Relation between intrinsic connections and isofrequency contours in the inferior colliculus of the big brown bat, Eptesicus fuscus.

Information processing in the inferior colliculus depends on interactions between ascending pathways and intrinsic circuitry, both of which exist within a functional tonotopic organization. To determine how local projections of neurons in the inferior colliculus are related to tonotopy, we placed a small iontophoretic injection of biodextran amine at a physiologically characterized location in the inferior colliculus. We then used electrophysiological recording to place a grid of small deposits of Chicago Sky Blue throughout the same frequency range to specify an isofrequency contour. Using three-dimensional computer reconstructions, we analyzed patterns of transport relative to the physiologically determined isofrequency contour to quantify the extent of the intrinsic connection lamina in all three dimensions. We also performed a quantitative analysis of the numbers of cells in different regions relative to the biodextran amine injection. Biodextran amine-labeled fibers were mainly located dorsomedial to the injection site, confined within the isofrequency contour, but biodextran amine-labeled cells were mainly located ventrolateral to the injection site. When we counted numbers of labeled cells classified by morphological type, we found that both elongate and multipolar cells were labeled within the isofrequency contour. Because the dendrites of multipolar cells typically extend outside the isofrequency lamina, it is likely that they receive input from other isofrequency contours and relay it to more dorsomedial portions of their specific isofrequency contour, along with the frequency-specific projections of the elongate cells. Within a given isofrequency contour, there is a consistent organization in which intrinsic connections ascend from the ventrolateral portion to more dorsomedial points along the contour, forming a cascaded system of intrinsic feedforward connections that seem ideally suited to provide the delay lines necessary to produce several forms of selectivity for temporal patterns in inferior colliculus neurons.

Medial superior olive of the big brown bat: neuronal responses to pure tones, amplitude modulations, and pulse trains.

The structure and function of the medial superior olive (MSO) is highly variable among mammals. In species with large heads and low-frequency hearing, MSO is adapted for processing interaural time differences. In some species with small heads and high-frequency hearing, the MSO is greatly reduced in size; in others, including those echolocating bats that have been examined, the MSO is large. Moreover, the MSO of bats appears to have undergone different functional specializations depending on the type of echolocation call used. The echolocation call of the mustached bat contains a prominent CF component, and its MSO is predominantly monaural; the free-tailed bat uses pure frequency-modulated calls, and its MSO is predominantly binaural. To further explore the relation of call structure to MSO properties, we recorded extracellularly from 97 single neurons in the MSO of the big brown bat, Eptesicus fuscus, a species whose echolocation call is intermediate between that of the mustached bat and the free-tailed bat. The best frequencies of MSO neurons in the big brown bat ranged from 11 to 79 kHz, spanning most of the audible range. Half of the neurons were monaural, excited by sound at the contralateral ear, while the other half showed evidence of binaural interactions, supporting the idea that the binaural characteristics of MSO neurons in the big brown bat are midway between those of the mustached bat and the free-tailed bat. Within the population of binaural neurons, the majority were excited by sound at the contralateral ear and inhibited by sound at the ipsilateral ear; only 21% were excited by sound at either ear. Discharge patterns were characterized as transient ON (37%), primary-like (33%), or transient OFF (23%). When presented with sinusoidally amplitude modulated tones, most neurons had low-pass filter characteristics with cutoffs between 100 and 300 Hz modulation frequency. For comparison with the sinusoidally modulated sounds, we presented trains of tone pips in which the pulse duration and interstimulus interval were varied. The results of these experiments indicated that it is not the modulation frequency but rather the interstimulus interval that determines the low-pass filter characteristics of MSO neurons.

Neural measurement of sound duration: control by excitatory-inhibitory interactions in the inferior colliculus.

