- Email: [email protected]
Interaction of excitation and inhibition in inferior collicular neurons of the big brown bat, Eptesicus fuscus
Hearing Research 169 (2002) 140^150 www.elsevier.com/locate/heares
Interaction of excitation and inhibition in inferior collicular neurons of the big brown bat, Eptesicus fuscus Yong Lu 1 , Philip H.-S. Jen
Division of Biological Sciences, University of Missouri-Columbia, Columbia, MO 65211, USA Received 22 January 2002; accepted 9 April 2002
Abstract Neurons in the inferior colliculus (IC) of the midbrain receive convergent excitatory and inhibitory inputs from both lower and higher auditory nuclei. Interaction of these two opposing inputs shapes different response properties of IC neurons. In this study, we examined this interaction of excitation and inhibition in IC neurons using a probe (excitatory pulse) and a masker (inhibitory pulse) under different stimulation conditions. Inhibition of probe-elicited responses by a masker, i.e. masking, occurred when the masker was presented at certain inter-pulse intervals (the temporal window) in relation to the probe. At the best inter-pulse interval, masking was maximal such that a neuron had the minimal number of impulses, the longest response latency, and the smallest excitatory frequency tuning curve. The temporal window for masking expanded with increasing masker duration. The inhibition decreased with increasing probe intensity but increased with increasing masker intensity. Increasing masker intensity also produced progressive shrinkage in excitatory frequency tuning curves. Similarly, increasing probe intensity produced progressive shrinkage of inhibitory frequency tuning curves. Possible mechanisms underlying the time and intensity dependence of inhibition are discussed. / 2002 Elsevier Science B.V. All rights reserved. Key words: Bat; Inferior colliculus; Intensity; Temporal window; Masking
1. Introduction In sensory signal processing, neural inhibition plays an important role in shaping response properties of sensory neurons. For example, neural inhibition contributes to color contrast and edge contrast in the retina, the lateral geniculate nucleus and the visual cortex (Eysel et al., 1987; Kelly, 1973; Singer and Creutzfeldt, 1970). In the somatosensory system, neural inhibition de¢nes the excitatory areas of peripheral and central somatosensory neurons (Mountcastle and Powell, * Corresponding author. Tel.: +1 (573) 882-7479; Fax: +1 (573) 884-5020. E-mail address: [email protected] (P.H.-S. Jen). 1 Present address: Department of Otolaryngology-HNS, University of Washington, Seattle, WA 98195, USA.
Abbreviations: BF, best frequency; EPSP, excitatory postsynaptic potential; FTC, frequency tuning curve; IC, inferior colliculus; IPI, inter-pulse interval; IPSP, inhibitory postsynaptic potential; MT, minimum threshold; PST, peristimulus time
1959). In the auditory system, Katsuki et al. (1958) suggested that increasing sharpness of excitatory frequency tuning curves (FTCs) along the central auditory pathway was due to synaptic inhibition from neighboring auditory neurons. Since then, many studies have examined the contribution of neural inhibition in auditory signal processing in the frequency domain using a forward masking paradigm. These studies showed that many central auditory neurons have inhibitory areas on one or both frequency £anks of their excitatory FTCs (Ehret and Merzenich, 1988; Fuzessery and Feng, 1983; Jen and Zhang, 1999; Lu and Jen, 2001; Suga and Tsuzuki, 1985; Suga et al., 1997; Zhou and Jen, 2000). These inhibitory areas contribute to sharpening of excitatory FTCs (see review by Suga, 1995). All these examples are evidence that sensory signal processing is determined by the interplay between excitatory and inhibitory inputs regardless of sensory modality. In the ascending auditory pathway, the central nucleus of the inferior colliculus (IC) receives and integrates excitatory and inhibitory inputs from many lower
0378-5955 / 02 / $ ^ see front matter / 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 2 ) 0 0 4 5 7 - 4
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
auditory nuclei (Aitkin, 1986; Casseday and Covey, 1996; Pollak and Casseday, 1989) and from the auditory cortex (Herbert et al., 1991; Hu¡man and Henson, 1990; Saldan‹a et al., 1996; Winer et al., 1998). For this reason, temporal and spectral signal processing of IC neurons is a¡ected by the timing and intensity of excitatory and inhibitory inputs. Presumably, inhibition only occurs when both excitatory and inhibitory inputs arrive at IC neurons within a certain time window. Furthermore, inhibitory inputs with stronger intensity and longer duration should be more e¡ective in producing inhibition of auditory response of IC neurons than inhibitory inputs with weaker intensity and shorter duration. The main objective of this study was to determine how the interaction of excitation and inhibition might shape di¡erent response properties of IC neurons. To achieve this objective, we examined how the response of IC neurons elicited by a probe (an excitatory pulse) might be inhibited by a masker (an inhibitory pulse) under di¡erent stimulation conditions. Speci¢cally, we examined variation in the number of impulses, response latency, and excitatory FTC of IC neurons with interpulse interval (IPI). We then studied the e¡ect of masker duration on rate^IPI functions. We also studied how inter-pulse intensity di¡erence might a¡ect excitatory and inhibitory FTCs of IC neurons.
