Onset Neurones in the Anteroventral Cochlear Nucleus Project to the Dorsal Cochlear Nucleus
- 618 Downloads
Considerable circumstantial evidence suggests that cells in the ventral cochlear nucleus, that respond predominantly to the onset of pure tone bursts, have a stellate morphology and project, among other places, to the dorsal cochlear nucleus. The characteristics of such cells make them leading candidates for providing the so-called “wideband inhibitory input” which is an essential part of the processing machinery of the dorsal cochlear nucleus. Here we use juxtacellular labeling with biocytin to demonstrate directly that large stellate cells, with onset responses, terminate profusely in the dorsal cochlear nucleus. They also provide widespread local innervation of the anteroventral cochlear nucleus and a small innervation of the posteroventral cochlear nucleus. In addition, some onset cells project to the contralateral dorsal cochlear nucleus.
Keywordsstellate cells anteroventral cochlear nucleus dorsal cochlear nucleus wideband inhibitor onset responses
The cochlear nucleus consists of three functionally and anatomically separate divisions: the anteroventral cochlear nucleus (AVCN), the posteroventral cochlear nucleus (PVCN), and the dorsal cochlear nucleus (DCN) (Rose et al. 1959). Responses of auditory nerve fibers to simple tone burst stimuli are rather homogenous (Kiang et al. 1965), whereas cells in the cochlear nucleus provide diverse responses to simple tones (e.g., Young 1984). In the ventral division, many morphological cell types have been distinguished. Some have very distinctive morphologies, such as the granular, Golgi, and octopus cells and the globular and spherical bushy cells. However, others form part of a diverse class of stellate or multipolar cells which vary in size, from small to giant, with either smooth dendrites or varying densities of spines (Osen 1969; Brawer et al. 1974). Bushy cells provide more-or-less faithful transmission of auditory nerve firing patterns (Pfeiffer 1966) and project this information into pathways for comparison of the activity from the two ears to extract cues for localizing sounds. Stellate cells have more diverse physiological properties, producing a variety of onset or chopping-type responses (Young et al. 1988; Blackburn and Sachs 1989; Smith and Rhode 1989; Winter and Palmer 1990).
Two main classes of stellate cells in VCN have been described based on a number of different criteria. Some of the criteria are purely structural while others are functional, but there seems to be a good correspondence between them. The first class has variously been termed type I (sparse synapses formed on the soma) (Cant 1981), small planar or marginal cells (Doucet and Ryugo 1997), type T (axon projecting into the trapezoid body) (Oertel et al. 1990), glycine negative (Alibardi 1998, 2001), and cells with chopping responses (Smith and Rhode 1989). The second class has been termed type II (dense covering of synapses on the soma), large radiate cells, type D (dorsal projecting axons), glycine positive, and cells with onset responses (by the same authors). Chopping describes regular, sustained discharge characteristics determined by membrane biophysics rather than being related to the stimulus waveform. Onset cells, stimulated close to threshold, produce only a few spikes locked to the onset of the stimulus and have little if any sustained activity. Chopping responses are typical of small, smooth stellate cells and the difference between these cells and the large stellate cells with onset properties has been described elsewhere (e.g., Palmer et al. 2003). Three types of onset cells have been characterized in the guinea pig cochlear nucleus: these are the On-C (57% of the population), On-L (34%), and On-I (9%) (Winter and Palmer 1995). The On-C cells have more than one peak at the onset, the other two have a single peak at the onset, but, while the On-L and On-C cells can have sustained rates of >10 spikes/s, the On-I have sustained rates of <10 spikes/s.
MATERIALS AND METHODS
Experiments were carried out using male and female pigmented guinea pigs ranging from 350 to 600 g in accordance with the United Kingdom Animals (Scientific Procedures) Act 1986. Animals were anesthetized with urethane (1.3 g/kg IP, in 20% solution in 0.9% saline) and Hypnorm (0.2 ml IM, comprising fentanyl citrate 0.315 mg/ml and fluanisone 10 mg/ml). Atropine sulfate (0.06 mg/kg SC) was administered at the start of the experiment. Anesthesia was supplemented, on indication, with further doses of Hypnorm (0.2 ml IM).
