狠狠撸

狠狠撸Share a Scribd company logo
Responses from the trapezoid body in the Mongolian gerbil
We report results for 80 TB fibers. 8 fibers were recorded by midline approach
and the other 72 fibers was by trans-bulla approach. 26 fibers were sensitive to
ipsilateral acoustic stimulation and the other 54 fibers were sensitive to
contralateral stimulation.
Shotaro Karino and Philip X. Joris. Laboratory of Auditory Neurophysiology, K.U.Leuven, Belgium. Abstract No.858
Results: Responses to broadband noise
Methods
Conclusions
Results: Threshold and CF
Introduction
Gerbils were anesthetized with a 10:1 mixture of ketamine hydrochloride and
xylazine hydrochloride. In early recordings (N=4), we applied a midline approach to
the TB. The basioccipital bone was exposed after resection of the prevertebral
muscles. The TB was exposed by making a longitudinal slit on the bone, 3-5 mm
rostral to the jugular foramen. The advantage of this approach is that the bulla is left
intact on both sides. In later recordings (N=19), we applied a trans-bulla approach
to the TB. We made a hole on the exposed ventral wall of the bulla in one side to
get a view of the “roof” of the bulla. Then another small hole was made to expose
the overlying brainstem. Glass micropipettes, filled with 3 M NaCl were positioned
in the TB. The angle of penetration ranged from 0 to 30° mediolaterally relative to
the midsagittal plane in both midline and trans-bulla approaches. Earpieces were fit
over the transversely cut ear canals, and the acoustic assembly was calibrated with
a probe tube near the eardrum. All stimuli were computed at a sample rate of 125
kHz and compensated for the transfer function. The stimuli were specified in sound
pressure level (SPL) (dB re 20 ?Pa). The neural signal was amplified and filtered
(300 Hz to 3 kHz), and spikes were converted to standard pulses and time stamped
to an accuracy of 1 ?s. A threshold tuning curve was obtained by an automated
tracking program, from which characteristic frequency (CF), spontaneous rate (SR),
and Q10 value (CF divided by bandwidth at 10 dB above threshold) were
measured.
Subsequently, a rate-level function was obtained by presenting short tone bursts at
CF (25 ms / 100 ms, 200 repetitions, 5/10 dB steps).
Finally, the response to broadband noise (0.05 - 15 kHz) was obtained at a number
of SPLs, if time allowed.
The bottom 16 panels show PST histograms that were unexpected and
which are - in the cat - not associated with the TB. These included
several kinds of onset responses: Onset with some sustained
response (O), Pure Onset (OI), and Onset chopper (OC). A number of
unclassified PST histograms were encountered as well (X), usually
showing long latency or sluggish responses. However, some well-
timed, short-latency responses were also classified as "X" because
the PST histogram was not consistent across SPLs (M and N).
The correlation index (CI) is the peak value of the normalized shuffled
autocorrelogram (SAC). It provides a means to quantify temporal coding to any
stimulus that generates sustained responses (Joris et al. 2006). A CI of unity
indicates absence of temporal coding. In cat, temporal coding to noise is
enhanced in the TB relative to the auditory nerve (Louage et al., 2005). This
seems to be the case in the gerbil as well, with the provisio that CI values for the
auditory nerve of the gerbil are not available.
A: Tuning curves of all the
TB fibers. Superimposed
thresholds (thick line) are
those obtained by a
behavioral experiment (Ryan
1976).
B: Thresholds at CF
classified by PSTH types
(inset).
C: Relationship between CF
and spontaneous rate (SR).
D: Scatterplots of Q10 dB
values as a function of fiber
CF.
Results: Pure tone metrics
A: Relationship between CF and
significant Rmax to CF tones. Rmax
provides a convenient parameter for
comparison of phase-locking in multiple
populations at different CFs. Of 80 TB
fibers, 5 had Rmax >= 0.9, which we
call “high-sync”. All of them had CFs
below 700 Hz.