2. Materials and methods Seven big brown bats (four males, three females, 24^30 g body weight, b.w.) were used for this study. The surgical procedures have been described in previous studies (Jen et al., 1987). Brie£y, the £at head of a 1.8 cm nail was glued onto the exposed skull of each Nembutal-anesthetized bat (45^50 mg/kg b.w.) with acrylic glue and dental cement 1 or 2 days before the recording session. Exposed tissue was treated with an antibiotic (Neosporin) to prevent in£ammation. During recording, the bat was administered the neuroleptanalgesic, Innovar-Vet (fentanyl 0.08 mg/kg b.w., droperidol 4 mg/kg b.w.) and was placed inside a bat holder (made of wire mesh) which was suspended in an elastic sling. After the bat’s head was immobilized, small holes were bored in the skull above the IC for insertion of micropipette electrodes ¢lled with 3 M KCl to record sound-activated IC neurons. The bat’s head was oriented such that its eye^snout line was pointed to 0‡ in azimuth and 0‡ in elevation relative to the position of the loudspeaker. Recordings were conducted inside a double-walled sound-proof room (temperature 28^30‡C) with its ceiling and inside walls covered with 3 inch convoluted polyurethane foam to reduce echoes. A local anesthetic
141
(lidocaine) was applied to the open wound area. The recording depth was read from the scale of a microdrive (David-Kopf). A common indi¡erent electrode (silver wire) was placed at the nearby temporal muscles. The experiments were conducted in compliance with NIH publication No. 85-23, ‘Principles of Laboratory Animal Care’ and with the approval of the Institutional Animal Care and Use Committee (#1438) of the University of Missouri-Columbia. As in previous studies (Jen et al., 1987), continuous sine waves from an oscillator (KH model 1200) were formed into 4 ms pulses (0.5 ms rise^decay times, at 2 pps) by a homemade pulse generator (electronic switch) driven by a stimulator (Grass S88). These pulses were ampli¢ed after passing through a decade attenuator (HP 350D) before being fed to a small condenser loudspeaker (AKG model CK 50, 1.5 cm diameter, 1.2 g). The loudspeaker was placed 23 cm away from the bat and 30‡ contralateral to the recording site. Calibration of the loudspeaker was conducted with a Bru«el and Kjaer 1/4 inch microphone (4135) placed at the position where the bat’s head would be during recording. The output was expressed in dB SPL referred to 20 WPa root mean square. A frequency^response curve of the loudspeaker was plotted to determine the maximal available stimulus intensity at each frequency. Frequency-modulated (FM) pulses were generated by means of a ramp generator. Upon isolation of an acoustically evoked IC neuron with 4 ms pulses, its excitatory best frequency (BF) and minimum threshold (MT) were determined by changing the sound frequency and intensity. At the MT, the neuron, on average, responded with 50% probability to excitatory BF pulses. A probe was then delivered at the excitatory BF and 10 dB above the excitatory MT to elicit excitatory response of the isolated neuron. An inhibitory FTC was then plotted with combinations of frequency and intensity of a masker that decreased the neuron’s excitatory response by at least 20%. Because big brown bats use downward-sweep FM pulses for orientation, we determined the temporal window of masking using FM pulses sweeping downward across the 10 dB bandwidth of excitatory and inhibitory FTCs as a probe and a masker, respectively (i.e. the bandwidth of the excitatory and inhibitory FTCs at 10 dB above the MT, see Fig. 1A). However, if a neuron did not respond or responded poorly to FM pulses, BF pulses delivered at 10 dB above excitatory and inhibitory MTs were used as the probe and masker. The masker was delivered 10 ms before (expressed as negative IPIs), simultaneously (expressed as zero IPI), and 4 ms after (expressed as positive IPIs) the probe at 1 ms increments. When a neuron had low- and high-£ank inhibitory FTCs, BF or FM pulses of both inhibitory FTCs were sequentially used as the masker. The neu-
HEARES 3935 2-7-02
142
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