A tracheotomy was performed, followed by bilateral exposure of the ear canal. The animal was mounted in a stereotaxic frame in which the ear bars were replaced with plastic speculae to allow visualization of the tympanic membrane and delivery of sound stimuli. A craniotomy was made which extended from approximately 2 mm behind the nuccal ridge to lambda and from approximately 2 mm to the right of the midline to approximately 6 mm to the left. The dura was reflected and the surface of the brain covered by a solution of 1.5% agar in 0.9% saline to prevent desiccation and to aid stability. The angle of the head was adjusted such that the surface of the skull in the rostrocaudal axis was displaced downward by 20° from horizontal toward the front of the animal. The animal’s temperature was maintained at 37°C throughout the experiment by means of a thermostatically controlled heating blanket. Respiratory rate and heart rate were monitored routinely.
Stimulation recording, and juxtacellular labeling
Experiments were carried out in a sound-attenuated booth. Stimuli were delivered through a sealed acoustic system comprising custom-modified Radio Shack 40-1377 tweeters (M. Ravicz, Eaton Peabody Laboratory, Boston, MA) coupled to damped 4 mm diameter probe tubes which fitted into the speculum. In every experiment, a probe tube microphone was used to calibrate the sound system, close to the tympanic membrane, using a Brüel and Kjaer 4134 microphone fitted with a calibrated 1 mm probe tube. The sound system response on each side was flat to within ±10 dB from 100 to 35,000 Hz.
Stimuli consisted of 50 ms bursts of wideband noise or pure tone (0.1–40 kHz) which were presented to the left ear (ipsilateral to the recording site) every 200 ms. All stimuli were generated by an array processor (Tucker–Davis Technologies, Alachua, FL, Model AP2) and output was at rates of at least 100 kHz. The maximum output level was set to approximately 100 dB SPL.
Single cell activity was initially recorded via tungsten-in-glass microelectrodes (Bullock et al. 1988) which were positioned using a piezoelectric motor (Inchworm IW-711-00, Burleigh Instruments, New York). The microelectrodes were angled at 45° and were placed stereotaxically. Once an area of onset responding cells had been identified, the tungsten electrode was withdrawn and replaced with an aluminosilicate glass capillary (1 mm outer diameter, Clarke Electromedical, Pangbourne, UK), pulled and broken back to give a tip impedance of 15–30 MΩ, and filled with 1.5% biocytin (Sigma, St. Louis, MO) in 0.5 M sodium chloride. The glass electrodes were not used for the initial mapping because their tips were easily damaged and their wider shafts tended to produce more tissue damage than the slimmer tungsten electrodes.
When a single cell with onset characteristics (Winter and Palmer 1995) was isolated, the cell’s characteristic frequency and minimum threshold for response were determined. The cell was characterized physiologically by determining its frequency response area, by constructing rate-level functions to best frequency tone and to noise stimuli, and by constructing poststimulus time histograms of responses evoked by best frequency tone and noise stimuli which were delivered at 20 and 50 dB above the characteristic frequency response threshold.
Following physiological characterization, the cell was labeled with biocytin according to the juxtacellular method of Pinault (1996). Briefly, biocytin was ejected from the recording pipette under physiological control using +3 to +11 nA square wave current pulses, of 200 ms duration (50% duty cycle), which were injected using the current injection circuit of the microelectrode amplifier. Adequate current injection, sufficiently close to the cell, caused action potentials to be evoked robustly during the depolarizing epochs. The current strength was titrated carefully to ensure that the cell remained firing throughout the labeling but was not damaged by overdriving; strong labeling occurred when current injection-associated firing was maintained for 4–15 min. We gained the impression that the duration of current injection, during which action potentials were elicited (with currents in the range +5 to +10 nA), was the most significant factor governing the distance through which a neurone might be traced subsequently.