In the inferior colliculus (IC) of the big brown bat, a subpopulation of cells ( approximately 35%) are tuned to a narrow range of sound durations. Band-pass tuning for sound duration has not been seen at lower levels of the auditory pathway. Previous work suggests that it arises at the IC through the interaction of sound-evoked, temporally offset, excitatory and inhibitory inputs. To test this hypothesis, we recorded from duration-tuned neurons in the IC and examined duration tuning before and after iontophoretic infusion of antagonists to gamma-aminobutyric acid-A (GABA(A)) (bicuculline) or glycine (strychnine). The criterion for duration tuning was that the neuron's spike count as a function of duration had a peak value at one duration or a range of durations that was >/=2 times the lowest nonzero value at longer durations. Out of 21 units tested with bicuculline, duration tuning was eliminated in 15, broadened in two, and unaltered in four. Out of 10 units tested with strychnine, duration tuning was eliminated in four, broadened in one, and unaltered in five. For units tested with both bicuculline and strychnine, bicuculline had a greater effect on reducing or abolishing duration tuning than did strychnine. Bicuculline and strychnine both produced changes in discharge pattern. There was nearly always a shift from an offset response to an onset response, indicating that in the predrug condition, inhibition arrived simultaneously with excitation or preceded it. There was often an increase in the length of the spike train, indicating that in the predrug condition, inhibition also coincided with later parts of excitation. These findings support two hypotheses. First, duration tuning is created in the IC. Second, although the construction of duration tuning varies in some details among IC neurons, it follows three rules: 1) an excitatory and an inhibitory event are temporally linked to the onset of sound but temporally offset from one another; 2) the duration of some inhibitory event must be linked to the duration of the sound; 3) an excitatory event must be linked to the offset of sound.

Timing in the auditory system of the bat.

Echolocating bats use audition to guide much of their behavior. As in all vertebrates, their lower brainstem contains a number of parallel auditory pathways that provide excitatory or inhibitory outputs differing in their temporal discharge patterns and latencies. These pathways converge in the auditory midbrain, where many neurons are tuned to biologically important parameters of sound, including signal duration, frequency-modulated sweep direction, and the rate of periodic frequency or amplitude modulations. This tuning to biologically relevant temporal patterns of sound is created through the interplay of the time-delayed excitatory and inhibitory inputs to midbrain neurons. Because the tuning process requires integration over a relatively long time period, the rate at which midbrain auditory neurons respond corresponds to the cadence of sounds rather than their fine structure and may provide an output that is closely matched to the rate at which motor systems operate.

Arctiid moths and bat echolocation: broad-band clicks interfere with neural responses to auditory stimuli in the nuclei of the lateral lemniscus of the big brown bat.

Clicks emitted by arctiid moths interfere with the ranging ability of echolocating bats. To identify possible neural correlates of this interference, we recorded responses of single units in the nuclei of the lateral lemniscus to combinations of a broad-band click and a test signal (pure tones or frequency-modulated sweeps). In 77% of 87 units tested, clicks interfered with neural responses to the test stimuli. The interference fell into two categories: latency ambiguity and suppression. Units showing latency ambiguity responded to both the click and the test signal. However, when the click occurred within a window of approximately 3 ms before the onset of the test signal, the latency of the response to the test signal was affected. Units that were suppressed did not respond to clicks. Nevertheless, when a click was presented immediately before or simultaneously with a test signal, the response to the test signal was eliminated. Both types of units were found throughout the lateral lemniscus except for the columnar division of the ventral nucleus, where all units tested exhibited latency ambiguity. There is a close match between the single unit data and previous studies of range difference discrimination in the presence of clicks.

The columnar region of the ventral nucleus of the lateral lemniscus in the big brown bat (Eptesicus fuscus): synaptic arrangements and structural correlates of feedforward inhibitory function.

Neurons of the columnar region of the ventral nucleus of the lateral lemniscus of Eptesicus fuscus respond with high-precision constant-latency responses to sound onsets and possess remarkably broad tuning. To study the synaptic basis for this specialized monaural auditory processing and to elucidate the excitatory or inhibitory nature of the input and output circuitry, we have used classical transmission electron microscopy, and postembedding immunocytochemistry for gamma aminobutyric acid (GABA) and glycine on serial semithin sections. The dominant putatively excitatory perisomatic input is provided by large calyx-like terminals that possess round synaptic vesicles and asymmetric synaptic contacts. Additionally, calyces contact the dendrites of neighboring neurons. Putatively inhibitory small boutons possess pleomorphic or flattened synaptic vesicles and symmetrical contacts and are sparsely distributed on somata and dendrites. Almost all neurons are glycine-immunoreactive. There is a moderate amount of glycine-immunoreactive puncta; GABA-immunoreactive puncta are rare. This suggests that (1) there is a fast robust excitatory synaptic input via calyx-like perisomatic endings, (2) calyx-like endings distribute frequency-specific excitatory input across isofrequency sheets by virtue of parallel synapses to somata and adjacent dendrites, and thus, dendritic integration may contribute to the broadening of frequency tuning, (3) the columnar region forms an inhibitory glycinergic feedforward relay in the ascending auditory pathway, a relay that is probably involved in creating filters for time-varying signals.