ron’s number of impulses and latency in response to excitatory FM or BF pulses at 10 dB above the MT were obtained at each IPI. These data were then plotted against the IPI to obtain the neuron’s rate^IPI and latency^IPI functions. The neuron’s temporal window of masking was de¢ned as the range of IPIs within which a neuron’s number of impulses was decreased by a masker by at least 50% of the maximal response. The two IPIs corresponding to the beginning and end of the temporal window were arbitrarily de¢ned as the start IPI and end IPI of the temporal window. Within the temporal window, there was a best IPI or a range of IPIs at which masking was maximal (see Fig. 1B). We then studied the e¡ect of masker duration on the temporal window of masking with a 4 ms probe and three di¡erent masker durations (1.5, 4, and 8 ms). To determine the e¡ect of inter-pulse intensity di¡erences on masking of responses of IC neurons, we plotted the inhibitory FTCs at di¡erent probe intensities. Speci¢cally, a series of inhibitory FTCs was plotted when 4 ms BF probe was delivered at the best IPI and at 10 dB increments above the excitatory MT. The number of impulses was also obtained when the intensity of probe and masker was increased in 10 dB increments above the excitatory and inhibitory MTs. Using the number of impulses elicited by the probe alone as a control, the percent masking was then calculated by dividing the change in the number of impulses under masking conditions by the control. Whenever possible, a series of excitatory FTCs was plotted with the masker delivered at the best IPI and at 10 dB increments above the inhibitory MT. Recorded action potentials were ampli¢ed, band-pass ¢ltered (Krohn-Hite 3500), and fed through a window discriminator (WPI 121) before being sent to an oscilloscope (Tektronix 5111) and an audio monitor (Grass AM6). They were then sent to a computer (Gateway 2000, 486) for acquisition of peristimulus time (PST) histograms (bin width: 500 Ws, sampling period : 300 ms) to 16 stimulus presentations. The total number of impulses in each PST histogram was used to quantify a neuron’s response under each stimulus condition.
3. Results Masking of auditory response was studied in 78 IC neurons (48 neurons studied with FM pulses and 30 neurons with BF pulses) which were isolated at depths of 346^1992 Wm. The excitatory BF ranged from 17.1 to 91.3 kHz and excitatory MT ranged from 10 to 85 dB SPL. When stimulated with 4 ms BF or FM pulses, these neurons either discharged one to three impulses (phasic responders, 43, 55%), four to seven impulses (phasic bursters, 26, 33%) or discharged
impulses throughout the entire 4 ms pulses (tonic responders, 9, 12%), similar to what we have reported earlier (Jen and Schlegel, 1982; Jen and Feng, 1999; Lu et al., 1997). Inhibitory FTCs were obtained at both the low and the high £ank (49, 63%) (Fig. 1A), or only at the low £ank (17, 22%) or high £ank (12, 15%) of the excitatory FTCs. Because the e¡ect of IPI, pulse duration and intensity on masking of response of IC neurons determined with FM pulses was similar to that determined with BF pulses, all data are combined together in the following presentation. 3.1. The e¡ect of IPI on masking Fig. 1A shows an IC neuron which had inhibitory FTCs (hatched) at the low and high £ank of its V-shaped excitatory FTC (un¢lled circles). Both inhibitory MTs were higher than excitatory MT but low£ank inhibitory MT was lower than high-£ank inhibitory MT. As shown in the right panel, this neuron discharged 42 impulses to a probe sweeping across the 10 dB bandwidth of excitatory FTC (used as a control, con). When a masker sweeping across the 10 dB bandwidth of low-£ank inhibitory FTC was delivered prior to the probe, the number of impulses progressively decreased to seven as the IPI decreased from 310 to 0 ms. When the masker was delivered after the probe, the number of impulses progressively increased up to 45 at an IPI of +4 ms. The neuron’s rate^IPI function is shown in Fig. 1B. This neuron’s temporal window of masking is 2.1 ms with a start IPI at 31.1 ms and an end IPI at +1 ms (indicated by two un¢lled arrows in Fig. 1B). At the best IPI of 0 ms, masking was maximal such that the number of impulses was minimal (indicated by the ¢lled arrow in Fig. 1B). Fig. 2 shows variations in the number of impulses and response latency of another IC neuron determined at di¡erent IPIs with a masker delivered at the BF of low- (left panel) and high-£ank (right panel) inhibitory FTC. When a low-£ank masker was used, the neuron’s number of impulses decreased with decreasing IPI and reached a minimum at an IPI of 34 ms. It then increased to a maximum at an IPI of +2 ms (Fig. 2A, ¢lled circles refer to left ordinate). However, the neuron’s latency did not change when the IPI varied between 310 and 36 ms but increased to the longest latency at IPIs of 34 and 32 ms. The latency then decreased to the shortest at an IPI of 0 ms and remained unchanged with further variation in IPI (Fig. 2A, un¢lled circles refer to right ordinate). The temporal window for this masking e¡ect was 4.2 ms (between the two un¢lled arrows in Fig. 2A) and the best IPI was at 34 ms (solid arrow in Fig. 2A). The average temporal window and the best IPI for masking of responses of
