties of auditory cortical neurons is incomplete. For example, it remains unclear what stimulus features are most important for these cells. Determining these features is challenging given the considerable size of the relevant stimulus parameter space as well as the unpredictable nature of many neurons' responses to complex stimuli due to nonlinear integration across frequency. Here we used an adaptive stimulus optimization technique to obtain the preferred spectral input for neurons in macaque primary auditory cortex (AI). This method uses a neuron's response to progressively modify the frequency composition of a stimulus to determine the preferred spectrum. This technique has the advantage of being able to incorporate nonlinear stimulus interactions into a "best estimate" of a neuron's preferred spectrum. The resulting spectra displayed a consistent, relatively simple circumscribed form that was similar across scale and frequency in which excitation and inhibition appeared about equally prominent. In most cases, this structure could be described using two simple models, the Gabor and difference of Gaussians functions. The findings indicate that AI neurons are well suited for extracting important scale-invariant features in sound spectra and suggest that they are designed to efficiently represent natural sounds.

Duration selectivity of neurons in the inferior colliculus of the big brown bat: tolerance to changes in sound level

At and above the level of the inferior colliculus (IC), some neurons respond maximally to a limited range of sound durations, with little or no excitatory response to durations outside of this range. Such neurons have been termed "duration tuned" or "duration selective." In this study we examined the effects of varying signal amplitude on best duration, width of tuning, and first spike latency of duration tuned neurons in the IC of the big brown bat, Eptesicus fuscus. Response areas as a function of stimulus duration and intensity took a variety of forms, including open (V-shaped), narrow and level tolerant (U-shaped), or closed (O-shaped). The majority (82%) of duration tuned neurons had narrow U-shaped or O-shaped duration response areas. Those with narrow U-shaped response areas retained their duration tuning across a broad dynamic range, < or = 50 dB above threshold, whereas those with O-shaped response areas were narrowly tuned to both stimulus duration and amplitude. For about one-half (55%) of the neurons with either a U- or O-shaped response areas, best duration (BD) changed by <1 ms across the range of suprathreshold amplitudes tested. Changes in BD most often took the form of a shift to slightly shorter durations as stimulus level increased. For the majority (65%) of U- and O-shaped neurons, 50% width of duration tuning changed by <2 ms with increasing amplitude. Latency of response at BD remained stable across changes in sound level, suggesting that the relative strengths of excitatory and inhibitory inputs to duration tuned neurons remain in balance over a wide dynamic range of sound pressure levels.

Infant Rodent Ultrasounds ? A Gate to the Understanding of Sound Communication

Components of the communication system between infant and adult rodents based on ultrasonic vocalizations (USVs) of infants are analyzed. USVs are most often emitted from a pup lost outside the nest in response to changes of: (i) body temperature, (ii) contact with adults/littermates, (iii) handling, and (iv) smell. These changes modulate the state of arousal and the emotional/motivational states and, as a result, USVs are produced. Acoustic properties of USVs seem to reflect the degrees of changes in arousal and emotion/motivation. Adult rodents are aroused by perceiving the USVs, locate the sender and show a phonotaxic approach to the sender. Acoustic properties of USVs in the frequency and time domains are described based on which adult rodents discriminate the USVs from other ultrasounds and take the USVs or adequate models of them as preferred goals of their phonotaxic approach. The preferred approach to adequate USVs is modulated by emotions/motivations, the sex of the receiver, hormonal states, experience with pups and neurotransmitter systems of the brain. The phonotaxis can be understood as the appetitive component of a pup-caring instinct. The consummatory act of the instinct is the retrieval of the lost pup. This retrieval is independent of USV presence, but it closes the communication loop activated by the emission of USVs. Communication with USVs can be used as a tool to investigate genetic and brain mechanisms of behavioral control.

Contrast tuning in auditory cortex

The acoustic features useful for converting auditory information into perceived objects are poorly understood. Although auditory cortex neurons have been described as being narrowly tuned and preferentially responsive to narrowband signals, naturally occurring sounds are generally wideband with unique spectral energy profiles. Through the use of parametric wideband acoustic stimuli, we found that such neurons in awake marmoset monkeys respond vigorously to wideband sounds having complex spectral shapes, preferring stimuli of either high or low spectral contrast. Low contrast-preferring neurons cannot be studied thoroughly with narrowband stimuli and have not been previously described. These findings indicate that spectral contrast reflects an important stimulus decomposition in auditory cortex and may contribute to the recognition of acoustic objects.