Neural tuning to sound duration in the inferior colliculus of the big brown bat, Eptesicus fuscus.

Neural tuning to different sound durations may be a useful filter for identification of certain sounds, especially those that are biologically important. The auditory midbrains of mammals and amphibians contain neurons that appear to be tuned to sound duration. In amphibians, neurons are tuned to durations of sound that are biologically important. The purpose of this study was to characterize responses of neurons in the inferior colliculus (IC) of the big brown bat, Eptesicus fuscus, to sounds of different durations. Our aims were to determine what percent of neurons are duration tuned and how best durations are correlated to durations of echolocation calls, and to examine response properties that may be relevant to the mechanism for duration tuning, such as latency and temporal firing pattern; we also examined frequency tuning and rate-level functions. We recorded from 136 single units in the central nucleus of the IC of unanesthetized bats. The stimuli were pure tones, frequency-modulated sweeps, and broadband noise. The criterion for duration tuning was an increase in spike count of > or = 50% at some durations compared with others. Of the total units sampled, 36% were tuned to stimulus duration. All of these units were located in the caudal half of the IC. Best duration for most units ranged from < 1 to 10 ms, but a few had best durations up to > or = 20 ms. This range is similar to the range of durations of echolocation calls used by Eptesicus. All duration-tuned neurons responded transiently. The minimum latency was always longer than the best duration. Duration-tuned units have best durations and best frequencies that match the temporal structure and frequency range of the echolocation calls. Thus the results raise the hypothesis that neurons in the IC of Eptesicus, and probably the auditory midbrain of other vertebrates, are tuned to biologically important sound durations. We suggest a model for duration tuning consisting of three components: 1) inhibitory input that is correlated with the onset of the stimulus and is sustained for the stimulus duration; 2) transient excitation that is correlated with the offset of the stimulus; and 3) transient excitation that is correlated with the onset of the stimulus but is delayed in time relative to the onset of inhibition. For the neuron to fire, the two excitatory events must coincide in time; noncoincident excitatory events are not sufficient.

Neural selectivity and tuning for sinusoidal frequency modulations in the inferior colliculus of the big brown bat, Eptesicus fuscus.

Most communication sounds and most echolocation sounds, including those used by the big brown bat (Eptesicus fuscus), contain frequency-modulated (FM) components, including cyclical FM. Because previous studies have shown that some neurons in the inferior colliculus (IC) of this bat respond to linear FM sweeps but not to pure tones or noise, we asked whether these or other neurons are specialized for conveying information about cyclical FM signals. In unanesthetized bats, we tested the response of 116 neurons in the IC to pure tones, noise with various bandwidths, single linear FM sweeps, sinusoidally amplitude-modulated signals, and sinusoidally frequency-modulated (SFM) signals. With the use of these stimuli, 20 neurons (17%) responded only to SFM, and 10 (9%) responded best to SFM but also responded to one other test stimulus. We refer to the total 26% of neurons that responded best to SFM as SFM-selective neurons. Fifty-nine neurons (51%) responded about equally well to SFM and other stimuli, and 27 (23%) did not respond to SFM but did respond to other stimuli. Most SFM-selective neurons responded to a limited range of modulation rates and a limited range of modulation depths. The range of modulation rates over which individual neurons responded was 5-170 Hz (n = 20). Thus SFM-selective neurons respond to low modulation rates. The depths of modulations to which the neurons responded ranged from +/-0.4 to +/-19 kHz (n = 15). Half of the SFM-selective neurons did not respond to the first cycle of SFM. This finding suggests that the mechanism for selective response to SFM involves neural delays and coincidence detectors in which the response to one part of the SFM cycle coincides in time either with the response to a later part of the SFM cycle or with the response to the first part of the next cycle. The SFM-selective neurons in the IC responded to a lower and more limited range of SFM rates than do neurons in the nuclei of the lateral lemniscus of this bat. Because the FM components of biological sounds usually have low rates of modulation, we suggest that the tuning of these neurons is related to biologically important sound parameters. The tuning could be used to detect FM in echolocation signals, modulations in high-frequency sounds that are generated by wing beats of some beetles, or social communication sounds of Eptesicus.