Histology and anatomical reconstruction
Following up to 9 h of survival (during which time we undertook recordings from the contralateral inferior colliculus for a different study), the animal was given an overdose of sodium pentobarbitone and perfused transcardially with 500 ml of phosphate buffer (PB) pH 7.4 containing 4% paraformaldehyde and 0.5% glutaraldehyde. The brain was removed and stored in the same fixative overnight at 4°C.
The following day, the brain was embedded in a mixture of gelatin and egg albumin and serial 50 μm coronal sections were cut using a vibratome. The freely floating sections were washed twice in PB and incubated overnight at 4°C in PB containing 0.3% Triton X-100 and avidin–biotin peroxidase complex (ABC Elite, Vector Laboratories, Burlingame, CA). The sections were washed twice in PB before being incubated for 10 min with 0.05% diaminobenzidine (DAB), 0.005% hydrogen peroxide, 0.0015% nickel ammonium sulfate, and 0.0015% cobalt chloride in PB. The sections were washed twice and mounted on coated slides.
Three-dimensional reconstructions were undertaken using computer software (Neurolucida, Microbrightfield, Colchester, VT) connected to a microscope (Axioskop2, Carl Zeiss). DAB-stained material was traced using a 40× objective lens (NA 0.95) within the traced boundaries of the CN and the brain stem in each section. The perfusion and sectioning did not produce any shrinkage that we could measure, but the sections did become thinner after they were mounted and had dried onto the slides. We compensated for this by resetting the Z value of the reconstruction after drawing each section so that the coordinates for each new section started exactly 20 μm further on than the start of the previous section. When making the reconstructions, we introduced a shrinkage compensation factor of 2.5× to compensate for the fact that the sections were cut at 50 μm.
Twenty neurones in AVCN were characterized physiologically, labeled, and recovered histologically. Five of these neurones corresponded to the On-C type and two to the On-L type (Smith and Rhode 1986; Winter and Palmer 1995). This report describes only these onset cells, the others are described in Palmer et al. (2003). In common with the earlier studies, our onset neurones frequently exhibited higher response thresholds than those associated with other CN neurones (e.g., chop-S neurones). Considerably lower thresholds were frequently measured in non-onset neurones with similar best frequencies (BFs), which were encountered within a few hundred microns of our onset neurones. This fact suggested that the higher thresholds of the onset cells did not reflect cochlear pathology. Our onset cells had minimum thresholds that varied from 23 to 70 dB SPL and best frequencies varied from 0.98 kHz to 11.1 kHz.
Axon terminations in AVCN
The On-C cells possessed single axons that arose either from an axon hillock, which was contiguous with the soma, or from the proximal part of a large dendrite within approximately 15 μm of the soma. Figure 5 shows the pattern of projection, within VCN, of the neurone whose physiological response profile is shown in Figure 2. On-C axons (black) branched profusely, forming a local plexus of fine branches with boutons en passant and terminaux. Although some of the local ramification was present in the same volume as the dendritic arbor (red), the local axonal termination was more extensive and, in every case, innervated regions of the AVCN which were outside those containing the cell’s dendrites (Figs. 5, 6, 7). Despite the length of the axon, for all four of the reconstructed On-C cells, local terminations appeared to be arranged largely within a volume of tissue representing BFs that were within 1 octave of the injected cell’s BF. They appeared to be centered on the same isofrequency slab as the dendrites (Fig. 9A,B).
There was some variability in the pattern of local termination. For example, the local axonal ramification of the neurone in Figures 5 and 9A (No. 282, BF-6.3 kHz) is arranged in a diagonally oriented cylinder whose axis is displaced medially and rostrally from the soma. In contrast, the local innervation of the other On-C cells is arranged more diffusely (Figs. 6, 7, 9B).
