| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TRANSLATIONAL PHYSIOLOGY
1The Bionic Ear Institute, East Melbourne Victoria; and 2School of Psychological Science and 3Department of Electronic Engineering, La Trobe University, Melbourne, Victoria, Australia
Submitted 7 June 2007; accepted in final form 6 October 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
,b). The neurons of the VCN are arranged in a tonotopic order, with low-frequency AN fibers innervating the most ventral and lateral regions of the VCN and high-frequency fibers innervating more dorsal and medial regions (Rose et al. 1959). On the basis of cytoarchitecture, there are several neuron types found in the VCN, each one different in anatomy (Hackney et al. 1990; Osen 1969) and function (Evans and Nelson 1973; Shofner and Young 1985). The posteroventral cochlear nucleus (PVCN) in several animal species predominantly contains two types of cells that are known to project to higher auditory brain centers, such as the central nucleus of the inferior colliculus (CIC) of the midbrain, through both direct and polysynaptic pathways (Adams 1979a; Beyerl 1978; Cant 1992; Oliver 1987; Osen 1972; Schofield and Cant 1996). One such type, the T-stellate neurons (also known as type I multipolar neurons), are thought to be the prime encoders of monaural frequency (Brawer 1974; Cant 1981; Osen 1969; Paolini et al. 2004, 2005) and project directly to the CIC (Adams 1979a; Beyerl 1978; Cant 1992; Oliver 1987; Osen 1972; Schofield and Cant 1996). Another cell type, the octopus cells, are located in the most posterior regions of the PVCN and are a major source of inhibition to the CIC through the ventral nucleus of the lateral lemniscus (Adams 1997; Friauf and Ostwald 1988; Nayagam et al. 2005, 2006; Saint Marie and Baker 1990; Schofield and Cant 1997; Thompson 1998). The anteroventral cochlear nucleus (AVCN) is known to contain a high proportion of bushy cells (Harrison and Warr 1962; Osen 1969; Tolbert and Morest 1982a,b), which do not directly project to the CIC (Adams 1979b, 1983; Ryugo et al. 1981). However, these cells project to the superior olivary complex (Cant and Casseday 1986; Warr 1966), which in turn sends projections to the CIC (Browner and Webster 1975; Roth et al. 1978).
In several animal species, the CIC is tonotopically arranged from low to high frequencies in a dorsomedial–ventrolateral direction (Aitkin et al. 1972; FitzPatrick 1975; Huang and Fex 1986; Semple and Aitkin 1979). Although there are some differences in the anatomical characteristics of the VCN across species (Cant and Benson 2003), given that both the VCN and the CIC are tonotopically organized, it is likely that the projections between the structures are frequency-specific, which is supported by previous anatomical studies (Oliver 1987; Osen 1972). What remains to be investigated is whether these tonotopic projections are functionally valid for VCN stimulation. The knowledge gained by assessing this could provide significant further information on how to electrically stimulate the VCN with the auditory brain stem implant (ABI).
The ABI differs from the cochlear implant (CI) in that it stimulates the surface of the cochlear nucleus (CN) rather than along the tonotopic gradient of the cochlea (Otto et al. 1998). Commercially available implants consist of either 12 (MED-EL C40+, MED-EL, Innsbruck, Austria) or 21 (Nucleus 24 ABI, Cochlear, Sydney, NSW, Australia) active stimulating electrodes fixed on a Dacron fabric mesh that fits on the surface of the CN in the lateral recess of the fourth ventricle. The ABI has mainly been used to restore hearing in patients with Neurofibromatosis Type II (NF2), a genetic condition usually marked by bilateral tumor growths on the VIIIth cranial nerve (Otto et al. 2002). Damage to this nerve after removal of these tumors may result in profound sensorineural hearing loss. Although patients with this condition are unable to benefit from the CI designed to stimulate the spiral ganglion neurons of the cochlea, they may benefit from stimulation of the VCN when implanted with the ABI (Edgerton et al. 1982). In other conditions where a CI will not be effective, such as a temporal bone fracture with cochlear nerve avulsion, cochlear ossification, and possibly cochlear nerve aplasia, patients may also benefit from an ABI (Colletti and Shannon 2005). When compared with CI recipients several clinical studies have shown that NF2 ABI users obtain only limited sound perception (Otto et al. 1998, 2002; Schwartz et al. 2003). The typical ABI user will benefit from environmental sound perception, although in most cases is unable to understand speech without lipreading (Otto et al. 1998, 2002; Schwartz et al. 2003). A recent study (Colletti and Shannon 2005) reported that some non-NF2 patients can achieve significantly higher speech perception than NF2 patients. Although they suggested a separate pathway for speech and modulation coding, which might be damaged in NF2 patients, it is unclear why only some non-NF2 patients receive benefit.
Previous studies involving CIs (Friesen et al. 2001) and ABIs (Kuchta et al. 2004) have shown that the ability to perceive speech is largely correlated with the number of independent channels of frequency information. One way this may be achieved is by stimulation of localized groups of neurons within isofrequency laminae of the VCN. However, the present surface implant technology may have limited access to the tonotopic map located within the VCN. A penetrating electrode array may overcome this limitation, as suggested by previous studies showing low thresholds, high dynamic ranges, and topographic specificity in higher brain centers to VCN stimulation (El-Kashlan 1999; El-Kashlan et al. 1991; McCreery et al. 1998, 2007; Takahashi et al. 2005).
In this investigation we examined the frequency specificity of activation of CIC neurons by microstimulation of the VCN across multiple sites. Our approach was to directly compare CIC multiunit cluster responses obtained by electrical stimulation to those obtained acoustically. Unlike previous studies (El-Kashlan 1999; El-Kashlan et al. 1991; McCreery et al. 1998, 2007; Takahashi et al. 2005), we stimulated specific frequency regions in the VCN that were predetermined using acoustic stimulation. Multichannel electrodes were inserted in both structures along their tonotopic axes and, after predetermining the characteristic frequencies (CFs) of the multiunit clusters at each electrode site, we stimulated each VCN site with single biphasic charge-balanced pulses. We recorded multiunit spike activity in the CIC and analyzed multiunit spike rate in response to stimulation, which allowed accurate measures of electrical thresholds, dynamic ranges, and frequency specificity of activation. The hypothesis of our study was that CIC clusters respond with lowest current threshold to electrical stimulation of a frequency-matched site in the VCN. We also hypothesized that this frequency specificity should depend on location of stimulation within the VCN. Our results have significant implications for ABIs.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Male Hooded Wistar rats (n = 8) weighing between 350 and 450 g were anesthetized with urethane in water (20% wt/vol; Sigma–Aldrich, Castle Hill, NSW, Australia) via the intraperitoneal cavity. The animals were placed in a stereotaxic frame (David Kopf Instruments, Tujunga, CA) and fitted with hollow ear bars. Animal temperature was regulated at 37°C and monitored continuously using a DC homoeothermic blanket. Contralateral craniotomies were performed to access both the VCN and the inferior colliculus (IC). The cerebellum was aspirated after removing the outer duramater to expose the brain stem and VCN. Under visual control, using the lateral recess of the fourth ventricle as a guide, 32-channel electrodes (Neuronexus Technologies, Ann Arbor, MI) were inserted both into the ipsilateral (left) VCN and the contralateral (right) CIC along their tonotopic axes. In all experimental animals, the VCN electrodes were inserted in a caudorostral direction to access the central parts of the PVCN, which have a high concentration of T-stellate cells (see RESULTS). The CIC electrodes for all experiments were inserted at a 10° rostrocaudal angle. A low-impedance silver reference electrode was placed under the skin and served as a common reference for spike recording. All surgical procedures and protocols were approved by the La Trobe University Animal Ethics Committee (Protocol # 03/13) and the St. Vincent's Hospital Animal Ethics Committee (Protocol # 35/06).