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
143
Fig. 1. (A) The excitatory (un¢lled circles) and two inhibitory (hatched) FTCs of an IC neuron of the big brown bat, Eptesicus fuscus. Inhibitory FTCs were obtained using a forward masking paradigm. For this neuron, 4 ms FM pulses sweeping across the frequency bandwidth at 10 dB (un¢lled and ¢lled triangles) above the MTs of excitatory and inhibitory FTCs were used as a probe and a masker, respectively (probe: ¢lled arrow sweeping from 45 to 39 kHz; maskers: un¢lled arrows sweeping from 37 to 23 kHz across the low-£ank inhibitory FTC and from 64 to 49 kHz across the high-£ank inhibitory FTC). (B) A rate^IPI function showing variation in the neuron’s number of impulses under masking conditions when determined with a low-£ank masker at di¡erent IPIs. The masker was delivered prior to (negative IPI values), simultaneously (zero IPI), or after (positive IPI values) the probe. The horizontal dashed line indicates 50% of maximal response. The range of IPIs indicated between two un¢lled arrows is the temporal window for masking within which the number of impulses discharged to the probe is reduced at least 50%. The ¢lled arrow indicates the IPI of maximal masking at which the number of impulses was minimal. The response pattern (in peristimulus time histogram, PST) and the number of impulses (N) of this neuron obtained with the probe alone (used as a control, con) and under masking conditions at di¡erent IPIs are shown in the right panel. Envelopes of the probe and masker are shown at the bottom of each PST histogram.
53 IC neurons were 6.3 ms and 31.1 ms when determined with a low-£ank masker. When a high-£ank masker was used, the neuron’s number of impulses decreased to none at IPIs between 31 and 32 ms (Fig. 2B, solid arrows) before increasing again to a maximum at an IPI of +3 ms (Fig. 2B, ¢lled circles refer to left ordinate). The latency of this neuron lengthened with decreasing IPI. The latency, however, was not measurable at IPIs between 32 and 31 ms because of complete inhibition. The latency then decreased to the shortest at an IPI of 1 ms and remained unchanged with further variation in IPI (Fig. 2B, un¢lled circles refer to right ordinate). The temporal window for this masking e¡ect was 6.0 ms (between the two un¢lled arrows in Fig. 2B) and the best IPIs were between 32 and 31 ms (Fig. 2B, ¢lled arrows). The average temporal window and the best IPI for masking responses of 46 IC neurons were 7 ms and 31.8 ms when determined with a high-£ank masker. Fig. 2C,D show the neuron’s excitatory FTCs obtained with a probe alone (Fig. 2Ca) and under di¡erent masking conditions. The excitatory FTC elevated
and decreased with systematic decrease in IPI (Fig. 2Ca^c, Dg^i). The excitatory FTC was either the smallest (Fig. 2Cd) or not measurable (Fig. 2Dj) at the best IPI. The excitatory FTC then increased and shifted downward with further variation in IPI (Fig. 2Ce,f, Dk,l). While the above V- or U-shaped rate^IPI functions were observed for most (71%) IC neurons (e.g. Figs. 1B and 2A,B), three other types of rate^IPI functions were observed for the remaining neurons (Fig. 3). The rate^ IPI functions of 17% neurons only crossed the 50% maximal masking line (the dashed line in Fig. 3) once such that only the end IPI (un¢lled arrow) and the best IPI (¢lled arrow) were measurable (Fig. 3A). The rate^ IPI functions of 10% of IC neurons crossed the 50% maximal masking line three times so that they had two temporal windows and one best IPI for masking (Fig. 3B). The rate^IPI functions of 2% of IC neurons were always above the 50% maximal masking line such that a temporal window for masking could not be determined (Fig. 3C). Fig. 4 shows the distributions of start IPIs and end
HEARES 3935 2-7-02
144
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
Fig. 2. Rate^IPI and latency^IPI functions showing variation in the number of impulses (solid circles refer to left ordinate) and response latency (un¢lled circles refer to right ordinate) of an IC neuron determined under masking conditions using low-£ank (A) and high-£ank BF pulses as maskers (B) at di¡erent IPIs (abscissa). The neuron’s excitatory FTCs plotted at di¡erent IPIs are shown below in C and D (see Fig. 1 for legends and text for details).
