| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Zoologisches Institut, Universität München, D-80333 Munich, Germany; and 2 Neurobiology Group, Department of Biological Sciences, University of Illinois at Chicago, Chicago, Illinois 60607-7060
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Grothe, Benedikt, Thomas J. Park, and Gerd Schuller. Medial superior olive in the free-tailed bat: response to pure tones and amplitude-modulated tones. J. Neurophysiol. 77: 1553-1565, 1997. In mammals with good low-frequency hearing and a moderate to large interear distance, neurons in the medial superior olive (MSO) are sensitive to interaural time differences (ITDs). Most small mammals, however, do not hear low frequencies and do not experience significant ITDs, suggesting that their MSOs participate in functions other than ITD coding. In one bat species, the mustached bat, the MSO is a functionally monaural nucleus, acting as a low-pass filter for the rate of sinusoidally amplitude-modulated (SAM) stimuli. We investigated whether the more typical binaural MSO of the Mexican free-tailed bat also acts as an SAM filter. We recorded from 60 MSO neurons with their best frequencies covering the entire audiogram of this bat. The majority revealed bilateral excitation and indirect evidence for inhibition (EI/EI; 55%). The remaining neurons exhibited reduced inputs, mostly lacking ipsilateral inputs (28% I/EI; 12% O/EI; 5% EI/O). Most neurons (64%) responded with a phasic discharge to pure tones; the remaining neurons exhibited an additional sustained component. For stimulation with pure tones, two thirds of the cells exhibited monotonic rate-level functions for ipsilateral, contralateral, or binaural stimulation. In contrast, nearly all neurons exhibited nonmonotonic rate-level functions when tested with SAM stimuli. Eighty-eight percent of the neurons responded with a phase-locked discharge to SAM stimuli at low modulation rates and exhibited low-pass filter characteristics in the modulation transfer function (MTF) for ipsilateral, contralateral, and binaural stimulation. The MTF for ipsilateral stimulation usually did not match that for contralateral stimulation. Introducing interaural intensity differences (IIDs) changed the MTF in unpredictable ways. We also found that responses to SAMs depended on the carrier frequency. In some neurons we measured the time course of the ipsilaterally and contralaterally evoked inhibition by presenting brief frequency-modulated sweeps at different ITDs. The duration and timing of inhibition could be related to the SAM cutoff for binaural stimulation. We conclude that the response of the MSO in the free-tailed bat is created by a complex interaction of inhibition and excitation. The different time constants of inputs create a low-pass filter for SAM stimuli. However, the MSO output is an integrated response to the temporal structure of a stimulus as well as its azimuthal position, i.e., IIDs. There are no in vivo results concerning filter characteristics in a "classical" MSO, but our data confirm an earlier speculation about this interdependence based on data accessed from a gerbil brain slice preparation.
The superior olivary complex (SOC) is the first station of binaural interaction within the mammalian ascending auditory pathway. The medial superior olive (MSO), one of several subdivisions within the SOC, is traditionally thought to function in the context of sound localization (Goldberg and Brown 1968 Five free-tailed bats were used in this study. During surgery the bats were anesthetized with pentobarbital sodium (15 mg/kg) and methoxyflurane inhalation. Skin and muscles were deflected from the upper part of the skull and a metal rod that was later used to secure the bat's head during recordings was mounted to the skull with the use of cyanoacrylate and dental cement. A small hole (0.5-1 mm diam) was cut over the inferior colliculus on one side. The stereotaxic procedure described by Schuller et al. (1986) Anatomic connections of MSO neurons in the
free-tailed bat
The main connections of the free-tailed bat MSO have been described earlier (Grothe et al. 1994
Distribution of BF
We recorded from 60 MSO units. Their BFs covered the range from 9 to 81 kHz, with a bias to BFs between 20 and 30 kHz (21 = 35%). This corresponds well with the audiograms and the biosonar signal used by the free-tailed bat. Its echolocation call consists of a multiharmonic downward FM sweep that ends in a constant-frequency component (Fenton and Bell 1981
Binaural characteristics
Consistent with the bilateral inputs to the MSO, the majority of neurons (55%, 33 of 60) could be stimulated by presentation of pure tones or amplitude-modulated tones to either ear alone, but (except for 2 neurons) these cells also exhibited indirect evidence for inhibitory inputs from both ears (EI/EI). Evidence of inhibition uncluded phasic response patterns, nonmonotonic rate-level functions, low SAM cutoffs, and IID functions with decreasing response for certain IIDs. This, of course, only suggests inhibitory influences and will be discussed in detail in the DISCUSSION. About 40% of MSO units (24 of 60) could not be driven by the ipsilateral ear and for these units our data did not provide any evidence for facilitatory effects from the ipsilateral ear either. However, most of these units (17 of 24) exhibited prominent inhibition elicited by stimulation of the ipsilateral as well as by stimulation of the contralateral ear (I/EI neurons). In five of these I/EI-type units, ipsilateral induced inhibitory effects could only be seen when the signal at the ipsilateral ear was advanced or delayed a few milliseconds. The remaining neurons showed no effects from ipsilateral stimulation but could be excited from contralateral stimulation with signs of inhibitory effects (O/EI neurons). Three units were excited by ipsilateral stimulation but without apparent contralateral influences (EI/O units). The distribution of binaural types is summarized in Fig. 2B.