B: Relationship between CF and depth
of recorded units.
C: Relationship between CF and
maximal peak-to-sustained (P/S) ratios
in spike rates. Values of infinity are
located at the top of the panel.
D: Relationship between CF and
minimal peak-to-sustained (P/S) ratios
in spike rates. The P/S ratio was
calculated as the ratio of the response
in a 5ms window aligned with the peak
onset, and the sustained rate
subsequent to that window up to 25 ms
(Blackburn and Sachs, 1989).
PST histograms were evaluated qualitatively over a large range of SPLs,
and grouped according to the categories that have been established in
other well-studied mammals (cat, guinea pig).
The top 16 panels show PST histograms that we expected to find in the TB,
based on work in the cat (Smith et al. 1991; Smith et al. 1993; Spirou et al.
1990): phase-locked (PHL), primary-like (PL), primary-like-with-notch (PLn),
and chopper (CHOP). In the TB of the cat, these PST histograms are
associated with bushy cells (PHL, PL, PLn) and stellate cells (CHOP).
Results: PostStimulus-Time Histogram
In the cat and guinea pig, the Oc PST histograms are associated with
commissural multipolar neurons with wide dynamic range (Arnott et al. 2004;
Smith et al. 2005). We measured dynamic range by fitting a model to the rate-
level functions (Sachs et al. 1989): the average 10%-90% range is similar for
Oc and PLN neurons. Combined with the location of recording, this argues
against the association of the Oc responses recorded in the TB with
multipolar commissural neurons in this species.
Effects of SPL on normalized SACs obtained from responses to broadband noise of four TB fibers.
TB fibers generally had low spontaneous rates, and showed a large range
of thresholds. Sharpness of tuning increases with CF but does not reach as
high values as obtained in the cat.
PHL
N=10
PL
N=4
PLN
N=18
CHOP
N=11
O
N=6
OI
N=6
OC
N=12
Unusual
N=13
For each neuron, synchronization to CF tones was measured with the vector
strength and checked for statistical significance. The maximal vector strength
(Rmax) was > 0.9 in a small number of neurons, consistent with the enhanced
synchronization that has also been reported in cats. Enhanced synchronization
was also observed in the tuning curve "tail“ (not shown here).
Responses were collected over a depth of roughly 1mm. Onset responses
occurred over a similar range of depths as the PHL, PL and PLN responses. We
hypothesize that these responses are also derived from axons of bushy cells.
To quantify onset responses, we calculated a "peak-sustained ratio". Besides
onset neurons, all other categories of PST histogram (except PL) can have pure
onset responses, as indicated by "infinite" ratios at one SPL at least. In O and
OI neurons, the ratio remains large over the entire range of SPLs tested.
Results: Rate-level functions
The Mongolian gerbil offers several advantages for auditory research. Unlike other
small mammals, such as bats or mice, the gerbil posesses excellent low frequency
hearing which makes it a more appropriate model for studies of hearing in humans.
Gerbils have been increasingly used to study the mechanisms of sensitivity to
interaural time differences. To properly interpret binaural data, it is important to also
know the monaural properties of the inputs to the binaural nuclei. To our knowledge,
there have been no studies of responses in the trapezoid body. Here, we report
responses of monaural fibers in the trapezoid body (TB) in the gerbil.
Normalized rate-level functions for each category of PST-histogram. For each curve, SPL is
plotted relative to the mimimal threshold of the tuning curve, and the rate is normalized to the
maximum.
Noise responses were obtained in a limited number of fibers: spike isolation was
often difficult in response to noise. We used shuffled autocorrelograms (SACs)
to quantify temporal coding to noise. Responses to reference and anticorrelated
noise were used to construct cross-stimulus autocorrelograms (XACs). Bin per
bin subtraction of XACs from SACs results in "difcors," which reveal response
components that change on inverting the polarity, and therefore reflect
synchronization to the fine structure of the effective stimulus waveform (Joris
2003; Louage et al. 2004).