Multichannel electrodes
Each multichannel electrode consisted of four silicon substrate shanks (200 µm apart), each 5 mm long, attached to a fiberglass circuit board with eight iridium electrode sites (413 µm2 surface area, 200 µm apart) on each shank (referred to later as shanks a–d with sites 1–8 on each shank; see Figs. 1 and 2). Before each experiment, to increase their charge storage capacity, all 32 sites of the VCN electrode were electrochemically activated using cyclic voltammetry as described by Anderson et al. (1989) (660B Potentiostat and CHI200 Picoamp Booster, CH Instruments, Austin, TX). On activation, the impedance of each site was changed from 1 M
to <100 k at 1 kHz, whereas each CIC electrode site had an impedance (unactivated) of 1–3 M at 1 kHz.
|
|
All stimuli generation and data acquisition were performed using Tucker Davis Technologies System III hardware (TDT, Alachua, FL), with a custom-designed software package using TDT's OpenEx client-server applications. Acoustically driven units were first located in both structures by presenting a broadband (1–44 kHz) Gaussian-distributed noise stimulus (50-ms bursts, 500-ms intertrial interval) while advancing the electrode arrays into the VCN and CIC using a motorized microdrive (Sutter Instrument, Novato, CA). Once the arrays were in position, responses to various pure tones (1–44 kHz, 1-kHz steps, 50-ms bursts, 300-ms intertrial interval) of different amplitudes [10- to 70-dB sound pressure level (SPL), 10-dB steps] were obtained to generate a response area for each recording site (10 repetitions for each frequency–intensity combination). Ear bars were calibrated before each experiment using a 1/8-in. Brüel & Kjær (Nærum, Denmark) microphone and measuring amplifier unit.
Electrical protocols
Each acoustically driven electrode site in the VCN was electrically stimulated with charge-balanced biphasic current pulses using the site above it on the same shank as a reference (120 µs per phase, 80-µs interphase gap, bipolar configuration). For the topmost site on each shank, the site immediately below it was used as a reference. Microstimulators (32-channel RX7 microstimulators; TDT) were used to deliver currents 
100 µA (typically maximum current delivered was 54 µA). The pulse rate on every channel was fixed to one pulse every 500 ms (50 repetitions for each current amplitude). Because the electrode sites had a surface area of 413 µm2 and our stimulus pulses were 120 µs per phase, even at 100 µA, the maximum charge density was 2.9 mC/cm2 and the charge per phase was 12 nC. This ensured that we were not above the electrochemical charge injection limit for iridium oxide (
3 mC/cm2) (Beebe and Rose 1988) and that we did not cause any neural damage. McCreery et al. (1990) showed that charge per phases of 5,000 nC could be used safely with the charge density set at 10 µC/cm2 and the pulse width set to 400 µs per phase at a stimulation rate of 50 Hz. At our stimulation rate (2 Hz), the threshold of damage would be even higher. Moreover, at the end of the stimulation paradigm in each experiment, acoustic response areas were repeated to ensure that the VCN neurons were still functional and showed similar tuning properties.
Data acquisition
Data were acquired at a sampling rate of 24.4 kHz using an on-line threshold-based spike discriminator. To determine when spikes occurred, the incoming signal on each channel was monitored in a 2-s window to get an estimate of the spontaneous baseline level of activity. If the signal crossed this baseline activity by >4.2 SDs, a spike detector was triggered, which recorded the next 31 samples (1.2 ms) of the signal waveform (band-pass filtered from 300 to 5,000 Hz) along with a timestamp value indicating the time of occurrence of the largest positive or negative peak of the spike. After recording 31 samples, the spike detector could be triggered again.
Data analysis
INITIAL PROCEDURE.
Multiunit activity from each of the electrode sites was analyzed off-line, using programs we developed in Matlab (The MathWorks, Natick, MA). A typical approach to analyze multiunit recordings is spike sorting (Lewicki 1998). One of the major challenges for any sorting technique has been to distinguish spike waveforms that occur as a result of two or more neurons firing simultaneously, resulting in fully or partially overlapping spikes (Bar-Gad et al. 2001). In structures like the VCN and the CIC, it is expected that when an acoustic stimulus is presented, not only one but several neurons close to a recording site will respond simultaneously. Therefore it was not possible to determine whether any spike waveform acquired by our system corresponded to a single neuron's spike; thus we analyzed all our data using the multiunit activity from each site. Also, the method of on-line spike detection meant that while the spike detector was recording a waveform, it could not be triggered by another spike until it had recorded 31 samples of the signal. This could result in the recording system not accurately keeping track of all the spikes, especially those that occurred very near in time to each other. Therefore a rethresholding procedure was used off-line to accurately determine the occurrence of spikes. This procedure involved finding the mean and SD of each 31-sample spike recorded by the system, and then checking how many times any particular sample out of the 31 exceeded 1.5SDs above the mean. Each threshold crossing was picked up by a Schmitt trigger function, which kept track of the occurrence of spikes. This provided a more accurate estimate for spike rate calculations for analysis of the data.
ACOUSTIC RESPONSE AREAS. Spike rate calculations were made for each frequency–amplitude combination of sound presented using a time window 0–50 ms from the stimulus onset. These were then plotted as a function of frequency and intensity to give a response area for each of the 64 recorded sites in the two structures. The CF for each site in the VCN and the CIC was determined from its response area as the frequency that elicited maximum activity at the lowest threshold (dB SPL). Acoustic threshold was determined by visual inspection of the peristimulus time histograms (PSTHs) and by selecting the dB SPL where spike discharge significantly increased above spontaneous levels within a window of 0–50 ms from stimulus onset.