Axon terminations in PVCN
Three of the On-C cells gave rise to varicose axonal plexuses in the most caudal part of PVCN. However, the route by which the axon reached the PVCN differed in each of our reconstructed neurones. In the neurone in Figure 5 (No. 282, BF-6.3 kHz), a branch of the ipsilateral DCN innervation turned ventrorostrally to arborize in the caudoventral part of PVCN (see arrows) beneath the ventral boundary of rostral DCN. In contrast, in the neurone shown in Figure 6 (No. 212, BF-5.9 kHz), an axonal branch arose from the local arbor and projected caudoventrally, giving rise to a varicose axonal plexus in the same caudoventral region of PVCN (indicated by arrows). Interestingly, a further axonal branch arose from this PVCN plexus and extended back rostrodorsally, running along the lateral margin of the CN to terminate within the volume of the neurone’s local axonal arbor. The neurone in Figure 7 (No. 249, BF-1.2 kHz) innervated the same region of PVCN (see arrows) via terminating branches of the axon en passage to the ipsilateral DCN. Only the PVCN innervation of unit 212 (Fig. 6) gave rise to the recurrent fiber that extended rostrally to re-enter AVCN.
In AVCN, PVCN, and DCN, when there was sufficient background labeling, it was possible to see the terminal branches of an On-C axon encircling pale somata and giving rise to en passant and terminal varicosities in close proximity to them (Fig. 8B).
Axon terminations in DCN
The contralaterally projecting axonal branches were traced from section to section. There was no evidence of axonal branching either before the neurone became untraceable in the heavily myelinated fiber tract beneath the fourth ventricle or, in the case of unit 282, before the fiber innervated the contralateral CN.
Whereas we were unable to distinguish the soma and dendrites of the On-L cells from those of our On-C cells, there were differences in their axonal projection. The somata were ovoid and gave rise to gently tapered, aspinous dendrites that were finely branched at their terminations. Unit 286 gave rise to four primary dendrites (total length-6970 μm); unit 284 gave rise to six primary dendrites (total length-7096 μm).
Like the On-C cells, our On-L cells gave rise to local, extensively branched, varicose axonal processes which innervated a volume which was more extensive than that occupied by the cells’ dendrites (Fig. 9C,D). The local axonal tree was flattened orthogonal to the frequency axis of the nucleus. This flattening meant that the axon was largely located among cells with BFs that were less than one octave above that of the injected cell.
Three-dimensional reconstruction necessitated tracing every axonal branch using the 40X lens. In neither of our fully reconstructed On-L cells was there any evidence of axonal projection toward the midline as shown by the five On-C cells. The On-L axons were darkly stained and we are confident that we would have seen a midline branch had one been present.
In this study we used the juxtacellular labeling technique to clarify an important component of the internal circuitry of the mammalian cochlear nucleus. We have shown that the On-C/L cells are probably the same as the wideband inhibitor or radiate neurone or D-multipolar that have been described by others. Although these cells have been studied by combined anatomical/physiological methods before, we have gone beyond previous studies. Thus, we have provided a much more detailed description of the axonal distribution, both within the ipsilateral CN and also to the contralateral CN. This anatomy has been combined with both a description in vivo of the physiological response profile and attempts to relate the morphology to a detailed map of the isofrequency sheets in a way not tried before. Different aspects of the study will be discussed in turn.
Local axonal projections
Both On-C and On-L cells gave rise to substantial local axonal arbors in PVCN and AVCN. The axons into PVCN took a number of different routes, but it is presumably the locus of termination which is important and not the trajectory. For both cell types we observed terminal axonal branches forming rings or clusters of en passant and terminal varicosities that encircled somata which were not labeled specifically (Fig. 8B). This observation suggests that the On-C and On-L axons may form axosomatic synaptic contacts of the kind commonly associated with inhibitory synaptic input. This suggestion is supported by both the imrnunohistochemical evidence for glycinergic transmission in VCN large stellate cells which innervate DCN (Doucet et al. 1999; Alibardi 2001) and the presence of glycinergic terminals on the somata of fusiform cells in the DCN (Osen et al. 1990). There is also extensive physiological evidence for strychnine-sensitive inhibitory input to type II cells in DCN (Davis and Young 2000).