Discharge patterns
The majority of MSO neurons (64%, 38 of 60) responded with a phasic discharge pattern when stimulated with a pure tone 20 dB above threshold. Most of these units (35) exhibited a phasic ON response; three showed a phasic OFF response. Sustained discharge patterns were found in 21 (36%) of units, mostly in the form of a primary-like discharge including a prominent ON response (19%), or as a tonic response without a prominent ON component (14%). Two neurons exhibited a chopper and two other neurons a pauser discharge pattern. The distribution of response patterns is shown in Fig. 2C.
Rate-level functions: pure tones
We measured rate-level functions from 43 MSO neurons with the use of ipsilateral, contralateral, or binaural stimulation. The rate-level functions were measured with the use of 100-ms pure tones, with the stimulus intensity increasing in 10-dB steps from threshold up to 40 dB above threshold. Because of the small interaural distance in this small animal, there is an acoustic cross talk from one to the other ear that occurs at ~45-50 dB. Therefore purely monaural rate-level functions can only be measured over a range of ~40 dB. For higher levels, binaural effects complicate the interpretation. In general, we observed a similar proportion of monotonic and nonmonotonic functions in response to pure tones. Of 41 neurons tested with the use of contralateral stimulation, 17 exhibited monotonic rate-level functions whereas 24 had nonmonotonic rate-level functions. Of 23 neurons tested for ipsilateral stimulation alone, 12 cells revealed monotonic rate-level functions and 11 neurons showed nonmonotonic rate-level functions.
Rate-level functions: SAM stimuli
As shown above, a substantial population of cells had monotonic rate-level functions for stimulation with pure tones. The situation changed dramatically when MSO neurons were tested with the use of SAM stimuli (N = 23). Only 1 of 9 neurons for ipsilateral stimulation, and 1 of 14 neurons for contralateral stimulation, had a monotonic rate-level function for SAM tones with a 200-Hz modulation rate. All others had nonmonotonic rate-level functions for SAM stimuli, independent of their rate-level function for pure tones. Figure 4A shows the distribution of rate-level functions for SAM stimuli. Figure 4B shows rate-level functions from the same unit as in Fig. 3B but now for SAM stimuli. Whereas this neuron exhibited monotonic rate-level functions for pure tones, it revealed nonmonotonic rate-level functions for SAM stimuli.
IIDs affect the response to SAM
The example in Fig. 5 shows how stimulating both ears separately with the same intensity could generate very different spike rates. The rate-level functions revealed that high intensities at either ear evoked a strong suppression, most likely due to inhibition (see DISCUSSION). To test whether the inhibition evoked from stimulating one ear can suppress spikes evoked from stimulating the other ear, we presented binaural stimuli and varied the IID. In 22 MSO neurons with EI/EI response characteristics, we tested the response to SAM stimuli at different IIDs. In 17 of these units, IIDs in either direction (favoring the ipsilateral or the contralateral ear) of
MSO neurons phase lock to SAM at low modulation rates
We tested the response of 53 MSO neurons to 100-ms SAM stimuli with the use of the units' BFs as the carrier frequency and varying the modulation rate between 20 and 1,000 Hz. We used 100% modulation depth only. The vast majority of neurons (45) exhibited a discharge that was phase locked to the envelope of low-frequency amplitude modulations with a minimum of 0.5 spikes/cycle, mostly with
MSO units exhibit low-pass filter characteristics for
SAM rate
The main feature of the SAM selectivity of the 45 neurons that exhibited phase-locked responses to SAM was a low-pass filter characteristic for the modulation rate. Figure 7 shows typical examples of responses from an EI/EI neuron (Fig. 7, left) and an I/EI neuron (Fig. 7, right) in response to binaural SAM stimulation with different modulation rates (equal intensities at both ears). The PSTHs, derived from 20 presentations, illustrate the precise phase locking of these neurons to modulation rates <200 Hz. To each SAM cycle the unit responded with one spike, sometimes with two spikes. At 200-Hz modulation rate the units still phase locked but with <1 spike/cycle. Only the ON response (2-3 spikes to the 1st modulation cycle) was not affected by the increasing modulation rate. In contrast, the response to the following cycles ("late response") decreased progressively and at >500 Hz no substantial response remained.