ELECTRICAL RESPONSE AREAS. To directly compare CIC responses to electrical and acoustic stimulation, spike rate calculations in response to electrical stimulation were made using a window of 2–25 ms from stimulus onset. This ensured that the stimulus artifact recorded in the CIC would not be counted in the calculations. By plotting these spike rates as a function of the current amplitude and CF of each stimulated VCN site, we were able to generate an electrical response area (see RESULTS) for each CIC site and compare its tuning characteristics directly to the tuning characteristics of the acoustic response area.
ELECTRICAL THRESHOLD, SATURATION, AND DYNAMIC RANGE.
Methods to obtain electrical thresholds vary and range from visual inspection of the PSTHs to curve fitting incorporating spontaneous and maximum firing rates. For our data, we chose an objective mathematical method described by Koppl and Yates (1999) and Nizami (2002). This method calculates the thresholds along with saturation firing rates and therefore dynamic ranges can also be found. First, a least-squares regression method was used to fit a sigmoid to the stimulus level versus firing rate plots for all electrically evoked responses. Endpoints of the dynamic range were then determined by calculating the current levels required to produce firing rates 10% above threshold and 10% below saturation (Koppl and Yates 1999; Nizami 2002; Yates et al. 2000). This method was selected because it has been shown to successfully model mammalian and avian AN rate-level functions (Koppl and Yates 1999; Sachs and Abbas 1974; Sachs et al. 1989; Yates 1990; Yates et al. 2000) and the threshold and saturation estimates obtained from our data using this method matched well to the values obtained by visual inspection of the electrical PSTHs. Dynamic ranges are presented in µA and in dB.
Histological analysis
At the end of each experiment, the animal was deeply anesthetized and perfused transcardially using 0.1 M phosphate-buffered saline and 10% neutral buffered formalin (Sigma–Aldrich). After tissue fixation, the brain was removed and snap-frozen. Serial coronal sections of 60 µm were collected using a freezing sledge microtome (Thomas Scientific, Swedesboro, NJ). Sections were placed on gelatin-coated slides, stained with thionine, and coverslipped. Positions of the electrodes through the VCN and CIC were confirmed using bright-field microscopy and VCN shank locations were reconstructed by manual drawings using a combination of histological data and known visual placements of the electrodes. Outlines of the CN were determined in the horizontal and parasagittal planes with reference to the rat brain atlas (Paxinos and Watson 2005). On viewing the coronal histological sections, the placement of the VCN probe was estimated and dots representing the shanks were added to the outlines, giving a relative indication of where the VCN electrodes were placed within the boundaries of the model outlines (see Fig. 1). Histological data were obtained in six of the eight experimental animals. Due to multiple VCN penetrations, histological data could not be verified for the remaining two animals.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Figure 1A summarizes the placement of electrode shanks at their entry point into the CN in the horizontal plane. In four of the six animals, electrodes were inserted either toward the medial edge or the lateral edge of the VCN, whereas in two animals, placements were made in the central regions of the VCN. In one experiment (07_007), although the electrode was inserted into the dorsal CN (DCN) at its entry point, it was inserted with a greater caudorostral angle compared with the other experiments. As a result, few of the dorsally located electrode sites from this experiment were either in the DCN or close to the granule cell lamina between the DCN and the VCN (indicated in Fig. 1B, black dotted lines). However, the electrode sites used for stimulation in this experiment were located in the VCN. Multiunit activity was recorded from individual sites located within the VCN across the tonotopic plane: low-frequency–sensitive neurons in the ventral regions and high-frequency–sensitive neurons in the more dorsal regions (Fig. 1B, i–vi). Minimal tissue damage was noted in the VCN as a result of these penetrating electrode shanks by visual inspection of the histological sections close to where the electrode tracks were seen (Fig. 1C, i and ii). Electrode placements in the IC were verified histologically within its central nucleus (Fig. 2; Paxinos and Watson 2005). Multiunit activity recorded across the CIC showed the extent of the tonotopic gradient with clusters sensitive from low frequencies to high frequencies in the dorsoventral direction.
Characteristics of VCN electrical stimulation
In 334 of 512 possible sites (32 sites in each structure, 8 animals), multiunit cluster responses to frequencies at multiple intensity levels were obtained (response areas: 141, VCN; 193, CIC). Because our experiments involved sophisticated methods of surgery and recording techniques, due to experimental constraints it was on occasion not possible to stimulate all VCN sites in a given experiment. Of the VCN sites, 126 (89.4%) were electrically stimulated, of which 100 initiated a response in the CIC. The remaining 26 VCN sites did not show any response in the CIC even when stimulated at the highest current amplitude (100 µA). For the CIC, all 193 sites showed a response to electrical stimulation of at least one site in the VCN. To further analyze CIC responses to VCN stimulation, electrode sites in both structures were paired according to their CF. Those CIC sites that had a CF equal to or 1 kHz away from VCN sites in a given experiment were classified together in one group called "CF Aligned." The other group consisted of CIC sites that responded with the lowest threshold to stimulation of a single VCN site. These were classified into the "Lowest Threshold Aligned" group, consistent with Lim and Anderson (2006). VCN sites that did not have a CF-matched site in the CIC (n = 32) were not paired with any CIC sites and not included in the analysis. For both the groups, analyses were performed on the data obtained from stimulation of 68 VCN sites.
A total of 364 VCN–CIC pairs belonging to the CF Aligned group were analyzed for electrical stimulation (Fig. 3). In response to increasing levels of both acoustic and electrical stimulation, CIC sites exhibited an increase in discharge (Fig. 3, A and B) with multiunit spike activity generally observed between 4 and 60 ms from stimulus onset. In contrast to acoustic stimulation, a well-timed onset peak was seen with higher stimulus currents in PSTHs from electrical stimulation, 4–5 ms from stimulus onset (Fig. 3B), possibly a result of antidromic activation in the VCN.
|
|
|
; Fig. 6). Monotonic responses (Fig. 6, Ai and Bi) were classified in cases where spike rates did not reach a saturation level and kept increasing with stimulus levels, whereas those that did saturate were classified as plateau-type responses (Fig. 6, Aii and Bii). Nonmonotonic responses (Fig. 6, Aiii and Biii) were classified in cases where spike rates first increased with stimulus levels but dropped to <50% of the maximum spike rate at higher stimulus levels. A final group of multiunit clusters exhibited both monotonic and nonmonotonic rate-level characteristics (Fig. 6, Aiv and Biv), classified as complex responses. The distribution of all response types seen with both acoustic (n = 138) and electrical (n = 289) stimulation is shown in Fig. 6C. In most CIC multiunit clusters (56.5%), acoustic rate-level functions did not reach a saturation point (Fig. 6C, left). This was in contrast to that seen on electrical stimulation of CF-aligned sites where the majority of clusters (63.7%) exhibited spike rate saturation over our typical stimulated current range (0–54 µA; Fig. 6C, right).