Whole-cell patch-clamp recording reveals subthreshold sound-evoked postsynaptic currents in the inferior colliculus of awake bats.

The inferior colliculus receives excitatory and inhibitory input from parallel auditory pathways that differ in discharge patterns, latencies, and binaural properties. Processing in the inferior colliculus may depend on the temporal sequence in which excitatory and inhibitory synaptic inputs are activated and on the resulting balance between excitation and inhibition. To explore this issue at the cellular level, we used the novel approach of whole-cell patch-clamp recording in the midbrain of awake bats (Eptesicus fuscus) to record EPSCs or IPSCs. Sound-evoked EPSCs were recorded in most neurons. These EPSCs were frequently preceded by an IPSC, followed by an IPSC, or both. These findings help explain the large latency range and transient responses that characterize inferior colliculus neurons. The EPSC was sometimes followed by long-lasting oscillatory currents, suggesting that a single brief sound sets up a pattern of altered excitability that persists far beyond the duration of the initial sound. In three binaural neurons, ipsilateral sound evoked a large IPSC that partially or totally canceled the EPSC evoked by contralateral sound. In one binaural neuron with ipsilaterally evoked IPSCs, contralaterally evoked IPSCs occurred in response to frequencies above and below the neuron's best frequency. Thus, both monaural and binaural interactions can occur at single inferior colliculus neurons. These results show that whole-cell patch-clamp recording offers a powerful means of understanding how subthreshold processes determine the responses of auditory neurons.

Distribution of GABAA, GABAB, and glycine receptors in the central auditory system of the big brown bat, Eptesicus fuscus.

Quantitative autoradiographic techniques were used to compare the distribution of GABAA, GABAB, and glycine receptors in the subcortical auditory pathway of the big brown bat, Eptesicus fuscus. For GABAA receptors, the ligand used was 35S-t-butylbicyclophosphorothionate (TBPS) for GABAB receptors, 3H-GABA was used as a ligand in the presence of isoguvacine to block binding to GABAA sites; for glycine, the ligand used was 3H-strychnine. In the subcortical auditory nuclei there appears to be at least a partial complementarity in the distribution of GABAA receptors labeled with 35S-TBPS and glycine receptors labeled with 3H-strychnine, GABAA receptors were concentrated mainly in the inferior colliculus (IC) and medial geniculate nucleus, whereas glycine receptors were concentrated mainly in nuclei below the level of the IC. Within the IC, there was a graded spatial distribution of 35S-TBPS binding; the most dense labeling was in the dorsomedial region, but very sparse labeling was observed in the ventrolateral region. There was also a graded spatial distribution of 3H-strychnine binding. The most dense labeling was in the ventral and lateral regions and the weakest labeling was in the dorsomedial region. Thus, in the IC, the distribution of 35S-TBPS was complementary to that of 3H-strychnine. GABAB receptors were distributed at a low level throughout the subcortical auditory nuclei, but were most prominent in the dorsomedial part of the IC.

Spatial tuning of neurons in the inferior colliculus of the big brown bat: effects of sound level, stimulus type and multiple sound sources.

We examined factors that affect spatial receptive fields of single units in the central nucleus of the inferior colliculus of Eptesicus fuscus. Pure tones, frequency- or amplitude-modulated sounds, or noise bursts were presented in the free-field, and responses were recorded extracellularly. For 58 neurons that were tested over a 30 dB range of sound levels, 7 (12%) exhibited a change of less than 10 degrees in the center point and medical border of their receptive field. For 28 neurons that were tested with more than one stimulus type, 5 (18%) exhibited a change of less than 10 degrees in the center point and medial border of their receptive field. The azimuthal response ranges of 19 neurons were measured in the presence of a continuous broadband noise presented from a second loudspeaker set at different fixed azimuthal positions. For 3 neurons driven by a contralateral stimulus only, the effect of the noise was simple masking. For 11 neurons driven by sound at either side, 8 were unaffected by the noise and 1 showed a simple masking effect. For the remaining 2, as well as for 5 neurons that were excited by contralateral sound and inhibited by ipsilateral sound, the peak of the azimuthal response range shifted toward the direction of the noise.

A neuroethological theory of the operation of the inferior colliculus.