Synchronization to stimulus fine structure.
A and B: typical SACs and XACs (thick and thin lines,
respectively). All correlograms are normalized and
therefore asymptote toward unity at long delays.
C and D: Oscillatory Difcors.
E and F: Fourier spectrum of difcors.
Synchronization to stimulus envelope.
A and B: typical SACs and XACs (thick and thin
lines, respectively).
C and D: Difcors.
Correlograms reveal coding of fine-structure to noise and envelope.
Although coding of
fine-structure was encountered, coding of envelope predominated,
presumably because our sample is biased towards high CFs.
Relationship between CF and CI.
The top limit of CIs of cat TB fibers is shown
with a solid line (Louage et al. 2005). The
top limit of CIs of cat AN fibers is shown with
a dashed line (Louage et al. 2004).
Most of the 80 TB fibers of gerbil showed a
range of CIs similar to that in the cat,
however, some data indicate higher values.
? In the trapezoid body of the gerbil, a surprisingly large number of onset
responses are recorded from monaural fibers.
? The occurrence of “high-sync” responses (maximal vector strength > 0.9) was
not as prominent as in the TB of the cat.
? The onset responses observed occurred over a similar range of depths as the
primary-like responses (PHL, PL and PLN responses). We hypothesize that there
is more diversity in the PST histograms of AVCN neurons in the gerbil than in the
cat.
REFERENCES
Arnott R, et al. J Assoc Res Otolaryngol 5: 153-170, 2004.
Blackburn CC and Sachs MB. J Neurophysiol 62: 1303-1329, 1989.
Joris PX. J Neurosci 23: 6345-6350, 2003.
Joris PX, et al. Hear Res 216-217: 19-30, 2006.
Louage DH, et al. J Neurophysiol 91: 2051-2065, 2004.
Louage DH, et al. J Neurosci 25: 1560-1570, 2005.
Ryan A. J Acoust Soc Am 59: 1222-1226, 1976.
Sachs MB, et al. Hear Res 41: 61-69, 1989.
Smith PH, et al. J Neurophysiol 79: 3127-3142, 1998.

More Related Content

Responses from the trapezoid body in the Mongolian gerbil

  • 1. Responses from the trapezoid body in the Mongolian gerbil We report results for 80 TB fibers. 8 fibers were recorded by midline approach and the other 72 fibers was by trans-bulla approach. 26 fibers were sensitive to ipsilateral acoustic stimulation and the other 54 fibers were sensitive to contralateral stimulation. Shotaro Karino and Philip X. Joris. Laboratory of Auditory Neurophysiology, K.U.Leuven, Belgium. Abstract No.858 Results: Responses to broadband noise Methods Conclusions Results: Threshold and CF Introduction Gerbils were anesthetized with a 10:1 mixture of ketamine hydrochloride and xylazine hydrochloride. In early recordings (N=4), we applied a midline approach to the TB. The basioccipital bone was exposed after resection of the prevertebral muscles. The TB was exposed by making a longitudinal slit on the bone, 3-5 mm rostral to the jugular foramen. The advantage of this approach is that the bulla is left intact on both sides. In later recordings (N=19), we applied a trans-bulla approach to the TB. We made a hole on the exposed ventral wall of the bulla in one side to get a view of the “roof” of the bulla. Then another small hole was made to expose the overlying brainstem. Glass micropipettes, filled with 3 M NaCl were positioned in the TB. The angle of penetration ranged from 0 to 30° mediolaterally relative to the midsagittal plane in both midline and trans-bulla approaches. Earpieces were fit over the transversely cut ear canals, and the acoustic assembly was calibrated with a probe tube near the eardrum. All stimuli were computed at a sample rate of 125 kHz and compensated for the transfer function. The stimuli were specified in sound pressure level (SPL) (dB re 20 ?Pa). The neural signal was amplified and filtered (300 Hz to 3 kHz), and spikes were converted to standard pulses and time stamped to an accuracy of 1 ?s. A threshold tuning curve was obtained by an automated tracking program, from which characteristic frequency (CF), spontaneous rate (SR), and Q10 value (CF divided by bandwidth at 10 