Gamma-aminobutyric acid circuits shape response properties of auditory cortex neurons

Neurons containing gamma aminobutyric acid (GABA) are widely distributed throughout the primary auditory cortex (AI). We investigated the effects of endogenous GABA by comparing response properties of 110 neurons in chinchilla AI before and after iontophoresis of bicuculline, a GABA(A) receptor antagonist, and/or CGP35348, a GABA(B) receptor antagonist. GABA(A) receptor blockade significantly increased spontaneous and driven discharge rates, dramatically decreased the thresholds of many neurons, and constricted the range of thresholds across the neural population. Some neurons with 'non-onset' temporal discharge patterns developed an onset pattern that was followed by a long pause. Interestingly, the excitatory response area typically expanded on both sides of the characteristic frequency; this expansion exceeded one octave in a third of the sample. Although GABA(B) receptor blockade had little effect alone, the combination of CGP35348 and bicuculline produced greater increases in driven rate and expansion of the frequency response area than GABA(A) receptor blockade alone, suggesting a modulatory role of local GABA(B) receptors. The results suggest that local GABA inhibition contributes significantly to intensity and frequency coding by controlling the range of intensities over which cortical neurons operate and the range of frequencies to which they respond. The inhibitory circuits that generate nonmonotonic rate-level functions are separate from those that influence other response properties of AI neurons.

Corticofugal modulation of midbrain sound processing in the house mouse

An understanding of the neural mechanisms responsible for auditory information processing is incomplete without a careful examination of substantial descending pathways. This study focuses on the functional role of corticofugal projections. Our work with the house mouse reveals that the focal electrical stimulation of the primary auditory cortex leads to profound changes in auditory response properties in the central nucleus of the inferior colliculus of the midbrain. Cortical stimulation does not impact on the collicular best frequencies when the best frequencies of stimulated cortical neurons and recorded collicular neurons are similar. Rather, collicular best frequencies are shifted toward the stimulated cortical best frequencies when cortical and collicular frequencies are different. Such a shift is unrelated to the differences in minimum thresholds between cortical and collicular neurons. In addition to frequency-specific shifts in collicular best frequencies, cortical stimulation elevates collicular minimum thresholds and reduces both dynamic ranges and response magnitudes if cortical and collicular best frequencies are different. If cortical and collicular best frequencies are similar but minimum thresholds are different, collicular minimum thresholds are shifted toward the stimulated cortical thresholds; dynamic ranges and response magnitudes may either increase or decrease in this scenario. Our results suggest that the corticofugal adjustment has a centre-surround organization with regard to both cortical best frequencies and cortical minimum thresholds. The midbrain processing of sound components in the centre of cortical feedback is largely enhanced while processing in the surround is suppressed.

Corticofugal reorganization of the midbrain tonotopic map in mice

Previous studies have indicated that frequency maps (tonotopies) in mammalian auditory brain centers are plastic. Here, we examined this plasticity in the mouse auditory midbrain through focal stimulation of the primary auditory cortex. Cortical activation shifted midbrain frequency tunings toward the best frequencies of the stimulated cortical neurons if these were either higher or lower than the cortical ones. Such corticofugal adjustments appear to minimize the difference between cortical and collicular frequency tuning within the critical bandwidths of the auditory system. Consequently, the neural representation is enhanced for the frequencies to which the cortical neurons were tuned. Our data suggest that the auditory cortex reorganizes midbrain tonotopy on the basis of which cortical frequencies are stimulated, mostly probably through corticofugal projections.

Signal detection in amplitude-modulated maskers. II. Processing in the songbird's auditory forebrain