A general statement of the function of the inferior colliculus is lacking, even after more than three decades of electrophysiological investigation. A neuroethological theory is proposed that accounts for a large and diverse body of evidence. Although aimed at characterizing the inferior colliculus in mammals, the theory also applies generally to the auditory midbrain in vertebrates. The theory has two hypotheses: (1) Tuning processes in the inferior colliculus are related to the biological importance of sounds. (2) There is a change in timing properties at the inferior colliculus, from rapid input to slowed output; this transformation is related to the timing of specific behaviors. Expressed in neuroethological terms, at least some neurons in the inferior colliculus are tuned to sign-stimuli, and the processing of these sign stimuli triggers fixed action patterns for hunting, escape or vocal communication. The resulting temporal transformation adjusts the pace of sensory input to the pace of behavior. Evidence for the theory comes from anatomical, neurophysiological and behavioral studies and includes: (1) massive convergence of parallel auditory pathways at the inferior colliculus, (2) interaction of the inferior colliculus with motor systems, (3) tuning of auditory midbrain neurons to biologically important sounds, (4) the slow pace of neural processing at the inferior colliculus, (5) the slow pace of motor output. The theory has the following implications. Neurons in the inferior colliculus are filters for sounds that require immediate action, such as certain sounds made by prey, predators or conspecifics. Neural processing in the inferior colliculus is species specific, resulting in filtering for these kinds of sounds. Specific action patterns should be correlated with the activity of neurons in the inferior colliculus. Motor activities may modify neural processing in inferior colliculus neurons. The rate at which information is transmitted to the thalamus is regulated by the inferior colliculus.

Convergence and divergence of ascending binaural and monaural pathways from the superior olives of the mustached bat.

The lateral superior olive and medial superior olive give rise to pathways that terminate in the dorsal nucleus of the lateral lemniscus and central nucleus of the inferior colliculus. In most mammals, neurons in both the medial and lateral superior olives are binaural, but in the mustached bat most neurons in the medial superior olive are monaural. The aims of this study were to determine how the inputs to the medial superior olive contribute to its monaurality and to determine whether the ascending projections from the lateral and medial superior olives overlap or remain segregated at their targets. Injections of two different tracers were placed in tonotopically matched areas of the lateral and medial superior olives in the same animal. Retrograde transport from injections in the medial superior olive labeled spherical cells in the contralateral anteroventral cochlear nucleus and principal cells in the ipsilateral medial nucleus of the trapezoid body. Few cells were labeled in ipsilateral cochlear nucleus. Anterograde transport resulted in tonotopically specific distributions of label with the same laterality as in nonecholocating mammals. In the dorsal nucleus of the lateral lemniscus, label from the lateral and medial superior olives largely overlapped. In the inferior colliculus, label from the lateral and medial superior olives largely overlapped. In the inferior colliculus, label from the two sources overlapped in the high and low frequency ranges, but in the frequency range around 60 kHz, label from the medial superior olive extended more dorsally than that from the lateral superior olive. These results indicate that projections of the lateral and medial superior olives overlap extensively at their targets.

Frequency tuning and response latencies at three levels in the brainstem of the echolocating bat, Eptesicus fuscus.

To determine the level at which certain response characteristics originate, we compared monaural auditory responses of neurons in ventral cochlear nucleus, nuclei of lateral lemniscus and inferior colliculus. Characteristics examined were sharpness of frequency tuning, latency variability for individual neurons and range of latencies across neurons. Exceptionally broad tuning curves were found in the nuclei of the lateral lemniscus, while exceptionally narrow tuning curves were found in the inferior colliculus. Neither specialized tuning characteristic was found in the ventral cochlear nuclei. All neurons in the columnar division of the ventral nucleus of the lateral lemniscus maintained low variability of latency over a broad range of stimulus conditions. Some neurons in the cochlear nucleus (12%) and some in the inferior colliculus (15%) had low variability in latency but only at best frequency. Range of latencies across neurons was small in the ventral cochlear nucleus (1.3-5.7 ms), intermediate in the nuclei of the lateral lemniscus (1.7-19.8 ms) and greatest in the inferior colliculus (2.9-42.0 ms). We conclude that, in the nuclei of the lateral lemniscus and in the inferior colliculus, unique tuning and timing properties are built up from ascending inputs.

Neural tuning for sound duration: role of inhibitory mechanisms in the inferior colliculus.