|
|
; Swadlow 1974; Swadlow et al. 1978) these properties are not always sufficient to discern antidromic spikes from orthodromic spikes; thus we labeled our responses as "presumed" antidromic activity. Of 1,313 total CIC responses to stimulation of the 100 VCN sites, 565 responses (43%) showed the presence of antidromic activity. In 541 (96%) of these responses, the threshold of antidromic activity to VCN stimulation was found to be equal to or higher than the threshold of the delayed orthodromic activity. Mean electrical threshold of all antidromic responses was found to be 30.72 ± 13 µA (mean ± SD). Although we have included these responses in our analysis, they did not influence our overall threshold measures. Electrical thresholds versus acoustic thresholds
To directly compare the acoustic and electrical tuning at threshold levels for all CIC sites, both on-CF and off-CF, we plotted regression lines for the thresholds elicited by acoustic stimulation at given frequencies against the electrical thresholds elicited by stimulation of VCN sites with CFs equal to those frequencies (Fig. 8). Only CIC sites that responded to stimulation of at least three VCN sites were analyzed (n = 135), of which 25.93% exhibited a high correlation (Pearson's coefficient of correlation, r > 0.7), whereas 11% showed moderate correlation (0.5 < r 0.7) between acoustic and electrical thresholds (Hinkle et al. 1998). Negatively correlated sites accounted for about 35% of the total number, whereas 20% of the total sites fell in the –0.2 to –0.5 range (Fig. 8I). As seen from the single experiment plots in Fig. 8, A–H, the majority of negatively correlated CIC sites came from two experiments with centrally located VCN electrode positions (Fig. 1, Exp 06_002 and Exp 06_003), whereas the remaining experiments had CIC sites that showed positive correlation between acoustic and electrical thresholds.
|
The data were analyzed on a VCN shank by shank basis across all experiments to test whether location of VCN stimulation correlated with the degree of frequency-specific CIC activation. For each VCN site stimulated that had at least one CF-matched site in the CIC (n = 68), the tonotopic gradient of the CIC was examined for the lowest threshold response (Fig. 9 A). For a perfect tonotopic point-to-point mapping between the VCN and the CIC, each frequency region in the VCN when stimulated should elicit the lowest threshold response in the same frequency region in the CIC (Fig. 9A, solid line). For each point in Fig. 9A, the frequency difference between the stimulated VCN site and the lowest threshold CIC site was calculated. The average frequency difference for each VCN shank was used as a measure of shank effectiveness in eliciting frequency-specific CIC activation. VCN shanks with 
67% of the lowest threshold VCN–CIC pairs that had frequency differences 3 kHz were classified into the "Group 1 shanks" (closed circles, Fig. 9, A and B). Remaining VCN shanks were classified into a second group ("Group 2 shanks"; open circles, Fig. 9, A and B). The overall degree of frequency-specific CIC activation was found to be higher for the Group 1 shanks, which were placed more centrally, on the medial edge or more posterior in the VCN, whereas more lateral and anterior VCN placements (Group 2 shanks) showed a lower degree of frequency specificity (Fig. 9C).
|
In addition to being frequency specific, the mean threshold elicited in the CIC for the Group 1 shanks (12.8 ± 6.4 µA) was found to be significantly lower (P < 0.001) than the mean threshold for the Group 2 shanks (21.2 ± 11.5 µA). The mean dynamic range was found to be larger (P < 0.001) for the Group 2 shanks (6.67 ± 3.2 dB) compared with the Group 1 shanks (4.93 ± 2.7 dB) (Fig. 9, E and F; mean ± SD).
Effect of broadness of tuning on frequency specificity
Figure 10 shows the relationship between changes in electrical thresholds for all CIC sites from the threshold of a CF-matched VCN–CIC pair, plotted against their corresponding frequency differences from the VCN site in that pair. In cases where there was more than one CF-matched CIC site for a given VCN site, the change in threshold was calculated from the one that responded with the lowest threshold. From the acoustic response area of each CIC site, the Quality factor at 10 dB above threshold (Q10) was calculated as the CF divided by the bandwidth in kHz, as a measure of sharpness of tuning (Kelly et al. 1991). Sites that gave a Q10 1 (n = 72) were classified as broadly tuned, whereas those with a Q10 >1 (n = 121) were classified as narrowly tuned. For the CIC sites that responded to VCN sites in the Group 1 shanks, both broadly tuned (Fig. 10, A1 and A2) and narrowly tuned (Fig. 10, B1 and B2) CIC sites showed a trend for threshold differences to increase with a rise in frequency difference (narrowly tuned, r = 0.41, P < 0.01; broadly tuned, r = 0.27, P < 0.01). For the Group 2 shanks, only the broadly tuned CIC sites exhibited this trend of increasing threshold differences with increasing frequency differences (r = 0.22, P < 0.05).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
), and in a tonotopic manner based on anatomical studies (Oliver 1987; Osen 1972), we expected that CIC neurons would be most sensitive to stimulation of VCN neurons with the same CF and, conversely, less sensitive to stimulation of neurons with different CFs. In support of this, a high degree of frequency specificity was seen in VCN–CIC pairs between the CFs of regions activated in the CIC with the lowest threshold and CFs of regions stimulated in the VCN. This frequency specificity was not only present at threshold levels but was also maintained across all current levels in our stimulation range. A major factor contributing to this high degree of frequency specificity was the placement of the electrodes within the VCN. Best results were obtained when electrodes were placed toward the medial, central, or posterolateral regions rather than more anterolateral regions of this nucleus.
The posterolateral and central placements would stimulate regions within the PVCN where a high concentration of T-stellate cells are found (Smith and Rhode 1989). Moreover, a high proportion of CIC sites exhibited maximum spike discharge to stimulation of the central and posterolateral VCN regions, which is consistent with a large number of T-stellate cell projections arising from these regions that correspond to the main excitatory inputs to the contralateral CIC. Medial and a few of the extreme lateral placements close to the free surface of the VCN would most likely stimulate the granule cell domain (Mugnaini et al. 1980a,b; Weedman et al. 1996). These cells are known to project to the fusiform cells (also known as pyramidal cells) of the DCN (Mugnaini et al. 1980b; Ryugo et al. 1995), which in turn project to the CIC (Berrebi and Mugnaini 1991). The medial extent of the granule cell domain is more developed in rats (Mugnaini et al. 1980b), thus increasing the likelihood of granule cell stimulation with medial shank placements in our experiments.