SAM filter: interaction of excitation and inhibition
So far we have presented data indicating that most MSO units have low-pass filter characteristics for SAM stimuli that are comparable with those shown for monaural MSO neurons in the mustached bat. To explain these filter characteristics in the mustached bat MSO, a scenario of monaural interaction of excitation and inhibition was proposed (Grothe 1994
Phase locking to SAM depends on the carrier frequency
In all recordings described so far, the BF of each neuron was used as the carrier frequency for the SAM stimuli. We tested the response to favorable SAM rates for different carrier frequencies in four neurons with clear phase locking. Three of these four units exhibited a strong dependency of the late response to SAM stimuli on the carrier frequency. Systematic variation of the carrier frequency revealed four major response types to different frequency ranges: 1) phasic responses (ON or OFF discharge patterns), 2) phase-locked responses, 3) sustained responses but without phase locking, and 4) complete inhibition of any response and spontaneous activity. Typically (as in Fig. 14), a sharp inhibitory area was at the high (2 of 3 cells) or the low (1 of 3 cells) end of the response area, creating a sharp border (in the neuron shown in Fig. 14 the discharge rate was reduced ~50% below spontaneous activity from 3.12 to 1.5 spikes/s). In all three units, best phase locking could be seen at frequencies close to the inhibitory border. Further away from the inhibitory sideband, the dominating responses were either sustained, but not phase-locked, or phasic responses. The fourth neuron responded with an OFF discharge to pure tones. This neuron phase locked to SAM stimuli over the entire frequency response range.
The main results from this study are as follows. 1) The majority of MSO cells we tested in the free-tailed bat receives excitatory and inhibitory projections from both ears, creating EI/EI response characteristics. Cells that varied from this pattern were usually lacking ipsilateral excitation and/or ipsilateral inhibition. 2) MSO neurons exhibit a clear low-pass filter characteristic for amplitude-modulated sounds. 3) Because of the binaural interaction in these units, the filter characteristics for SAM rate are IID dependent and thus should change with azimuthal position of a sound source.
General response patterns
In this study we recorded from neurons that basically covered the entire range of hearing of the free-tailed bat, ranging from 10 to 100 kHz (Schmidt et al. 1990 Inhibitory inputs to MSO neurons
We did not test for the presence of inhibitory projections to the free-tailed bat's MSO in a direct way, i.e., with the use of neuropharmacological techniques. However, there are four lines of evidence that inhibition is involved in processing acoustic stimuli in the free-tailed bat MSO, and all neurons tested showed at least one response feature indicating inhibitory influences.
Effect of IIDs
All of the binaural neurons we tested were sensitive to IIDs. The effect of changing IIDs was straightforward for neurons without ipsilateral excitatory inputs (I/EI neurons). I/EI neurons revealed IID functions comparable in many respects with those of LSO neurons in the same animal (compare Grothe and Park 1995 What is the function of a binaural MSO in bats?
According to the duplex theory of hearing (Raleigh 1907
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
, 1969; Langford 1984; Masterton and Diamond 1967; Spitzer and Semple 1995; Yin and Chan 1990). Moreover, it has been suggested that the MSO is not involved in processing other stimulus parameters commonly related to pattern recognition, such as coding of the temporal structure of a stimulus envelope (Langner 1992; Masterton 1992). Although alternative functions such as processing of temporally coded spectral information (Loeb et al. 1983, on the basis of a concept of Licklider 1951) have been suggested, experimental evidence has not been available. There is strong evidence that the MSO in mammals with low-frequency hearing and with sufficient interaural distance acts as an interaural time difference (ITD) detector by comparing the exact arrival times of the binaural inputs. Several studies showed that MSO neurons may function as coincidence detectors because they respond maximally if the phase-locked low-frequency inputs from the two ears arrive simultaneously ("in phase") and minimally if the inputs arrive asynchronously ("out of phase"; Goldberg and Brown 1969; Langford 1984; Spitzer and Semple 1995; Yin and Chan 1990).
; Poljak 1926; Schweizer 1981; Schweizer and Radtke 1980; Zook and Casseday 1982a,b). These results suggest that MSO neurons, at least in small mammals, serve functions additional to or other than coding ITDs. To date, there are two lines of evidence that MSO neurons are involved in the processing of other stimulus parameters instead or in addition to ITDs. First, it has been shown that the MSO in the mustached bat is particularly well suited for filtering amplitude modulations (Grothe 1994). However, the MSO in this bat is unique (as far as we know) in that it is a functionally monaural nucleus lacking significant ipsilateral inputs (Covey et al. 1991; Grothe et al. 1992). Second, intracellular recordings from a gerbil brain slice preparation revealed that MSO units are not only sensitive to the timing of the bilateral inputs, but also exhibit filter characteristics for the repetition rate of pulse trains (Grothe and Sanes 1994). However, it has not yet been tested in vivo whether these filter characteristics are functional in a binaural MSO. These results raise three questions.