Duration is a biologically important feature of sound. Some neurons in the inferior colliculus of the big brown bat, Eptesicus fuscus, are tuned to sound duration, but it is unclear at what level the tuning originates or what neural mechanisms are responsible for it. The application of antagonists of the inhibitory neurotransmitters gamma-aminobutyric acid or glycine to neurons in the inferior colliculus eliminated duration tuning. Whole-cell patch-clamp recordings of synaptic currents suggested that inhibition produces a temporal frame within which excitation can occur. A model is proposed in which duration tuning arises when an early, sustained inhibitory input interacts with a delayed, transient excitatory input.

Monaural interaction of excitation and inhibition in the medial superior olive of the mustached bat: an adaptation for biosonar.

In most mammals, the superior olive is the first stage for binaural interaction. Neurons in the medial superior olive (MSO) receive excitatory input from both ears and are sensitive to interaural time or phase differences of low-frequency sounds. The mustached bat (Pteronotus parnellii parnellii), a small echolocating species with high-frequency hearing, probably does not use interaural time or phase differences as cues for sound localization. Although the mustached bat has a large MSO, there is some evidence that it is functionally different from the MSO in nonecholocating mammals. Most MSO neurons in the mustached bat are monaural, excited by a contralateral sound. Their responses are phasic and correlated with either the onset or the offset of a sound. As a first step in determining the origin of these phasic monaural responses, we traced the connections of the MSO by using both retrograde and anterograde transport methods. Excitatory inputs to the MSO originate from spherical cells in the anteroventral cochlear nucleus, almost exclusively from the contralateral side. Glycinergic inhibitory input is relayed from the contralateral cochlear nucleus through the medial nucleus of the trapezoid body. To investigate the interactions of the contralateral excitatory and inhibitory inputs at the level of the MSO cell, we recorded sound-evoked responses and applied glycine or its antagonist by using microiontophoresis. The results show that the phasic response to a contralateral sound is created by interaction of a sustained excitatory input with a sustained inhibitory input, also from the contralateral ear. Whether the response is to the onset or offset of a sound is determined by the relative timing between the excitatory and inhibitory inputs. Thus, in MSO of the mustached bat, the ipsilateral excitatory pathway from the cochlear nucleus seen in animals with low-frequency hearing is virtually absent, and the MSO is adapted for timing analysis by using input from only the contralateral ear.

Frequency tuning properties of neurons in the inferior colliculus of an FM bat.

We examined frequency tuning characteristics of single neurons in the inferior colliculus of the echolocating bat, Eptesicus fuscus, in order to determine whether there are different classes of spectral selectivity at this level and to relate frequency tuning properties to the design of the echolocation signal. In unanesthetized but tranquilized animals, we recorded responses from 363 single units to pure tones, frequency-modulated (FM) sweeps, or broad-band noise. Most units were selective for stimulus type; 50% responded only to pure tones, 14% responded only to FM sweeps, and 5% responded only to noise. The remainder responded to two or more types of stimuli. Tuning curves could be classified as follows: 1) V-shaped tuning curves (57%) were the most common type; 2) closed tuning curves (20%) had thresholds at both low and high sound levels; 3) narrow filters (14%) had Q values above 20 at 10 dB and 30 dB above threshold or 10 dB and 40 dB above threshold; 4) frequency-opponent tuning (6%) was found in units with high spontaneous activity; within a center range of frequencies, firing rate increased above spontaneous level, but at higher or lower frequencies, firing rate decreased below spontaneous level; 5) double-tuned units (3%) had two best frequencies (BF). The most clear evidence of topographic distribution was seen for filter units, which were only found in the dorsal part of the 20-30 kHz isofrequency contour. Filter units were also the most clearly related to the echolocation signal of the bat. Their BFs were all within the range of the dominant frequency (approximately 20-30 kHz) that Eptesicus uses during the searching phase of echolocation.

The monaural nuclei of the lateral lemniscus in an echolocating bat: parallel pathways for analyzing temporal features of sound.