dB above threshold) were measured. Subsequently, a rate-level function was obtained by presenting short tone bursts at CF (25 ms / 100 ms, 200 repetitions, 5/10 dB steps). Finally, the response to broadband noise (0.05 - 15 kHz) was obtained at a number of SPLs, if time allowed. The bottom 16 panels show PST histograms that were unexpected and which are - in the cat - not associated with the TB. These included several kinds of onset responses: Onset with some sustained response (O), Pure Onset (OI), and Onset chopper (OC). A number of unclassified PST histograms were encountered as well (X), usually showing long latency or sluggish responses. However, some well- timed, short-latency responses were also classified as "X" because the PST histogram was not consistent across SPLs (M and N). The correlation index (CI) is the peak value of the normalized shuffled autocorrelogram (SAC). It provides a means to quantify temporal coding to any stimulus that generates sustained responses (Joris et al. 2006). A CI of unity indicates absence of temporal coding. In cat, temporal coding to noise is enhanced in the TB relative to the auditory nerve (Louage et al., 2005). This seems to be the case in the gerbil as well, with the provisio that CI values for the auditory nerve of the gerbil are not available. A: Tuning curves of all the TB fibers. Superimposed thresholds (thick line) are those obtained by a behavioral experiment (Ryan 1976). B: Thresholds at CF classified by PSTH types (inset). C: Relationship between CF and spontaneous rate (SR). D: Scatterplots of Q10 dB values as a function of fiber CF. Results: Pure tone metrics A: Relationship between CF and significant Rmax to CF tones. Rmax provides a convenient parameter for comparison of phase-locking in multiple populations at different CFs. Of 80 TB fibers, 5 had Rmax >= 0.9, which we call “high-sync”. All of them had CFs below 700 Hz. B: Relationship between CF and depth of recorded units. C: Relationship between CF and maximal peak-to-sustained (P/S) ratios in spike rates. Values of infinity are located at the top of the panel. D: Relationship between CF and minimal peak-to-sustained (P/S) ratios in spike rates. The P/S ratio was calculated as the ratio of the response in a 5ms window aligned with the peak onset, and the sustained rate subsequent to that window up to 25 ms (Blackburn and Sachs, 1989). PST histograms were evaluated qualitatively over a large range of SPLs, and grouped according to the categories that have been established in other well-studied mammals (cat, guinea pig). The top 16 panels show PST histograms that we expected to find in the TB, based on work in the cat (Smith et al. 1991; Smith et al. 1993; Spirou et al. 1990): phase-locked (PHL), primary-like (PL), primary-like-with-notch (PLn), and chopper (CHOP). In the TB of the cat, these PST histograms are associated with bushy cells (PHL, PL, PLn) and stellate cells (CHOP). Results: PostStimulus-Time Histogram In the cat and guinea pig, the Oc PST histograms are associated with commissural multipolar neurons with wide dynamic range (Arnott et al. 2004; Smith et al. 2005). We measured dynamic range by fitting a model to the rate- level functions (Sachs et al. 1989): the average 10%-90% range is similar for Oc and PLN neurons. Combined with the location of recording, this argues against the association of the Oc responses recorded in the TB with multipolar commissural neurons in this species. Effects of SPL on normalized SACs obtained from responses to broadband noise of four TB fibers. TB fibers generally had low spontaneous rates, and showed a large range of thresholds. Sharpness of tuning increases with CF but does not reach as high values as obtained in the cat. PHL N=10 PL N=4 PLN N=18 CHOP N=11 O N=6 OI N=6 OC N=12 Unusual N=13 For each neuron, synchronization to CF tones was measured with the vector strength and checked for statistical significance. The maximal vector strength (Rmax) was > 0.9 in a small number of neurons, consistent