In contrast, the anterolateral placements would predominantly excite the AVCN, which is known to contain a high proportion of bushy cells (Harrison and Warr 1962; Osen 1969; Tolbert and Morest 1982a,b). Because bushy cells do not project directly to the CIC (Adams 1979b, 1983; Ryugo et al. 1981), higher levels of current would be required to evoke a CIC response for these anterolateral placements as current spreads to neighboring T-stellate regions. This indiscriminant activation of VCN T-stellate neurons may explain the high proportion of unmatched VCN–CIC pairs with lowest threshold from anterolateral VCN placements. This was manifested predominantly in narrowly tuned CIC neurons, whereas broadly tuned neurons were generally unaffected. This alludes to the possibility that narrowly tuned CIC neurons may receive projections from the narrowly tuned T-stellate neurons of the PVCN, whereas broadly tuned CIC neurons may receive polysynaptic projections from a wide variety of cells types, including those from the AVCN. Furthermore, a large proportion of these anterolateral placements did not elicit a response in CF-aligned VCN–CIC pairs and very few CIC sites responded with maximum spike discharge to stimulation of these regions. This suggests that the current spread from these placements was confined to the AVCN. It is possible that in those cases, even the highest current amplitude (54 µA) was not enough for current to spread to the central PVCN T-stellate regions.
From a CIC perspective, over a third of the CIC sites analyzed showed a moderate to high correlation between their acoustic tuning and electrical tuning. However, for CIC sites in the experiments that had central VCN placements, a poor correlation was observed (Exp 06_002 and 06_003). In these experiments, we found that although lowest threshold was obtained by stimulating a CF-matched site in the VCN, when VCN sites away from CF were stimulated, the thresholds did not increase substantially. One possible explanation for this could be related to the nature of current spread in the VCN. Central regions of the VCN are known to have a high proportion of AN fibers entering via the AN trunk (Arnesen and Osen 1978; Arnesen et al. 1978; Lorente de Nó 1933a,b; Osen 1970). The high probability of stimulating these fibers of passage may affect electrical tuning in the CIC at higher current levels.
Electrical stimulation thresholds obtained in our study were consistent with the results obtained by McCreery et al. (2000) who obtained CIC activation through chronic VCN stimulation at current levels as low as 6 µA. We also found that two VCN sites within an experiment that had the same CF as a CIC site, elicited different electrical threshold values and some VCN sites when stimulated, did not elicit any response in CF-matched CIC sites. This lack of stimulus driven activity in higher-order auditory structures resulting from stimulation of some CN sites was also noted by Takahashi et al. (2005) who observed a lack of auditory cortical activation from a large proportion of CN sites. In this paradigm, we made use of bipolar stimulation to minimize current spread across frequency laminae. It is possible that neurons within the CIC receive convergent input from several points within a VCN isofrequency lamina. It is unknown whether it is sufficient to stimulate only a part of an isofrequency lamina to convey meaningful frequency information in higher-order centers. There may be benefit in stimulating a greater number of sites within an isofrequency lamina to maximize frequency-specific activation in the CIC.
Back-propagation of fibers may also influence the responses obtained within the CIC to electrical stimulation of the VCN. Multiunit clusters in the CIC often exhibited a well-timed onset response to VCN electrical stimulation, which occurred at latencies in the order of 4–5 ms from stimulus onset. This time-locked onset response may be a result of antidromic activation of descending projections known to exist from the IC to the CN (Coomes and Schofield 2004; Faye-Lund 1988; Okoyama et al. 2006; Schofield 2001; Schofield and Cant 1999).
Implications for an ABI
Our study examined the degree of frequency specificity of CIC activation achievable by electrical stimulation of the VCN. The results of this study are directly applicable to the development of ABIs. We have identified the following four major implications from our research.
VCN ELECTRODE PLACEMENT.
The need for the electrodes to access the tonotopic organization of the VCN is an important consideration in achieving better performance. Based on our results, it is evident that the placement of the implant array into the VCN is a critical factor in conveying frequency-specific information to higher-order brain centers. We have shown using a penetrating electrode array that medial, central, and posterolateral regions of the VCN when stimulated produce greater frequency-specific CIC activation compared with the anterolateral regions. This is consistent with McCreery et al. (1998) who showed that central regions of the PVCN are suitable for penetrating ABI implantation and McCreery et al. (2007) who recently showed that the rostrolateral and rostromedial region of the VCN when stimulated chronically elicited significantly lower degrees of CIC frequency specificity compared with the caudolateral and caudomedial VCN regions. Based on our results, if a surface array is placed on the posterolateral surface of the VCN in a dorsoventral direction using the present commercial ABI, some frequency-specific stimulation may be achievable but a penetrating array in the PVCN would result in an increased likelihood of frequency-specific CIC activation. However, a correct placement of the electrode array is not sufficient on its own to guarantee best results. One must consider the effects of deafness periods on the structure and function of the VCN. Hearing loss can evoke significant morphological and physiological changes within the auditory brain stem, and these changes become greater with duration of deafness (Hardie and Shepherd 1999). In particular, hearing loss is associated with a reduction in overall CN volume and an increase in neural density (Hardie and Shepherd 1999), which are likely to influence the effectiveness of the ABI.
CURRENT LOCALIZATION.
We have shown some similarity between electrical and acoustic response areas in the CIC but this was not found in all cases. In fact, regardless of whether CIC neurons were broadly or narrowly tuned acoustically, they responded with similar thresholds to stimulation of VCN regions within a 2-kHz range for our classified Group 1 shanks, which were placed medially, centrally, or posterolaterally in the VCN. For the Group 2 shanks that were placed more anterolaterally, similar threshold CIC activation was within a 6-kHz range. Although bipolar stimulation of the VCN was implemented using electrode sites with relatively small surface areas, we did not always achieve localized, frequency-specific CIC activation. Perhaps more sophisticated methods like tripolar or even quadrupolar stimulation (Jolly et al. 1996) or using different pulse widths and shapes may be an advantage over bipolar stimulation. However, one must consider to what extent stimulation of passing fibers might be a limiting factor in achieving frequency-specific activation.
We realize that our results were obtained by stimulating a rat VCN, where the frequency laminae would be less widespread as compared with a human and, in addition, the electrode dimensions that we used in this study were in the order of micrometers instead of millimeters. Humans and several other primate species have their best hearing frequencies (indicated by audiograms) centered around 4–8 kHz and the upper hearing limit of most humans is centered around 17 kHz (Heffner 2004). This is in contrast to the rat, which has an upper hearing limit of 80 kHz (Kelly and Masterton 1977). Thus we would expect the human VCN to have a tonotopic map with frequencies shifted toward lower-frequency scales compared with those of the rat.
ABI POWER REQUIREMENTS.