In echolocating bats, three cell groups in the lateral lemniscus are conspicuous for their large size and high degree of differentiation. These cell groups are the intermediate nucleus (INLL), columnar nucleus (VNLLc), and multipolar cell area (VNLLm). All receive projections from the contralateral cochlear nucleus. Previous anatomical studies suggest the hypothesis that these nuclei are important for analyzing the temporal structure of sound. To investigate this possibility, we recorded responses of single units in the INLL, VNLLc, and VNLLm of Eptesicus fuscus. The results show that each cytoarchitectural division contains a complete tonotopic representation. Certain response properties are common to all three nuclei. First, virtually all units are monaural. Second, all are broadly tuned to frequency; their average Q10dB value of 9.1 is considerably lower than values measured in the inferior colliculus of Eptesicus. Third, most units have little or no spontaneous activity. Fourth, all have short integration times, responding robustly to stimuli less than 5 msec in duration. The broad tuning, lack of spontaneous activity, and short integration time all make these neurons well suited for the accurate encoding of temporal information. Although there are many similarities, there are also important differences among nuclei. The clearest evidence of specialization is in VNLLc. Neurons here are more broadly tuned than those in INLL or VNLLm, have no spontaneous activity, and always respond with one spike per stimulus. The latency of the spike is precisely locked to the stimulus onset, with variability from trial to trial as low as 0.03 msec. In addition, the latency remains constant over large variations in frequency or intensity. In INLL and VNLLm, response patterns are about equally distributed between tonic, chopping, and phasic; there are no single-spike constant-latency responses of the type seen in VNLLc, although some choppers and pausers do respond with constant first-spike latency. The results indicate that VNLLc is specialized to encode very precisely the onset of sound; the other nuclei may encode ongoing properties of a sound.

Binaural properties of single units in the superior olivary complex of the mustached bat.

1. Previous studies of the superior olive of echolocating bats suggest that the lateral superior olive (LSO) retains the same structure and function as in other mammals but that the medial superior olive (MSO) is different in structure and possibly also in function. The present study is an examination of this idea in Pteronotus parnellii, a bat that has a large and well-defined MSO. 2. Using pure tones presented via earphones, we obtained data on frequency tuning for 60 single units and 96 multiunits in LSO and 94 single units and 154 multiunits in MSO. Of these we also obtained binaural response characteristics from 55 single units in LSO and 72 single units in MSO. 3. LSO and MSO each have a complete tonotopic representation, arranged in a sequence similar to that of other mammals studied. However, in both LSO and MSO there is an expanded representation of the frequencies around 60 kHz, the main frequency component of the bat's echolocation call; there is another expanded representation of the range around 90 kHz, the third harmonic of the call. The expansion of these frequency ranges suggests that the functions of LSO and MSO in Pteronotus are related to echolocation behavior. 4. The binaural characteristics of cells in LSO were essentially the same as those seen in other mammals. Most LSO units (93%) were excited by the ipsilateral ear and inhibited by the contralateral ear. The responses of nearly all LSO units were completely suppressed when the sound level at the two ears was equal. 5. The binaural characteristics of cells in MSO were different from those in nonecholocating mammals. Most MSO units (72%) were excited by the contralateral ear but were neither excited nor inhibited by the ipsilateral ear. Of the remaining units, 21% were excited by the contralateral ear and inhibited by the ipsilateral ear, and only 6% were excited by both ears. 6. The temporal discharge patterns of units in MSO differed from the tonic response pattern seen in LSO. Most MSO units had phasic response patterns, with a few spikes at the onset or offset of the stimulus; the response often changed from ON to OFF depending on stimulus frequency. 7. The results support the idea that in evolution LSO has remained unchanged, whereas MSO has undergone adaptation. The function of LSO in Pteronotus seems to be identical to that in other mammals, i.e., analysis of interaural sound level differences to derive azimuthal location. The function of MSO in Pteronotus must be different from that in nonecholocating mammals.(ABSTRACT TRUNCATED AT 400 WORDS)

Central acoustic tract in an echolocating bat: an extralemniscal auditory pathway to the thalamus.

To determine the sources and targets of auditory pathways that bypass the inferior colliculus in the mustache bat, we injected WGA-HRP in the medial geniculate body and related auditory nuclei of the thalamus as well as in the lower brainstem. We used electrophysiological methods to verify that the injection electrode was in an area responsive to sound. The only thalamic injections that produced retrograde transport to cells in auditory nuclei caudal to the inferior colliculus were those that included the suprageniculate nucleus. These injections labeled a group of large multipolar cells lying between the ventral nucleus of the lateral lemniscus and the superior olivary complex. Neurons in this cell group have also been shown to project to the deep layers of the superior colliculus in the mustache bat. The pathway revealed by these studies is almost identical to the "central acoustic tract" in which fibers course medial to the lateral lemniscus and bypass the inferior colliculus to reach the deep superior colliculus and the suprageniculate nucleus.