The functional role of this local inhibition can be inferred only indirectly, but there has been a proposal that broadly tuned On-C cells inhibit narrowly tuned chopper cells (Smith and Rhode 1989). There is also evidence from in vitro work that large D-type stellate cells inhibit the smaller T stellate cells. This evidence implies that widely tuned On-C cells provide choppers with inhibitory sidebands (Ferragamo et al. 1998). Pressnitzer et al. (2001) have suggested that some cell types in VCN, especially transient chopper (Chop-t)-type cells, may be involved in the psychophysical phenomenon of comodulation masking release. They have constructed a model circuit that depends upon wideband inhibitory input, which is similar to that in DCN. Both the On-C and On-L cell types may provide the basis of such an inhibitory circuit through their local projections into AVCN and PVCN. Only the On-L cells in our study innervated rostral AVCN (probably because of their position within the nucleus), but there is no evidence to suggest that this connection is functionally different from that to PVCN (Osen 1969; Hackney and Pick 1986).
Innervation of the ipsilateral DCN
Every labeled onset cell gave rise to projections into the ipsilateral DCN. Our sample of neurones suggests that there may be a tonotopic arrangement of the DCN termination of onset neurones. The DCN termination of the lowest-frequency On-C cell (unit 249, BF = 1.2 kHz) was found predominantly in the most lateral portion of the DCN, while midfrequency On-C cells gave rise to DCN terminations which were predominantly in the central portion of the nucleus. Previous studies have shown that cells with high BFs are located most medially in DCN, and low-BF cells are located laterally and ventrally in DCN (e.g., guinea pig: Stabler 1991; cat: Spirou et al. 1993). On-C cells therefore have a projection to the DCN that, while widespread, is nevertheless loosely consistent with the tonotopicity of the DCN principal cells.
The two reconstructed On-L cells show a similar pattern of DCN termination to that seen in On-C cells, although the terminal branches of the On-L neurones appear to occupy a more constrained laminar profile. Studies of the origin of the DCN projection to the inferior colliculus have identified the principal output neurones, which are large fusiform pyramidal and giant cells, as being located predominantly in layers II, III, and IV (Ryugo and Willard 1985; Spirou et al. 1993); thus these projection cells are likely to receive input from both On-C and On-L cells.
The innervation of the DCN by the VCN has been studied using small, focal injections of retrograde tracers into the DCN (rat: Doucet and Ryugo 1997; cat: Ostapoff et al. 1999). These studies have revealed both a restricted band of labeling in the VCN that corresponds tonotopically with the frequency sensitivity of the DCN injection site and labeling of large stellate somata outside the labeled band. Retrogradely labeled VCN somata included stellate cells but not bushy cells.
Within the band of label in the VCN, stellate cells are relatively small and are oriented in the plane of the tonotopic lamina (Doucet and Ryugo 1997; Ostapoff et al. 1999), while stellate cells, outside the band of label (sometimes very far removed from the labeled band), are large and have been described as “giant” (Ostapoff et al. 1999) or “radiate” (Doucet and Ryugo 1997). Not only are the stellate cells that give onset-type responses large and have morphologies consistent with those of “giant/radiate” cells, but the widespread innervation of the DCN by our onset cells means that onset cells from divergent frequency regions would likely have been labeled by focal DCN injections. Results from previous studies in this laboratory have shown consistency between the position of onset cells with different BFs and the tonotopic arrangement of the nucleus (see Fig. 4 and Jiang et al. 1996).
The functional significance of onset cell innervation of the DCN
Important functions that have been attributed to the DCN include extraction and processing of information pertaining to sound source location (Davis and Young 2000) and speech formants (Rhode and Greenberg 1994a, b). Principal output neurones of the DCN, termed type IV or fusiform neurones, have been found to be extremely sensitive to the frequency position of sharp spectral notches (for review see Young and Davis 2002). The ability of type IV neurones to signal these important stimulus characteristics depends upon neuronal circuitry intrinsic to the cochlear nucleus complex.