In the natural environment, acoustic signals have to be detected in ubiquitous background noise. Temporal fluctuations of background noise can be exploited by the auditory system to enhance signal detection, especially if spectral masking components are coherently amplitude modulated across several auditory channels (a phenomenon called 'comodulation masking release'). In this study of neuronal mechanisms of masking release in the primary auditory forebrain (field L) of awake European starlings (Sturnus vulgaris), we determined and compared neural detection thresholds for 20-ms probe tones presented in a background of sinusoidally amplitude modulated (10-Hz) noise maskers. Responses of a total of 34 multiunit clusters were recorded via radiotelemetry with chronically implanted microelectrodes from unrestrained birds. For maskers consisting of a single noise band centred around the recording site's characteristic frequency, a substantial reduction in detection threshold (21 dB on average) was found when probe tones were presented during envelope dips rather than during envelope peaks. Such effects could also explain results obtained for masking protocols where the on-frequency noise band was presented together with excitatory or inhibitory flanking bands that were either coherently modulated (in-phase) or incoherently modulated (phase-shifted). Generally, masking release for probe tones in maskers with flanking bands extending beyond the frequency range of a cell cluster's excitatory tuning curve was not substantially improved. Only some of the neurophysiological results are in agreement with behavioural data from the same species if only the average population response is considered. A subsample of individual neurons, however, could account for behavioural thresholds.

Modular organization of frequency integration in primary auditory cortex

Two fundamental aspects of frequency analysis shape the functional organization of primary auditory cortex. For one, the decomposition of complex sounds into different frequency components is reflected in the tonotopic organization of auditory cortical fields. Second, recent findings suggest that this decomposition is carried out in parallel for a wide range of frequency resolutions by neurons with frequency receptive fields of different sizes (bandwidths). A systematic representation of the range of frequency resolution and, equivalently, spectral integration shapes the functional organization of the iso-frequency domain. Distinct subregions, or "modules," along the iso-frequency domain can be demonstrated with various measures of spectral integration, including pure-tone tuning curves, noise masking, and electrical cochlear stimulation. This modularity in the representation of spectral integration is expressed by intrinsic cortical connections. This organization has implications for our understanding of psychophysical spectral integration measures such as the critical band and general cortical coding strategies.

Topographic and temporal indices of vowel spectral envelope extraction in the human auditory cortex

The auditory-evoked neuromagnetic field elicited by single vowel formants and two-formant vowels was recorded under active listening conditions using a 37-channel magnetometer. There were three single formants with formant frequencies of 200, 400, and 800 Hz, another single formant with a formant frequency of 2600 Hz, and three vowels that were constructed by linear superimposition of the high- onto one of the low-frequency formants. P50 m and N100 m latency values were inversely correlated with the formant frequency of single formants. A strong effect of formant frequency on source location was obtained along the postero-anterior axis, which is orthogonal to the well-established latero-medial tonotopic gradient. Regardless of whether single formants or first formants of vowels were considered, N100 m sources were more anterior and sustained field sources were more posterior for higher-frequency than for lower-frequency formants. The velocity of the apparent posterior-to-anterior movement across cortical surface of N100 m sources first reported by Rogers et al. [Rogers, R. L., Papanicolaou, A. C., Baumann, S. B., Saydjari, C., & Eisenberg, H. M. (1990). Neuromagnetic evidence of a dynamic excitation pattern generating the N100 auditory response. Electroencephalography and Clinical Neurophysiology,77, 237-240] decreased as a function of latency. The amount of deceleration was positively correlated with formant frequency. Responses to the vowels were superadditive, indicating that the processes elicited by the constituents of composite stimuli interact at one or more stages of the afferent auditory pathway. Such interaction may account for the absence of a lateral-to-medial tonotopic mapping of first formant frequency. The source topography found may reflect activity in auditory fields adjacent to AI with the strength of the contribution varying with formant frequency. Alternatively, it may reflect sharpness-of-tuning and inhibitory response-area asymmetry gradients along isofrequency stripes within AI. Either alternative may be interpreted in terms of a spectral blurring mechanism that abstracts spectral envelope information from the details of spectral composition, an important step towards the formation of invariant phonetic percepts.

Complex tone processing in primary auditory cortex of the awake monkey. II. Pitch versus critical band representation