Our results using penetrating electrodes show that charge deliveries as low as 0.36 nC (3 µA, 120 µs per phase) are sufficient to elicit a frequency-specific response in the CIC. Present surface electrodes have been reported to elicit thresholds in the order of 2 nC (Colletti and Shannon 2005). CIC multiunit clusters in our study exhibited average spike rate saturation at current levels close to 30 µA. These results imply that a penetrating ABI would have more efficient power requirements.
FACILITATION OF INFORMATION TRANSFER. A final implication of our results is that not all acoustically driven VCN stimulation sites when electrically stimulated produced a response in the CIC. This implies a need for an electrode design that will incorporate sufficient redundancy to optimize performance. In addition, individual stimulation of two or more VCN areas with similar CFs often elicited different thresholds and different rate-level functions in the CIC. We propose that a greater number of electrode sites will not only provide sufficient redundancy but also allow stimulation of multiple sites within VCN isofrequency laminae to maximize frequency-specific CIC activation.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: A. G. Paolini, Auditory Clinical Neuroscience Unit, The Bionic Ear Institute, 6th Floor Daly Wing, St. Vincent's Hospital, Victoria Parade, Fitzroy, VIC 3065, Australia (E-mail: tpaolini{at}bionicear.org)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Adams JC. Ascending projections to the inferior colliculus. J Comp Neurol 183: 519–538, 1979b.[CrossRef][Web of Science][Medline]
Adams JC. Multipolar cells in the ventral cochlear nucleus project to the dorsal cochlear nucleus and the inferior colliculus. Neurosci Lett 37: 205–208, 1983.[CrossRef][Web of Science][Medline]
Adams JC. Projections from octopus cells of the posteroventral cochlear nucleus to the ventral nucleus of the lateral lemniscus in cat and human. Audit Neurosci 3: 335–350, 1997.
Aitkin L. Rate-level functions of neurons in the inferior colliculus of cats measured with the use of free-field sound stimuli. J Neurophysiol 65: 383–392, 1991.
Aitkin LM, Fryman S, Blake DW, Webster WR. Responses of neurones in the rabbit inferior colliculus. I. Frequency-specificity and topographic arrangement. Brain Res 47: 77–90, 1972.[CrossRef][Web of Science][Medline]
Anderson DJ, Najafi K, Tanghe SJ, Evans DA, Levy KL, Hetke JF, Xue XL, Zappia JJ, Wise KD. Batch-fabricated thin-film electrodes for stimulation of the central auditory system. IEEE Trans Biomed Eng 36: 693–704, 1989.[CrossRef][Web of Science][Medline]
Arnesen AR, Osen KK. The cochlear nerve in the cat: topography, cochleotopy, and fiber spectrum. J Comp Neurol 178: 661–678, 1978.[CrossRef][Web of Science][Medline]
Arnesen AR, Osen KK, Mugnaini E. Temporal and spatial sequence of anterograde degeneration in the cochlear nerve fibers of the cat. A light microscopic study. J Comp Neurol 178: 679–696, 1978.[CrossRef][Web of Science][Medline]
Bar-Gad I, Ritov Y, Vaadia E, Bergman H. Failure in identification of overlapping spikes from multiple neuron activity causes artificial correlations. J Neurosci Methods 107: 1–13, 2001.[CrossRef][Web of Science][Medline]
Beebe X, Rose TL. Charge injection limits of activated iridium oxide electrodes with 0.2 ms pulses in bicarbonate buffered saline. IEEE Trans Biomed Eng 35: 494–495, 1988.[CrossRef][Web of Science][Medline]
Berrebi AS, Mugnaini E. Distribution and targets of the cartwheel cell axon in the dorsal cochlear nucleus of the guinea pig. Anat Embryol (Berl) 183: 427–454, 1991.[Medline]
Beyerl BD. Afferent projections to the central nucleus of the inferior colliculus in the rat. Brain Res 145: 209–223, 1978.[CrossRef][Web of Science][Medline]
Brawer JR, Morest DK, Kane EC. The neuronal architecture of the cochlear nucleus of the cat. J Neurophysiol 155: 251–300, 1974.
Browner RH, Webster DB. Projections of the trapezoid body and the superior olivary complex of the kangaroo rat (Dipodomys merriami). Brain Behav Evol 11: 322–354, 1975.[Web of Science][Medline]
Cant NB. The fine structure of two types of stellate cells in the anterior division of the anteroventral cochlear nucleus of the cat. Neuroscience 6: 2643–2655, 1981.[CrossRef][Web of Science][Medline]
Cant NB. The cochlear nucleus: neuronal types and their synaptic organization. In: The Mammalian Auditory Pathway: Neuroanatomy, edited by Webster DB, Popper AN, Fay RR. New York: Springer-Verlag, 1992, p. 66–116.
Cant NB, Benson CG. Parallel auditory pathways: projection patterns of the different neuronal populations in the dorsal and ventral cochlear nuclei. Brain Res Bull 60: 457–474, 2003.[CrossRef][Web of Science][Medline]
Cant NB, Casseday JH. Projections from the anteroventral cochlear nucleus to the lateral and medial superior olivary nuclei. J Comp Neurol 247: 457–476, 1986.[CrossRef][Web of Science][Medline]
Colletti V, Shannon RV. Open set speech perception with auditory brainstem implant? Laryngoscope 115: 1974–1978, 2005.[CrossRef][Web of Science][Medline]
Coomes DL, Schofield BR. Separate projections from the inferior colliculus to the cochlear nucleus and thalamus in guinea pigs. Hear Res 191: 67–78, 2004.[CrossRef][Web of Science][Medline]
Edgerton BJ, House WF, Hitselberger W. Hearing by cochlear nucleus stimulation in humans. Ann Otol Rhinol Laryngol Suppl 91: 117–124, 1982.[Medline]
El-Kashlan HK. Multichannel cochlear nucleus stimulation. Otolaryngol Head Neck Surg 121: 169–175, 1999.[CrossRef][Web of Science][Medline]
El-Kashlan HK, Niparko JK, Altschuler RA, Miller JM. Direct electrical stimulation of the cochlear nucleus: surface vs. penetrating stimulation. Otolaryngol Head Neck Surg 105: 533–543, 1991.[Web of Science][Medline]
Evans EF, Nelson PG. On the functional relationship between the dorsal and ventral divisions of the cochlear nucleus of the cat. Exp Brain Res 17: 428–442, 1973.[Web of Science][Medline]
Faye-Lund H. Inferior colliculus and related descending pathways in rat. Ups J Med Sci 93: 1–17, 1988.[Web of Science][Medline]
FitzPatrick KA. Cellular architecture and topographic organization of the inferior colliculus of the squirrel monkey. J Comp Neurol 164: 185–207, 1975.[CrossRef][Web of Science][Medline]
Friauf E, Ostwald J. Divergent projections of physiologically characterized rat ventral cochlear nucleus neurons as shown by intra-axonal injection of horseradish peroxidase. Exp Brain Res 73: 263–284, 1988.[Web of Science][Medline]
Friesen LM, Shannon RV, Baskent D, Wang X. Speech recognition in noise as a function of the number of spectral channels: comparison of acoustic hearing and cochlear implants. J Acoust Soc Am 110: 1150–1163, 2001.[CrossRef][Web of Science][Medline]
Hackney CM, Osen KK, Kolston J. Anatomy of the cochlear nuclear complex of guinea pig. Anat Embryol (Berl) 182: 123–149, 1990.[Medline]
Hardie NA, Shepherd RK. Sensorineural hearing loss during development: morphological and physiological response of the cochlea and auditory brainstem. Hear Res 128: 147–165, 1999.[CrossRef][Web of Science][Medline]
Harrison JM, Warr WB. A study of the cochlear nuclei and ascending auditory pathways of the medulla. J Comp Neurol 119: 341–380, 1962.[CrossRef][Web of Science][Medline]
Heffner RS. Primate hearing from a mammalian perspective. Anat Rec A Discov Mol Cell Evol Biol 281: 1111–1122, 2004.[Medline]
Hinkle DE, Wiersma W, Jurs SG. Correlation: a measure of relationship. In: Applied Statistics for the Behavioral Sciences (4th ed.). Boston, MA: Houghton Mifflin, 1998, p. 105–131.