A simplified schematic of this cochlear nucleus circuitry is shown in Figure 1. The cells designated as type IV and type II are classified according to the extent of the inhibitory inputs revealed within their frequency–intensity response areas (Evans and Nelson 1973; Shofner and Young 1985). Example response areas of actual neurones are also shown for the types II and IV and auditory nerve (these are for illustration only: the best frequencies in these examples are all relatively low but unrelated). The “V”-shaped response area of the auditory nerve fiber is transformed in the type II neurone by strong inhibition that overlaps at high sound levels to generate nonmonotonic responses as a function of sound level. This strong inhibition renders type II neurones insensitive to broadband signals (Young and Brownell 1976; Young and Voigt 1982; Shofner and Young 1985). Type II responses are associated with glycinergic “vertical cells” located just beneath the fusiform cell layer in DCN (Rhode 1999).
Type IV neurones have a response area that is even more restricted, often consisting of only an island of excitation severely constrained by inhibitory inputs. Hence, type IV neurones are strongly nonmonotonic as a function of tone sound level.
In addition to their type II innervation, which primarily influences behavior near the BF, principal cells are also inhibited at frequencies far from their BF (Nelken and Young 1994) and are thus subject to wideband inhibition, which is antagonized by strychnine (Davis and Young 2000). This wideband inhibition has been postulated to modulate the responses of principal neurones both directly, through glycinergic innervation of these cells, and indirectly, through glycinergic innervation of type II DCN interneurones.
A wealth of circumstantial evidence suggests that onset responding stellate cells in the ventral cochlear nucleus may give rise to the wideband inhibition of DCN principal cells. These onset cells integrate inputs from auditory nerve fibers with a broad range of BFs (Jiang et al. 1996; Palmer et al. 1996), acting as coincidence detectors with a relatively narrow coincidence window (Palmer and Winter 1996); they respond better to broadband signals than to pure tones; and they exhibit relatively wide dynamic ranges (Smith and Rhode 1986; Winter and Palmer 1995). Physiologically characterized, VCN onset cells have been found to exhibit large multipolar morphology (Smith and Rhode 1989), and VCN cells with this morphology have been found to be glycinergic (Doucet et al. 1999) and are thought to provide inhibition of other neurones in the cochlear nucleus such as the T stellate cells (Ferragamo et al. 1998).
Projection to the contralateral DCN
One of our On-C axons could be traced throughout its entire trajectory to the contralateral DCN, while three of the other On-C cells gave rise to an axon which followed the same course before fading and becoming untraceable in the white matter below the fourth ventricle. A previous study in rat (Friauf and Ostwald 1988) indicated that onset responding cells (“presumably stellate”) in the ventral cochlear nucleus projected to various superior olive and periolivary sites as well as to the contralateral lateral lemniscus. We found no evidence of such ramifications, even in the axon that was traced to the contralateral cochlear nucleus. Indeed, the projection of the onset cells of Friauf and Ostwald is reminiscent of those of octopus cells described by Adams (1979). Octopus cells have been reported to show onset-type responses (Godfrey et al. 1975; Rhode et al. 1983) and to project to the ventral nucleus of the lateral lemniscus (Schofield and Cant 1997).