Noninvasive neurophysiological studies in humans support the existence of an orthogonal spatial representation of pure tone frequency and complex tone pitch in auditory cortex [Langner et al., J. Comp. Physiol. A 181, 665-676 (1997)]. However, since this topographic organization is based on neuromagnetic responses evoked by wideband harmonic complexes (HCs) of variable fundamental frequency (f0), and thus interharmonic frequency separation (deltaF), critical band filtering effects due to differential resolvability of harmonics may have contributed to shaping these responses. To test this hypothesis, the present study examined responses evoked by three-component HCs of variable f0 in primary auditory cortex (A1) of the awake monkey. The center frequency of the HCs was fixed at the best frequency (BF) of the cortical site. Auditory evoked potential (AEP), multiunit activity, and current source density techniques were used to evaluate A1 responses as a function of f0 (=deltaF). Generally, amplitudes of nearly all response components increased with f0, such that maximal responses were evoked by HCs comprised of low-order resolved harmonics. Statistically significant increases in response amplitude typically occurred at deltaFs between 10% and 20% of center frequency, suggestive of critical bandlike behavior. Complex tone response amplitudes also reflected nonlinear summation in that they could not be predicted by the pure tone frequency sensitivity curves of the cortical sites. A mechanism accounting for the observed results is proposed which involves mutual lateral inhibitory interactions between responses evoked by stimulus components lying within the same critical band. As intracortical AEP components likely to be propagated to the scalp were also strongly modulated by deltaF, these findings indicate that noninvasive recordings of responses to complex sounds may require a consideration of critical band effects in their interpretation.

Inferior colliculus as candidate for pitch extraction: multiple support from statistics of bilateral spontaneous otoacoustic emissions

The fibrodendritic laminae of the central nucleus of the inferior colliculus (ICC) constitute a frequency map in stacked sheets that are consistently related to the psychoacoustic critical bandwidth (CB) [Schreiner and Langner, 1997. Nature 388, 383-386]. The recently observed co-occurrence of the CB and the double CB (2CB) suggested an adaptation of the ICC frequency map to the extraction of the fundamental frequency f(0) [Braun, 1999. Hear. Res. 129, 71-82]. The present study examined a possible influence of this frequency map upon efferent signaling towards the cochlea. The f(0) distribution of 2890 monaural and 2604 binaural pairs of human spontaneous otoacoustic emissions was analyzed by three statistical methods and in each case showed non-random behavior in the CB-2CB range. Single results were (1) a bias of right ear f(0) (mode at 349 Hz) and left ear f(0) (mode at 262 Hz) towards different ranges of speech f(0) (P<0.02); (2) a bias of binaural, but not monaural, f(0) towards five of 12 semitone bins, C-G-D-A-E, representing the most frequent tones in music (P<0.003); (3) a bias of binaural, but not monaural, f(0) fine-distribution towards the exact pitch frequencies used in music, according to the international standard A4=440 Hz (P=0.03). The results support a model of lamina-based f(0) extraction in the ICC and suggest a specific colliculo-cochlear feedback for f(0) enhancement.

Auditory midbrain laminar structure appears adapted to f0 extraction: further evidence and implications of the double critical bandwidth

The psychoacoustic 'critical bandwidth' (CB), e.g. approximately 2.6 semitones (= 0.22 octave) at 1.5-3 kHz, is known from many spectral integration phenomena. Cat data suggest that it is represented in the inferior colliculus (IC) (Ehret and Merzenich, 1985, Science 227, 1245-1247), where it is consistently related to the fibrodendritic laminae (Schreiner and Langner, 1997, Nature 388, 383-386). The recent discovery of the CB and the double CB (2CB) in the statistics of frequency spacing of spontaneous otoacoustic emissions (Braun, 1997. Hear. Res. 114, 197-203) has initiated further investigations of the novel phenomenon of 2CB. Meta-analysis of psychoacoustic valuation studies of pure-tone intervals again revealed the effects of CB and 2CB. Valuations showed a significant stepwise change with interval size: <CB unpleasant, CB-2CB pleasant, >2CB indifferent. Scrutiny of cat and human data indicated that for both species, at least in the midspectrum (1-3 kHz in humans), the tonotopic ranges within single IC laminae and the tonotopic distances between neighboring laminae may equal 1 CB (distances to second next laminae being 2CB). This unique architecture would provide the most economical neural convergence of period information from pairs of adjacent harmonic partials 3-6 of complex sound. The resulting summed postsynaptic potentials would thus contain a beat frequency equaling f0 of sound input and being detectable by the known neural behavior of characteristic periodicity response.