Huang CM, Fex J. Tonotopic organization in the inferior colliculus of the rat demonstrated with the 2-deoxyglucose method. Exp Brain Res 61: 506–512, 1986.[Web of Science][Medline]
Jolly CN, Spelman FA, Clopton BM. Quadrupolar stimulation for cochlear prostheses: modeling and experimental data. IEEE Trans Biomed Eng 8: 857–865, 1996.[Medline]
Kelly JB, Glenn SL, Beaver CJ. Sound frequency and binaural response properties of single neurons in rat inferior colliculus. Hear Res 56: 273–280, 1991.[CrossRef][Web of Science][Medline]
Kelly JB, Masterton B. Auditory sensitivity of the albino rat. J Comp Physiol Psychol 91: 930–936, 1977.[CrossRef][Web of Science][Medline]
Koppl C, Yates G. Coding of sound pressure level in the barn owl's auditory nerve. J Neurosci 19: 9674–9686, 1999.
Kuchta J, Otto SR, Shannon RV, Hitselberger WE, Brackmann DE. The multichannel auditory brainstem implant: how many electrodes make sense? J Neurosurg 100: 16–23, 2004.[Web of Science][Medline]
Lewicki MS. A review of methods for spike sorting: the detection and classification of neural action potentials. Network 9: 53–78, 1998.
Lim HH, Anderson DJ. Auditory cortical responses to electrical stimulation of the inferior colliculus: implications for an auditory midbrain implant. J Neurophysiol 96: 975–988, 2006.
Lim HH, Anderson DJ. Antidromic activation reveals tonotopically organized projections from primary auditory cortex to the central nucleus of the inferior colliculus in guinea pig. J Neurophysiol 97: 1413–1427, 2007.
Lorente de Nó R. Anatomy of the eighth nerve. The central projection of the nerve endings of the internal ear. Laryngoscope 43: 1–38, 1933a.
Lorente de Nó R. Anatomy of the eighth nerve. III. General plan of the structure of the primary cochlear nuclei. Laryngoscope 43: 327–350, 1933b.
McCreery DB, Agnew WF, Yuen TG, Bullara LA. Charge density and charge per phase as cofactors in neural injury induced by electrical stimulation. IEEE Trans Biomed Eng 37: 996–1001, 1990.[CrossRef][Web of Science][Medline]
McCreery DB, Lossinsky A, Pikov V. Performance of multisite silicon microprobes implanted chronically in the ventral cochlear nucleus of the cat. IEEE Trans Biomed Eng 54: 1042–1051, 2007.[CrossRef][Web of Science][Medline]
McCreery DB, Shannon RV, Moore JK, Chatterjee M. Accessing the tonotopic organization of the ventral cochlear nucleus by intranuclear microstimulation. IEEE Trans Rehabil Eng 6: 391–399, 1998.[CrossRef][Medline]
McCreery DB, Yuen TG, Bullara LA. Chronic microstimulation in the feline ventral cochlear nucleus: physiologic and histologic effects. Hear Res 149: 223–238, 2000.[CrossRef][Web of Science][Medline]
Mugnaini E, Osen KK, Dahl AL, Friedrich VL Jr, Korte G. Fine structure of granule cells and related interneurons (termed Golgi cells) in the cochlear nuclear complex of cat, rat and mouse. J Neurocytol 9: 537–570, 1980b.[CrossRef][Web of Science][Medline]
Mugnaini E, Warr WB, Osen KK. Distribution and light microscopic features of granule cells in the cochlear nuclei of cat, rat, and mouse. J Comp Neurol 191: 581–606, 1980a.[CrossRef][Web of Science][Medline]
Nayagam DA, Clarey JC, Paolini AG. Powerful, onset inhibition in the ventral nucleus of the lateral lemniscus. J Neurophysiol 94: 1651–1654, 2005.
Nayagam DA, Clarey JC, Paolini AG. Intracellular responses and morphology of rat ventral complex of the lateral lemniscus neurons in vivo. J Comp Neurol 498: 295–315, 2006.[CrossRef][Web of Science][Medline]
Nizami L. Estimating auditory neuronal dynamic range using a fitted function. Hear Res 167: 13–27, 2002.[CrossRef][Web of Science][Medline]
Okoyama S, Ohbayashi M, Ito M, Harada S. Neuronal organization of the rat inferior colliculus participating in four major auditory pathways. Hear Res 218: 72–80, 2006.[CrossRef][Web of Science][Medline]
Oliver DL. Projections to the inferior colliculus from the anteroventral cochlear nucleus in the cat: possible substrates for binaural interaction. J Comp Neurol 264: 24–46, 1987.[CrossRef][Web of Science][Medline]
Osen KK. Cytoarchitecture of the cochlear nuclei in the cat. J Comp Neurol 136: 453–484, 1969.[CrossRef][Web of Science][Medline]
Osen KK. Course and termination of the primary afferents in the cochlear nuclei of the cat. An experimental anatomical study. Arch Ital Biol 108: 21–51, 1970.[Web of Science][Medline]
Osen KK. Projection of the cochlear nuclei on the inferior colliculus in the cat. J Comp Neurol 144: 355–372, 1972.[CrossRef][Web of Science][Medline]
Otto SR, Brackmann DE, Hitselberger WE, Shannon RV, Kuchta J. Multichannel auditory brainstem implant: update on performance in 61 patients. J Neurosurg 96: 1063–1071, 2002.[Web of Science][Medline]
Otto SR, Shannon RV, Brackmann DE, Hitselberger WE, Staller S, Menapace C. The multichannel auditory brain stem implant: performance in twenty patients. Otolaryngol Head Neck Surg 118: 291–303, 1998.[CrossRef][Web of Science][Medline]
Paolini AG, Clarey JC, Needham K, Clark GM. Fast inhibition alters first spike timing in auditory brain stem neurons. J Neurophysiol 92: 2615–2621, 2004.