in contrast to the mustached bata "common," binaural MSO (Grothe et al. 1994). The data illustrate how the complex interaction of excitation and inhibition that occurs in the MSO influences the response to pure tones and to amplitude-modulated tones. We particularly used amplitude modulations because comparable data exist for the mustached bat (Grothe 1994). Also, amplitude modulation rate is an important feature of complex sounds, including human speech. In a subsequent paper we will address the question of how MSO neurons in the free-tailed bat respond to ITDs, relating our data to the common theories of coincidence detection and discussing the question of the evolution of ITD coding in mammals.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
was used to define the position of the skull and the position of the recording electrode.
. The electrodes were advanced with a piezoelectric drive (Burleigh) from outside the recording chamber. Spikes from single units were fed via a recording amplifier, a band-pass filter (0.3-5 kHz), and a window discriminator into a computer. The software used for controlling stimulus presentation and recording was programmed by M. Baumann and S. Kieslich.
) on the basis of peristimulus time histograms (PSTHs) with 64 bins per SAM cycle. ON responses (5 ms beginning with the 1st correlated spike) or the discharge to the first cycle (if 1 cycle was >5 ms) were excluded before the MTFs were calculated. Because of the limited time we were able to hold the cells in the awake animals and for reasons of comparison with the SAM data from the mustached bat MSO (Grothe 1994), we only focused on the upper cutoffs of the MTFs and ignored lower cutoffs that were visible in some neurons at <40-Hz modulation frequency. In some neurons the responses to different carrier frequencies for SAM and pure tones were tested to determine inhibitory sidebands, with the use of a suppression of spontaneous activity as criterion (20 dB above threshold at BF). In some units we measured relative delay and relative timing of inhibitory inputs with the use of brief (2 ms) downward frequency-modulated (FM) sweeps ranging from 5 kHz above a neuron's BF down to 5 kHz below BF. These sweeps cause the shortest response we could elicit (only 1, sometimes 2 spikes) and presumably also the shortest inhibitory impact. FM sweeps were presented at a variety of electronically induced ITDs. Each test signal was presented either 10 or 20 times.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
) by injections of HRP and fluorescent beads into the central nucleus of the inferior colliculus, the medial nucleus of the trapezoid body (MNTB), and different subdivisions of the cochlear nucleus. Our HRP injections into the MSO confirm these earlier results that exhibited projections to the MSO originating in both anteroventral cochlear nuclei (AVCNs) and in the ipsilateral MNTB. They also confirmed the bilateral projections to the central nucleus of the inferior colliculus. Additionally, the HRP injections into the MSO revealed a fourth MSO input from the ipsilateral lateral nucleus of the trapezoid body (LNTB) to the MSO. Figure 1 presents the connectional pattern of the free-tailed bat MSO receiving two excitatory inputs from the AVCNs and two putative inhibitory (glycinergic) inputs from the ipsilateral LNTB (see DISCUSSION), providing ipsilateral induced inhibition, and the ipsilateral MNTB, providing contralateral induced inhibition. These results are consistent with the MSO projections described for cats, mice, gerbils, and several other bats (Cant 1991; Cant and Hyson 1992; Kuwabara and Zook 1992).
View larger version (11K):
[in a new window]
FIG. 1.
Projections to the medial superior olive (MSO) in the Mexican free-tailed bat: the MSO receives direct excitatory (+) projections from both anteroventral cochlear nuclei (AVCNs). Additionally, it receives projections from the ipsilateral lateral nucleus of the trapezoid body (LNTB) and medial nucleus of the trapezoid body (MNTB). LNTB neurons are excited by projections from the ipsilateral (ipsi), MNTB neurons by projections from the contralateral (contra) AVCN. Both nuclei are glycinergic (for review, see Cant 1991 ), and thus most likely provide bilateral inhibitory (
) inputs to the MSO.
) with the fundamental at ~25 kHz (Simmons et al. 1978). The distribution of BFs is shown in Fig. 2A.
View larger version (28K):
[in a new window]
FIG. 2.
A: best frequencies (BFs) of MSO neurons varied from 13 up to 81 kHz. The largest proportion of cells had BFs in the range of 20-40 kHz, corresponding to the echolocation call of the free-tailed bat. B: majority of MSO cells receives all 4 inputs described in Fig. 1 (EI/EI). However, 1/3 of the neurons exhibited no evidence for ipsilateral excitation (I/EI); 12% additionally lacked ipsilateral inhibition (O/EI). Five percent failed to show effects of contralateral stimulation (EI/O). C: >60% of MSO neurons showed phasic discharge patterns in response to stimulation with pure tones. About 1/3 showed a sustained discharge throughout the duration of the pure tone.