Midbrain combinatorial code for temporal and spectral information in concurrent acoustic signals

All vocal species, including humans, often encounter simultaneous (concurrent) vocal signals from conspecifics. To segregate concurrent signals, the auditory system must extract information regarding the individual signals from their summed waveforms. During the breeding season, nesting male midshipman fish (Porichthys notatus) congregate in localized regions of the intertidal zone and produce long-duration (>1 min), multi-harmonic signals ("hums") during courtship of females. The hums of neighboring males often overlap, resulting in acoustic beats with amplitude and phase modulations at the difference frequencies (dFs) between their fundamental frequencies (F0s) and harmonic components. Behavioral studies also show that midshipman can localize a single hum-like tone when presented with a choice between two concurrent tones that originate from separate speakers. A previous study of the neural mechanisms underlying the segregation of concurrent signals demonstrated that midbrain neurons temporally encode a beat's dF through spike synchronization; however, spectral information about at least one of the beat's components is also required for signal segregation. Here we examine the encoding of spectral differences in beat signals by midbrain neurons. The results show that, although the spike rate responses of many neurons are sensitive to the spectral composition of a beat, virtually all midbrain units can encode information about differences in the spectral composition of beat stimuli via their interspike intervals (ISIs) with an equal distribution of ISI spectral sensitivity across the behaviorally relevant dFs. Together, temporal encoding in the midbrain of dF information through spike synchronization and of spectral information through ISI could permit the segregation of concurrent vocal signals.

Adjustable frequency selectivity of auditory forebrain neurons recorded in a freely moving songbird via radiotelemetry

One of the hearing system's basic properties that determines the detection of signals is its frequency selectivity. In the natural environment, a songbird may achieve an improved detection ability if the neuronal filters of its auditory system could be sharpened to adapt to the spectrum of the background noise. To address this issue, we studied 35 multi-unit clusters in the input layer of the primary auditory forebrain of nine European starlings (Sturnus vulgaris). Microelectrodes were chronically implanted in this songbird's cortex analogue and the neuronal activity was transmitted from unrestrained birds via a miniature FM transmitter. Frequency tuning curves (FTCs) and inhibitory sidebands were determined by presenting a matrix of frequency-level combinations of pure tones. From each FTC, the characteristic frequency (CF) and several parameters describing the neurons' filter characteristics were derived and compared to the same recording site's filter function while simultaneously stimulating with a continuous CF tone 20 dB above the response threshold. Our results show a significant improvement of frequency selectivity during two-tone stimulation, indicating that spectral filtering in the starling's auditory forebrain depends on the acoustic background in which a signal is presented. Moreover, frequency selectivity was found to be a function of the time over which the stimulus persisted, since FTCs were much sharper and inhibitory sidebands were largely expanded several milliseconds after response onset. Neuronal filter bandwidths during two-tone stimulation in the auditory forebrain are in good agreement with psychoacoustically measured critical bandwidths in the same species. Radiotelemetry proved to be a powerful tool in studying neuronal activity in freely behaving birds.

Auditory system: a neural substrate for frequency selectivity?

Recent results have provided new insights into the complexity of sound frequency representation in the auditory midbrain and may identify a neural substrate for behaviourally determined frequency resolution.

Frequency spacing of multiple spontaneous otoacoustic emissions shows relation to critical bands: a large-scale cumulative study

Multiple spontaneous otoacoustic emissions (SOAEs), recorded in one ear, are not randomly spaced on the frequency scale. Extent and origin of spacing order, however, are not clear. Therefore, the raw data of all human SOAE surveys were pooled, and the intervals of all possible emission pairs in each ear were in total outlined according to size on a distribution diagram (n = 5245, for intervals up to 2/3 octave). Prevalence was increased for intervals between the benchmarks of 1 and 2 critical bands (CB). This CB-2CB range was further characterized by preference of intervals with low-order frequency ratios like 5:4 or 6:5, whereas outside CB 2CB there were no such effects. The results are discussed in the context of current knowledge of the origin of critical bands. Experiments are proposed that test the hypothesis of an influence of the olivocochlear efferents on SOAE spacing.

Orderly cortical representation of vowels based on formant interaction

Psychophysical experiments have shown that the discrimination of human vowels chiefly relies on the frequency relationship of the first two peaks F1 and F2 of the vowel's spectral envelope. It has not been possible, however, to relate the two-dimensional (F1, F2)-relationship to the known organization of frequency representation in auditory cortex. We demonstrate that certain spectral integration properties of neurons are topographically organized in primary auditory cortex in such a way that a transformed (F1,F2) relationship sufficient for vowel discrimination is realized.