with the enhanced synchronization that has also been reported in cats. Enhanced synchronization was also observed in the tuning curve "tail“ (not shown here). Responses were collected over a depth of roughly 1mm. Onset responses occurred over a similar range of depths as the PHL, PL and PLN responses. We hypothesize that these responses are also derived from axons of bushy cells. To quantify onset responses, we calculated a "peak-sustained ratio". Besides onset neurons, all other categories of PST histogram (except PL) can have pure onset responses, as indicated by "infinite" ratios at one SPL at least. In O and OI neurons, the ratio remains large over the entire range of SPLs tested. Results: Rate-level functions The Mongolian gerbil offers several advantages for auditory research. Unlike other small mammals, such as bats or mice, the gerbil posesses excellent low frequency hearing which makes it a more appropriate model for studies of hearing in humans. Gerbils have been increasingly used to study the mechanisms of sensitivity to interaural time differences. To properly interpret binaural data, it is important to also know the monaural properties of the inputs to the binaural nuclei. To our knowledge, there have been no studies of responses in the trapezoid body. Here, we report responses of monaural fibers in the trapezoid body (TB) in the gerbil. Normalized rate-level functions for each category of PST-histogram. For each curve, SPL is plotted relative to the mimimal threshold of the tuning curve, and the rate is normalized to the maximum. Noise responses were obtained in a limited number of fibers: spike isolation was often difficult in response to noise. We used shuffled autocorrelograms (SACs) to quantify temporal coding to noise. Responses to reference and anticorrelated noise were used to construct cross-stimulus autocorrelograms (XACs). Bin per bin subtraction of XACs from SACs results in "difcors," which reveal response components that change on inverting the polarity, and therefore reflect synchronization to the fine structure of the effective stimulus waveform (Joris 2003; Louage et al. 2004). Synchronization to stimulus fine structure. A and B: typical SACs and XACs (thick and thin lines, respectively). All correlograms are normalized and therefore asymptote toward unity at long delays. C and D: Oscillatory Difcors. E and F: Fourier spectrum of difcors. Synchronization to stimulus envelope. A and B: typical SACs and XACs (thick and thin lines, respectively). C and D: Difcors. Correlograms reveal coding of fine-structure to noise and envelope. Although coding of fine-structure was encountered, coding of envelope predominated, presumably because our sample is biased towards high CFs. Relationship between CF and CI. The top limit of CIs of cat TB fibers is shown with a solid line (Louage et al. 2005). The top limit of CIs of cat AN fibers is shown with a dashed line (Louage et al. 2004). Most of the 80 TB fibers of gerbil showed a range of CIs similar to that in the cat, however, some data indicate higher values. ? In the trapezoid body of the gerbil, a surprisingly large number of onset responses are recorded from monaural fibers. ? The occurrence of “high-sync” responses (maximal vector strength > 0.9) was not as prominent as in the TB of the cat. ? The onset responses observed occurred over a similar range of depths as the primary-like responses (PHL, PL and PLN responses). We hypothesize that there is more diversity in the PST histograms of AVCN neurons in the gerbil than in the cat. REFERENCES Arnott R, et al. J Assoc Res Otolaryngol 5: 153-170, 2004. Blackburn CC and Sachs MB. J Neurophysiol 62: 1303-1329, 1989. Joris PX. J Neurosci 23: 6345-6350, 2003. Joris PX, et al. Hear Res 216-217: 19-30, 2006. Louage DH, et al. J Neurophysiol 91: 2051-2065, 2004. Louage DH, et al. J Neurosci 25: 1560-1570, 2005. Ryan A. J Acoust Soc Am 59: 1222-1226, 1976. Sachs MB, et al. Hear Res 41: 61-69, 1989. Smith PH, et al. J Neurophysiol 79: 3127-3142, 1998.