View larger version (31K):
[in a new window]
FIG. 3.
Rate-level functions of MSO neurons in response to pure tones. One third of neurons had monotonic rate-level functions for both ipsilateral and contralateral stimulation (m/m in A), as shown in the example in B. The 2nd third had a monotonic rate-level function for stimulation of 1 ear but a nonmonotonic function for stimulation of the other ear (m/n in A). The last third had nonmonotonic rate-level functions for stimulation at either ear (n/n in A).
View larger version (25K):
[in a new window]
FIG. 4.
Rate-level functions of MSO neurons in response to sinusoidally amplitude-modulated (SAM) tones. For this type of stimulation the vast majority showed nonmonotonic rate-level functions for stimulation at either ear alone (n/n). Only a few neurons showed monotonic rate-level functions for both ears (m/m) or for 1 ear (m/n). B: rate-level functions in response to SAM stimuli for the same neuron shown in Fig. 3B. In contrast to pure tone stimulation, SAM stimuli evoked nonmonotonic functions for either ear.
View larger version (28K):
[in a new window]
FIG. 5.
Example for the mismatch of rate-level functions between ipsilateral and contralateral stimulation with SAM stimuli (shown for 200-Hz SAM). Top: rate-level functions for ipsilateral, contralateral, and binaural [0 interaural intensity difference (IID)] stimulation. Bottom: corresponding peristimulus time histograms (PSTHs). Note that the response to binaural stimulation cannot be predicted from the 2 monaural functions.
10 dB resulted in a decrease of spike rate of 25% (up to 90%). In all cases the response dropped below that to monaural stimulation at the side with the fixed intensity, as shown for the EI/EI neuron in Fig. 6. Thus a large majority of MSO neurons exhibited a clear sensitivity for IIDs, probablyat least partlyas a consequence of the different strength of inhibition that was also apparent in the rate-level functions.
View larger version (19K):
[in a new window]
FIG. 6.
Normalized rate-level functions for monaural stimulation and IID functions from a typical EI/EI neuron. IIDs were assessed by keeping contralateral intensity fixed (50 and 40 dB SPL, respectively) and increasing ipsilateral intensity. First, when the intensity at the contralateral ear was still higher, raising ipsilateral intensity caused an increase in spike rate. However, by further increasing ipsilateral sound pressure, and thereby favoring the ipsilateral ear, the spike rate dropped to <60% of that obtained for monaural stimulation of the contralateral ear. This was true for pure tones and SAM (not shown).
1 spikes/cycle for certain modulation rates (Fig. 7, for example). All these units exhibited a selectivity for the modulation rate that will be discussed in detail below. Only eight units showed no phase locking at all. These units showed an ON response followed by a weak sustained response. In six of these eight neurons the response pattern as well as the number of spikes elicited by the SAM stimuli did not change when the modulation rate was varied. Thus these neurons were unselective for the modulation rate. In two of these units the response decreased for higher modulation rates.
View larger version (34K):
[in a new window]
FIG. 7.
Response of a typical EI/EI neuron (left) and a typical I/EI neuron (right) to different modulation rates for SAM. Representative PSTHs to binaurally presented 100-ms SAM stimuli are shown. The modulation rate was varied between 75 Hz (top) and 1,000 Hz (bottom). The envelope of the stimuli is given below each PSTH. The carrier frequency was at the neurons' BFs (21 and 48 kHz). Note the principal similarity between the 2 neurons.
) for the two neurons from Fig. 7 are shown in Fig. 8. Both neurons exhibited clear low-pass filter characteristics that were typical for the vast majority of MSO neurons. In contrast, not a single neuron we recorded in LNTB, MNTB, and lateral superior olive (LSO) exhibited such low-pass filter functions (data not shown).
View larger version (23K):
[in a new window]
FIG. 8.
Normalized modulation transfer functions (MTFs) of the 2 neurons shown in Fig. 7 for spikes per SAM cycle ( ) and vector strength ( ). Maximum spikes per cycle were 0.95 (A) and 2.15 (B), respectively. The neurons exhibited low-pass filter characteristics for both measurements.
).
View larger version (23K):
[in a new window]
FIG. 9.
Top: distribution of upper 50% cutoffs in the MTFs (spikes/cycle) for ipsilateral, contralateral, and binaural stimulation. For all 3 conditions, the majority of cells had cutoffs at <200-Hz SAM. The plot shows similar distributions for all 3 conditions with no obvious dependency on BF. Bottom: despite the similarity in the distribution of cutoffs, there is a considerable mismatch for the ipsilateral and contralateral cutoffs in individual neurons. Only 1 of 17 neurons tested for ipsilateral and contralateral cutoffs matched. Dotted lines: range of ±25% of the contralateral cutoff.