A commissural projection from one cochlear nucleus to the other has been shown previously (Shore et al. 1992). This involved large multipolar cells of the VCN that appeared to contact fusiform cells and other cell types in the fusiform and deep cell layers (Schofield and Cant 1996). Earlier physiological evidence also suggested an inhibitory pathway between the two cochlear nuclei (Mast 1970) and there is good evidence that it is glycinergic (Wenthold 1987; Wenthold et al. 1987). In the cat it has been proposed that On-C cells provide a wideband inhibition of projection neurones in the contralateral DCN (Joris and Smith 1998). More recently, an in vivo study has provided convincing evidence that On-C cells provide a monosynaptic inhibition of T stellate chopper cells in the contralateral and ipsilateral VCN (Needham and Paolini 2003). Stimulation of the cochlear nucleus in an isolated brain preparation evoked short latency (3–9 ms) inhibition consistent with mono- and disynaptic connections in all major cell types in the contralateral cochlear nucleus (Babalian et al. 1999). Our demonstration of an On-C commisural connection would be consistent with these inhibitory functions.
It is possible that On-C cells may enhance the firing rate of some contralateral projection neurones by contacting local inhibitory neurones and producing disinhibition (Alibardi 2000).
Careful examination of our On-L-containing sections yielded no evidence for a projection beyond the ipsilateral nucleus in these cells. In our sample population, there was a clear distinction between On-C cells—which had a contralateral projection—and On-L cells which did not. Antidromic stimulation of VCN in the rat also indicated that only a proportion of onset cells projected across the midline while others did not (Needham and Paolini 2003). However, at present it is still not possible to exclude the possibility that some On-L cells may also project across the midline to the other cochlear nucleus.
Either way, it seems likely the On-C cells are involved in the first binaural interaction underlying absolute sound localization of elevation and azimuth using spectral cues. They are apparently not involved in the azimuthal discrimination between two sound sources as this has a different mechanism (May 2000).
We are grateful to Didier Pinault, University of Strasbourg, for his helpful comments and advice during the initial implementation of the juxtacellular technique.
- 4.Alibardi, L 2001Fine structure and neurotransmitter cytochemistry of neurons in the rat ventral cochlear nucleus projecting to the ipsilateral dorsal cochlear cochlear nucleus.Anat. Anz.183459469Google Scholar
- 8.Bullock, DC, Palmer, SR, Rees, A 1988Compact and easy-to-use tungsten-in-glass microelectrode manufacturing workstation.Med. Biol. Eng. Comput.26669672Google Scholar
- 26.Osen, KK, Ottersen, OP, Stormmathisen, J 1990Colocalization of glycine-like and GABA-like immunoreactivities: A semiquantitative study of individual neurons in the dorsal cochlear nucleus of cat.Ottersen, OPStorm–Mathisen, J eds. Glycine Neurotransmission.John Wiley & Sons Ltd.New York417452Google Scholar
- 34.Rapisarda, C, Bacchelli, B 1977The brain of the guinea pig in stereotaxic coordinates.Arch. Sci. Biol.61137Google Scholar
- 36.Rhode, WS, Greenberg, S 1994aEncoding of amplitude modulation in the cochlear nucleus of the cat.J. Neurophysiol.7117971825Google Scholar
- 37.Rhode, WS, Greenberg, S 1994bLateral suppression and inhibition in the cochlear nucleus of the cat.J. Neurophysiol.71493514Google Scholar
- 39.Rose, JE, Galambos, R, Hughes, JR 1959Microelectrode studies of the cochlear nucleus of the cat.Bull. Johns Hopkins Hospital104211251Google Scholar
- 48.Stabler SE. The neural representation of simple and complex sounds in the dorsal cochlear nucleus of the guinea pig. Thesis, University of Nottingham, UK, 1991.Google Scholar
- 53.Young, ED 1984Response characteristics of neurones of the cochlear nuclei.Berlin, C eds. Hearing Sciences.College-Hill PressSan Diego423460Google Scholar
- 55.Young, ED, Davis, KA 2002Circuitry and functions of the dorsal cochlear nucleus.Oertel, DPopper, ANFay, RR eds. Integrative functions of the mammalian auditory pathway.Springer-VerlagNew York160206Google Scholar
- 58.Kiang, NYS, Watanabe, T, Clarke, EC, Clark, LF 1965Discharge patterns of single fibers in the cat’s auditory nerve.M.I.T.Cambridge, Mass.Google Scholar