Laminar fine structure of frequency organization in auditory midbrain

The perception of sound is based on signal processing by a bank of frequency-selective auditory filters, the so-called critical bands. Here we investigate how the internal frequency organization of the main auditory midbrain station, the central nucleus of the inferior colliculus (ICC), might contribute to the generation of the critical-band behaviour of its neurons. We find a unique spatial arrangement of the frequency distribution in the ICC that correlates with psychophysical critical-band characteristics. Systematic frequency discontinuities along the main tonotopic axis, in combination with a smooth frequency gradient orthogonal to the main tonotopic organization of cat ICC, reflect a layering of the frequency organization paralleling its anatomical laminae. This layered frequency organization is characterized by constant frequency ratios of corresponding locations on neighbouring laminae and may provide a spatial framework for the generation of critical bands and for signal processing within and across frequency bands for the analysis of sound.

Psychophysical frequency modulation thresholds in a FM-bat, Tadarida brasiliensis

Echolocating bats hunting flying insects discriminate complex temporal patterns of acoustic stimuli. For bats using frequency modulated sonar calls (FM bats), there are no behavioral data on the perception of sinusoidally frequency modulated (SFM) stimuli. Discrimination performance for SFM stimuli of varying modulation depth was measured in 4 Tadarida brasiliensis in a two-alternative, forced choice procedure. A center frequency of 40 kHz was modulated with rates between 10 and 2000 Hz. It was found that discrimination performance improved from a mean threshold modulation depth of 3.05 kHz at a modulation rate of 2000 Hz to 1.58 kHz at a modulation rate of 10 Hz. Psychoacoustical modulation depth thresholds of T. brasiliensis are thus distinctly larger than those observed in bat species emitting constant frequency (CF) components followed by an FM-sweep, in active echolocation experiments. The modulation thresholds of T. brasiliensis are discussed in connection with the ability of bats to discriminate insect wingbeats. A comparison between non-echolocating mammals and the FM bat T. brasiliensis shows that the ability to echolocate is not reflected in the modulation thresholds.

Receptive Fields of Neurons in the Owl's Auditory Brainstem Change Dynamically

Binaural neurons in the barn owl's auditory brainstem have spatial receptive fields. It is shown here that both the frequency tuning of these neurons and their tuning to interaural time difference (ITD), the prime cue for azimuthal sound localization, improves with time after stimulus onset, a process I shall term 'dynamic sharpening'. Thus, the receptive fields of these neurons also have a temporal dimension. Data were collected in four hierachically ordered nuclei concerned with the computation of ITD: the nucleus ventralis lemnisci lateralis, pars anterior (VLVa), and three subnuclei in the inferior colliculus. Dynamic sharpening in the frequency tuning curves is evident from a dynamic reduction of tuning width. When stimulated with a tone, the response of all neurons varies in a cyclic manner with ITD. The ITDs at the response peaks differ by one period of the stimulus tone. The responses with noise stimulation are similar to the responses with tonal stimulation in all but the hierarchically highest nucleus, the external nucleus of the inferior colliculus. In this nucleus are found neurons that, if stimulated with noise, respond maximally only to one ITD while the responses at the other peaks are suppressed. This sidepeak suppression is a dynamic process. Dynamic sharpening of ITD tuning is also evident from a dynamic reduction of the tuning width around each of the response peaks. The proportion of neurons showing dynamic sharpening of ITD tuning is the same in all collicular subnuclei, but is lower in VLVa. Therefore, a major component of dynamic sharpening of ITD tuning occurs at the first station of the inferior colliculus. Lateral inhibition is one of the mechanisms underlying dynamic sharpening. Part of the inhibition may be mediated by GABA (Fujita and Konishi, 1988). The function of dynamic sharpening of ITD tuning may be to increase the fine representation of auditory space in single neurons.

Effect of noise masking on the brain-stem and middle-latency auditory evoked potentials: central and peripheral components

Brain-stem (BAEP) and middle-latency (MLAEP) auditory evoked potentials were recorded in zero noise and in 3 levels of continuous ipsilateral broadband noise. New information is presented on the effects of noise on BAEP wave I. Latency of wave I was not changed by increasing noise, but wave V latency linearly increased. Amplitude of waves I and V decreased non-linearly. The amplitude decrease was equivalent for both waves and occurred only at the higher noise levels. The dissociation of latency effects for waves I and V indicates a central component for the effect of noise on latency. The parallel amplitude change for waves I and V suggests a largely peripheral component. The MLAEP Pa latency also increased with increasing noise further supporting a central mechanism.