View larger version (32K):
[in a new window]
FIG. 10.
Normalized MTFs for spikes per SAM cycle ( ) for ipsilateral, contralateral, and binaural stimulation (0 IID) for 2 EI/EI neurons. The neuron shown in A had similar cutoffs for all 3 conditions. In contrast, the neuron shown in B showed similar MTFs for binaural and contralateral stimulation, but a very different function for ipsilateral stimulation. Shaded areas: range of 50% cutoffs.
30 dB IID (favoring the ipsilateral ear) ~ 65-Hz modulation rate. Thus in this neuron there was a systematic decrease of MTF cutoff with increasing ipsilateral intensity. However, the five neurons tested did not respond in a uniform way. One other unit shifted like the neuron shown in Fig. 11, one neuron exhibited MTF shifts in the opposite direction, one responded in a nonlinear way, and one did not show any shifts in the MTF cutoff.
View larger version (25K):
[in a new window]
FIG. 11.
Normalized MTFs for spikes per SAM cycle ( ) for binaural stimulation at 3 different IIDs (+30, 0, 30; contralateral ear held constant). Positive IIDs: higher intensities at the ipsilateral ear. For more positive IIDs, the MTF systematically shifted to higher cutoffs. Shaded area: range of 50% cutoffs.
) that implies a temporal mismatch of excitation and inhibition that caused an overlap of the inhibition elicited by one SAM cycle with the excitation evoked by the following cycle (Fig. 12).
View larger version (30K):
[in a new window]
FIG. 12.
SAM filter mechanism as proposed for monaural neurons in the mustached bat's MSO. Temporal interplay of excitation and delayed inhibition will allow a phasic response to each SAM cycle, as long as the SAM rate is low (top). At higher amplitude modulation rates, the delayed inhibition evoked by 1 cycle overlaps with the excitation evoked by the next cycle, thereby suppressing the response to all but the 1st amplitude modulation cycle. Top line in each panel: hypothetical output of an MSO cell (schematic PSTH). Flat bars: excitatory [excitatory postsynaptic potential (EPSP)] and inhibitory [inhibitory postsynaptic potential (IPSP)] inputs from the contralateral ear. Bottom of each panel: envelope of the SAM stimulus and assumed threshold for eliciting excitatory and inhibitory inputs to the MSO neuron (adapted from Grothe 1994 ).
View larger version (28K):
[in a new window]
FIG. 13.
Response of an EI/EI neuron to short (2 ms) FM sweeps as a function of interaural time difference (ITD). The neuron responded with only 1-2 spikes per sweep to contralateral or ipsilateral stimulation. A: relative spike rate evoked by ipsilateral stimulation for a variety of electronically induced ITDs. B: relative spike rate evoked by contralateral stimulation for the same cell. Gray area: range of ITDs that caused a reduction of 50% (or more) in the normalized spike rate. See text for a more detailed explanation. C: correlation of duration of contralaterally driven inhibition derived from graphs like that in A and the 50% cutoff for the cell's contralateral MTF. Asterisk: data point corresponding to the cell shown in A.
View larger version (16K):
[in a new window]
FIG. 14.
PSTHs of an EI/EI neuron in response to 5 repetitions of a 200-Hz SAM stimulus (binaural stimulation, 0 IID) at different carrier frequencies. The response area was limited at both high and low frequencies by inhibitory sidebands, resulting in a discharge at ~50% below spontaneous activity (1.5 and 3.12 spikes/s, not shown). The neuron showed a weak sustained response at ~20 kHz, an ON-dominated response at 22 and 24 kHz, and phase locking (vector strength > 0.3) to the envelope only for carrier frequencies at the high portion of the frequency response area close to the high-frequency inhibitory sideband (26 kHz).
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
). There was a bias toward BFs in the range of 20-30 kHz, which corresponds to the fundamental of the echolocation call of these animals (Simmons et al. 1978). This result seems to be consistent with the BF representation in the MSO of the mustached bat, which is also biased to the frequencies dominating the biosonar signal (Grothe 1994). In this respect, the MSO in bats appears to differ from the MSO in other mammals, where mainly low frequencies are represented and only a small proportion of cells have BFs >4 kHz (Guinan et al. 1972a,b). These species differences may suggest that the MSO in bats (and maybe in other small mammals) is serving a function other than ITD detection of low-frequency sounds, the major function suggested for the MSO in larger mammals.
; Yin and Chan 1990) and the gerbil (Grothe and Sanes 1993, 1994), and ~11% in the dog (Goldberg and Brown 1969), but ~45% in the rat (Indbody and Feng 1981). In terms of response types, the MSO in the free-tailed bat appears to represent a mixed nucleus positioned somewhere between the mustached bat and the cat or gerbil MSO.