Paolini AG, Clarey JC, Needham K, Clark GM. Balanced inhibition and excitation underlies spike firing regularity in ventral cochlear nucleus chopper neurons. Eur J Neurosci 21: 1236–1248, 2005.[CrossRef][Web of Science][Medline]
Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates (5th ed.). San Diego, CA: Academic Press, 2005.
Rose JE, Galambos R, Hughes JR. Microelectrode studies of the cochlear nuclei of the cat. Bull Johns Hopkins Hosp 104: 211–251, 1959.[Web of Science][Medline]
Roth GL, Aitkin LM, Anderson RA, Merzenich MM. Some features of the spatial organization of the central nucleus of the inferior colliculus of the cat. J Comp Neurol 182: 661–680, 1978.[CrossRef][Web of Science][Medline]
Ryugo DK, Pongstaporn T, Wright DD, Sharp AH. Inositol 1,4,5-trisphosphate receptors: immunocytochemical localization in the dorsal cochlear nucleus. J Comp Neurol 358: 102–118, 1995.[CrossRef][Web of Science][Medline]
Ryugo DK, Willard FH, Fekete DM. Differential afferent projections to the inferior colliculus from the cochlear nucleus in the albino mouse. Brain Res 210: 342–349, 1981.[CrossRef][Web of Science][Medline]
Sachs MB, Abbas PJ. Rate versus level functions for auditory-nerve fibers in cats: tone-burst stimuli. J Acoust Soc Am 6: 1835–1847, 1974.
Sachs MB, Winslow RL, Sokolowski BH. A computational model for rate-level functions from cat auditory-nerve fibers. Hear Res 41: 61–69, 1989.[CrossRef][Web of Science][Medline]
Saint Marie RL, Baker RA. Neurotransmitter-specific uptake and retrograde transport of [3H]glycine from the inferior colliculus by ipsilateral projections of the superior olivary complex and nuclei of the lateral lemniscus. Brain Res 524: 244–253, 1990.[CrossRef][Web of Science][Medline]
Schofield BR. Origins of projections from the inferior colliculus to the cochlear nucleus in guinea pigs. J Comp Neurol 429: 206–220, 2001.[CrossRef][Web of Science][Medline]
Schofield BR, Cant NB. Projections from the ventral cochlear nucleus to the inferior colliculus and the contralateral cochlear nucleus in guinea pigs. Hear Res 102: 1–14, 1996.[CrossRef][Web of Science][Medline]
Schofield BR, Cant NB. Ventral nucleus of the lateral lemniscus in guinea pigs: cytoarchitecture and inputs from the cochlear nucleus. J Comp Neurol 379: 363–385, 1997.[CrossRef][Web of Science][Medline]
Schofield BR, Cant NB. Descending auditory pathways: projections from the inferior colliculus contact superior olivary cells that project bilaterally to the cochlear nuclei. J Comp Neurol 409: 210–223, 1999.[CrossRef][Web of Science][Medline]
Schwartz MS, Otto SR, Brackmann DE, Hitselberger WE, Shannon RV. Use of a multichannel auditory brainstem implant for neurofibromatosis type 2. Stereotact Funct Neurosurg 81: 110–114, 2003.[CrossRef][Web of Science][Medline]
Semple MN, Aitkin LM. Representation of sound frequency and laterality by units in central nucleus of cat inferior colliculus. J Neurophysiol 42: 1626–1639, 1979.
Shofner WP, Young ED. Excitatory/inhibitory response types in the cochlear nucleus: relationships to discharge patterns and responses to electrical stimulation of the auditory nerve. J Neurophysiol 54: 917–939, 1985.
Smith PH, Rhode WS. Structural and functional properties distinguish two types of multipolar cells in the ventral cochlear nucleus. J Comp Neurol 282: 595–616, 1989.[CrossRef][Web of Science][Medline]
Swadlow HA. Properties of antidromically activated callosal neurons and neurons responsive to callosal input in rabbit binocular cortex. Exp Neurol 43: 424–444, 1974.[CrossRef][Web of Science][Medline]
Swadlow HA, Waxman SG, Rosene DL. Latency variability and the identification of antidromically activated neurons in mammalian brain. Exp Brain Res 32: 439–443, 1978.[Web of Science][Medline]
Takahashi H, Nakao M, Kaga K. Accessing ampli-tonotopic organization of rat auditory cortex by microstimulation of cochlear nucleus. IEEE Trans Biomed Eng 52: 1333–1344, 2005.[CrossRef][Web of Science][Medline]
Thompson AM. Heterogeneous projections of the cat posteroventral cochlear nucleus. J Comp Neurol 390: 439–453, 1998.[CrossRef][Web of Science][Medline]
Tolbert LP, Morest DK. The neuronal architecture of the anteroventral cochlear nucleus of the cat in the region of the cochlear nerve root: electron microscopy. Neuroscience 7: 3053–3067, 1982a.[CrossRef][Web of Science][Medline]
Tolbert LP, Morest DK. The neuronal architecture of the anteroventral cochlear nucleus of the cat in the region of the cochlear nerve root: Golgi and Nissl methods. Neuroscience 7: 3013–3030, 1982b.[CrossRef][Web of Science][Medline]
Warr WB. Fiber degeneration following lesions in the anterior ventral cochlear nucleus of the cat. Exp Neurol 14: 453–474, 1966.[CrossRef][Web of Science][Medline]
Weedman DL, Pongstaporn T, Ryugo DK. Ultrastructural study of the granule cell domain of the cochlear nucleus in rats: mossy fiber endings and their targets. J Comp Neurol 369: 345–360, 1996.[CrossRef][Web of Science][Medline]
Yates GK. Basilar membrane nonlinearity and its influence on auditory nerve rate-intensity functions. Hear Res 50: 145–162, 1990.[CrossRef][Web of Science][Medline]
Yates GK, Manley GA, Koppl C. Rate-intensity functions in the emu auditory nerve. J Acoust Soc Am 107: 2143–2154, 2000.[CrossRef][Web of Science][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