Neuronal discharge rate is unsuitable for encoding sound intensity at the inferior-colliculus level

Rate-intensity functions from single neurons in the central nucleus of the inferior colliculus (ICC) of the cat in response to tone bursts and continuous noise were recorded. Only 6% of 64 quantitatively studied neurons had monotonic functions in response to tone bursts, 12.5% in response to continuous noise. The other neurons had functions with a single peak which could occur at any super-threshold level tested (3-80 dB), or with multiple peaks. In 78% of the neurons the rate-intensity functions in response to tones and noise were of different shape. We interpret this stimulus dependency of the shape of rate-intensity functions of most neurons as evidence against sound-intensity coding only on the basis of the shape of rate-intensity functions at the ICC level. Rate-intensity functions averaged from neurons with similar characteristic frequencies or from all neurons of our sample indicate a constant average discharge over at least 60 dB sound intensity. This excludes sound-intensity coding by means of an average neuronal discharge rate in the ICC.

Complex sound analysis (frequency resolution, filtering and spectral integration) by single units of the inferior colliculus of the cat

The central nucleus of the inferior colliculus (ICC) is a center of convergence of brainstem input and is critical for auditory information processing. Here, the analysis of complex sound spectra by single neurons in the ICC is investigated. Several measures of frequency resolution (excitatory/inhibitory tuning curves, effective bandwidths, critical ratio bands, critical bands derived using narrowband masking and two-tone separation paradigms) have been obtained from the responses of these neurons at sound pressure levels (SPL) up to 80 dB above the units' response thresholds (nearly 110 dB SPL). Among our results are the following: (1) Narrowband masking measures of critical bands from ICC neurons closely parallel behavioral measures using the same stimulus paradigm. (2) Frequency resolution power as measured by critical bandwidths varies little as a function of stimulus intensity. (3) Tuning curves of ICC neurons provide no simple basis for predicting the frequency filtering of the same neurons excited by complex sound spectra. (4) There is a frequency dependence of all measures of frequency resolution similar to that found in psychophysical determinations of critical bandwidths. That is, spatial frequency resolution in the cochlea is the origin for the resolution found in the ICC and in behavioral tests. (5) Lateral inhibition at the level of the ICC clearly plays a role in frequency resolution. (6) Frequency resolution is encoded by response rate changes of ICC neurons and is independent of tone response threshold, response latency, spontaneous activity, tone response type, binaural response type. It is concluded that spectral analysis of sound is established by processes, including lateral inhibition, independent of other basic response properties of neurons at the level of the ICC.

Frequency selectivity of hearing in the green treefrog, Hyla cinerea

Frequency selectivity of hearing was measured in the green treefrog, Hyla cinerea. A psychophysical technique based on reflex modification was used to obtain masked threshold estimates for pure tones (300-5,400 Hz) presented against two levels of broadband masking noise. A pure tone (S-1) presented 200 ms prior to a reflex-eliciting stimulus (S-2) inhibited the motor reflex response to S-2. The magnitude of this reflex modification effect varied systematically with the sound pressure level (SPL) of S-1, and threshold was defined as the SPL of S-1 at which the reflex modification effect disappeared. Masked thresholds were used to calculate critical ratios, an index of the auditory system's frequency selectivity. The frequency selectivity of the treefrog's hearing is greatest and critical ratios are lowest (22-24 dB) at about 900 and 3,000 Hz, the two spectral regions dominant in the male treefrog's species-specific advertisement call. These results suggest that the treefrog's auditory system may be specialized to reject noise at biologically-relevant frequencies. As in other vertebrates, critical ratios remain constant when background noise level is varied; however, the shape of the treefrog's critical ratio function across frequencies differs from the typical vertebrate function that increases with increasing frequency at a slope of about 3 dB/octave. Instead, the treefrog's critical ratio function resembles its pure tone audiogram. Although the shape of the treefrog's critical ratio function is atypical, the critical ratio values themselves are comparable to those of many other vertebrates in the same frequency range. Critical ratio values here measured behaviorally do not match critical ratio values previously measured physiologically in single eighth nerve fibers.(ABSTRACT TRUNCATED AT 250 WORDS)