). These neurons are known to exhibit sustained discharge for the duration of a stimulus. In contrast, the majority of MSO neurons responded with a phasic response pattern. Thus phasic response patterns in the MSO are most likely explained by additional inhibitory inputs. In support of this idea, it has been shown that glycinergic inhibition shapes phasic ON or OFF response patterns in the mustached bat's MSO. These discharge patterns could be reversibly altered to primary-like discharge patterns when glycinergic inhibition was blocked by iontophoretically applying strychnine, thus revealing the pattern of the excitatory input (Grothe 1994).
). Support for this idea comes from gerbil brain slice experiments. Increasing electric current by shocking fiber bundles projecting to the SOC caused glycinergic inhibition to block action potentials (Grothe and Sanes 1993).
). Intracellular in vitro recordings revealed a limited ability of gerbil MSO neurons to follow trains of electric stimulation applied to the fibers entering the SOC. Most neurons exhibited, again, cutoffs between 100- and 300-Hz repetition rate that could be shifted to higher levels by addition of strychnine into the artificial cerebrospinal fluid (Grothe and Sanes 1994). In the present study, the results from the use of FM sweeps are consistent with the idea that inhibition shapes SAM filter characteristics.
; Spangler et al. 1985; rodents: Kuwabara and Zook 1992; bats: Casseday et al. 1988; Covey et al. 1991). Additionally for the cat, several rodents, and several bats, a projection from the LNTB has been found (Cant 1991; Cant and Hyson 1992; Kuwabara and Zook 1992). These inputs are also found in the free-tailed bat (Grothe et al. 1994; this study). Both LNTB and MNTB neurons are glycinergic, and the MNTB is considered as one of the regions with the highest concentration of glycine in the entire mammalian brain (Wenthold and Hunter 1990).
) and gerbils (Spitzer and Semple 1995) show little evidence of a significant role for inhibition in the MSO. But the available data are more or less limited to those from recordings in which pure tones or ongoing noise with ITDs were used. It might be that it is more difficult to assay inhibitory effects when pure tone stimuli are used. This is apparent from the data presented here (more inhibitory effects were observed for SAM stimuli compared with pure tones) and has also been suggested recently by Brughera et al. (1996). A final solution to the questions of the role of inhibition in the MSO requires direct evidence by neuropharmacological experiments in vivo.
): increasing SPL at the ipsilateral ear progressively inhibits the neuron's response. IID sensitivity in LSO neurons is due to a subtraction of excitation from one ear and inhibition from the other ear (for review, see Irvine 1992). The difference is that whereas the inhibitory ear for LSO neurons is the contralateral ear, in MSO I/EI neurons there is excitation from the contralateral ear and inhibition from the ipsilateral ear. Thus these neurons prefer contralateral stimuli (opposite to LSO neurons), as do some MSO neurons in the dog (Goldberg and Brown 1969) and a substantial number of neurons in the rat MSO (Indbody and Feng 1981).
; Stevens and Newman 1934), ITDs are the main cue for localizing low-frequency sounds. Free-tailed bats only hear high-frequency sounds and have an interaural distance most likely too small to generate resolvable ITDs. The question of how MSO neurons in the free-tailed bat respond to ITDs will be addressed in a subsequent paper. However, the data presented here indicate that MSO neurons in the free-tailed bat are sensitive to other features than ITDs. They act as filters for the temporal structure of a stimulus, i.e., for amplitude modulations. Amplitude modulations are a common features in abiotic and biotic sounds (including communication signals and in the echoes of biosonar signals) and are thought to play an important role in pattern recognition(Langner 1992). In the context of hunting fluttering insects, bats might particularly benefit from analyzing the periodic components of a sound at an early stage of auditory processing (Casseday and Covey 1996). However, in contrast to the mustached bat's MSO, the output of the free-tailed bat's MSO appears to be an integrated one defined by stimulus pattern as well as stimulus position.
). Therefore the results presented here might well represent common properties of the MSO in all mammals, and thus ITD coding might not be the exclusive function of the MSO in general.
| |
ACKNOWLEDGEMENTS |
|---|
We thank C. Schulte, S. Kieslich, and H. König for technical support. We thank M. Reed and N. Suga for detailed discussion of our data. We particularly thank G. Neuweiler, D. R. F. Irvine, G. D. Pollak, and J. J. Blum for the special interest that strongly influenced our experiments.
This work was supported by SFB 204 and the Alexander-von-Humboldt Foundation.
| |
FOOTNOTES |
|---|
Address for reprint requests: B. Grothe, Zoologisches Institut, Luisenstr. 14, D-80333 Munich, Germany.
Received 16 April 1996; accepted in final form 20 November 1996.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
PerceptionCognition. Proceedings of the 18th Göttingen Conference,
edited by
N. Elsner,
and G. Roth
. Stuttgart, Germany: Thieme, 1990, p. 146This article has been cited by other articles:
|
|
 |
|
  A. Kutscher and E. Covey Functional Role of GABAergic and Glycinergic Inhibition in the Intermediate Nucleus of the Lateral Lemniscus of the Big Brown Bat J Neurophysiol, June 1, 2009; 101(6): 3135 - 3146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zhang and J. B. Kelly Responses of Neurons in the Rat's Ventral Nucleus of the Lateral Lemniscus to Amplitude-Modulated Tones J Neurophysiol, December 1, 2006; 96(6): 2905 - 2914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Dreyer and B. Delgutte Phase Locking of Auditory-Nerve Fibers to the Envelopes of High-Frequency Sounds: Implications for Sound Localization J Neurophysiol, November 1, 2006; 96(5): 2327 - 2341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Hurley and G. D. Pollak Serotonin Shifts First-Spike Latencies of Inferior Colliculus Neurons J. Neurosci., August 24, 2005; 25(34): 7876 - 7886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Park, A. Klug, M. Holinstat, and B. Grothe Interaural Level Difference Processing in the Lateral Superior Olive and the Inferior Colliculus J Neurophysiol, July 1, 2004; 92(1): 289 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Svirskis, V. Kotak, D. H. Sanes, and J. Rinzel Sodium Along With Low-Threshold Potassium Currents Enhance Coincidence Detection of Subthreshold Noisy Signals in MSO Neurons J Neurophysiol, June 1, 2004; 91(6): 2465 - 2473. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zhang and J. B. Kelly Glutamatergic and GABAergic Regulation of Neural Responses in Inferior Colliculus to Amplitude-Modulated Sounds J Neurophysiol, July 1, 2003; 90(1): 477 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Walton, H. Simon, and R. D. Frisina Age-Related Alterations in the Neural Coding of Envelope Periodicities J Neurophysiol, August 1, 2002; 88(2): 565 - 578. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Behrend, A. Brand, C. Kapfer, and B. Grothe Auditory Response Properties in the Superior Paraolivary Nucleus of the Gerbil J Neurophysiol, June 1, 2002; 87(6): 2915 - 2928. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Grothe, E. Covey, and J. H. Casseday Medial Superior Olive of the Big Brown Bat: Neuronal Responses to Pure Tones, Amplitude Modulations, and Pulse Trains J Neurophysiol, November 1, 2001; 86(5): 2219 - 2230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Hurley and G. D. Pollak Serotonin Effects on Frequency Tuning of Inferior Colliculus Neurons J Neurophysiol, February 1, 2001; 85(2): 828 - 842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Brand, R. Urban, and B. Grothe Duration Tuning in the Mouse Auditory Midbrain J Neurophysiol, October 1, 2000; 84(4): 1790 - 1799. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Casseday, D. Ehrlich, and E. Covey Neural Measurement of Sound Duration: Control by Excitatory-Inhibitory Interactions in the Inferior Colliculus J Neurophysiol, September 1, 2000; 84(3): 1475 - 1487. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Koch and B. Grothe Interdependence of Spatial and Temporal Coding in the Auditory Midbrain J Neurophysiol, April 1, 2000; 83(4): 2300 - 2314. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Hurley and G. D. Pollak Serotonin Differentially Modulates Responses to Tones and Frequency-Modulated Sweeps in the Inferior Colliculus J. Neurosci., September 15, 1999; 19(18): 8071 - 8082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Klug, E. E. Bauer, and G. D. Pollak Multiple Components of Ipsilaterally Evoked Inhibition in the Inferior Colliculus J Neurophysiol, August 1, 1999; 82(2): 593 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kuwada and R. Batra Coding of Sound Envelopes by Inhibitory Rebound in Neurons of the Superior Olivary Complex in the Unanesthetized Rabbit J. Neurosci., March 15, 1999; 19(6): 2273 - 2287. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Oswald, A. Klug, and T. J. Park Interaural Intensity Difference Processing in Auditory Midbrain Neurons: Effects of a Transient Early Inhibitory Input J. Neurosci., February 1, 1999; 19(3): 1149 - 1163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Grothe and T. J. Park Sensitivity to Interaural Time Differences in the Medial Superior Olive of a Small Mammal, the Mexican Free-Tailed Bat J. Neurosci., August 15, 1998; 18(16): 6608 - 6622. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Koch and B. Grothe GABAergic and Glycinergic Inhibition Sharpens Tuning for Frequency Modulations in the Inferior Colliculus of the Big Brown Bat J Neurophysiol, July 1, 1998; 80(1): 71 - 82. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