IPIs of temporal windows for masking responses of IC neurons determined with a 4 ms probe and a low- or high-£ank masker. While the start IPIs distributed over a wide range of IPIs without showing a large number at any speci¢c IPI, the distribution of end IPIs peaked between IPIs of 1 and 3 ms (Fig. 4A,B, un¢lled vs. ¢lled bars). The start IPIs of most (69%) IC neurons were between 37 and 31 ms (average : 35.5 U 3.2 ms for a low-£ank masker, 36.7 U 2.9 ms for a high-£ank
masker, Fig. 4, un¢lled bars). The end IPIs of most (76%) neurons were between 0 and 3 ms (average: 0.8 U 1.8 ms for a low-£ank masker, 0.3 U 1.5 ms for a high-£ank masker, Fig. 4, ¢lled bars). 3.2. The e¡ect of masker duration on rate^IPI functions Rate^IPI functions of IC neurons varied with masker duration. Fig. 5 shows rate^IPI functions of a represen-
Fig. 3. Three other types of rate^IPI functions obtained from IC neurons. (A) The rate^IPI function crosses the 50% maximal response dashed line at one point only. (B) The rate^IPI function crosses the 50% maximal response dashed line at three di¡erent points. (C) The rate^IPI function never crosses the 50% maximal response dashed line. Un¢lled arrow: crossing points. Filled arrow: IPI of maximal masking.
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
145
tative IC neuron determined with a 4 ms probe and three di¡erent durations of low- and high-£ank maskers. The start IPI of the neuron’s rate^IPI function shifted right-ward with decreasing masker duration. As a result, the temporal window for masking became small at shorter masker duration (Fig. 5, 8 ms, between ¢lled arrows; 4 ms, between un¢lled arrows; 1.5 ms, between un¢lled arrows). When determined with a 1.5 ms low-£ank masker, inhibition was so weak that the number of impulses was never below the 50% maximal response line (e.g. Fig. 5A, un¢lled squares). Under this stimulation condition, the neuron’s temporal window for masking was not measurable. Table 1 shows the distribution of temporal windows for masking of responses of IC neurons determined with a 4 ms probe and three masker durations. It is clear that the temporal window for masking responses of most IC neurons was longer than 4 ms when determined with a 4 or 8 ms masker but was either shorter than 4 ms or not measurable when determined with a 1.5 ms masker. Fig. 5. Rate^IPI functions of an IC neuron obtained with a 4 ms BF probe (excitatory pulse, EP) and three di¡erent durations of low- (A) and high-£ank (B) BF maskers (inhibitory pulse, IP; 8 ms, un¢lled circles, curve a; 4 ms, ¢lled circles, curve b; 1.5 ms; un¢lled squares, curve c). The interval between each set of identical arrows represents the temporal window of masking. Note that the temporal window for masking increases with increasing masker duration. A temporal window was not available for curve Ac which did not cross the 50% maximal response dashed line.
3.3. The e¡ect of inter-pulse intensity di¡erence on masking
Fig. 4. Distribution of the start and end IPIs of temporal windows of IC neurons determined under masking conditions using a low(A) and a high-£ank (B) masker. Note that most start IPIs (un¢lled bars) were between 37 and 31 ms and most end IPIs (¢lled bars) were between 0 and 3 ms. The average start and end IPI and size of temporal window are indicated by the double-headed arrow. NA: temporal windows were not measurable. N: total number of neurons studied (see text for details).
We studied the e¡ect of inter-pulse intensity di¡erence on masking by plotting inhibitory FTCs of 36 neurons with di¡erent probe intensities. Inhibitory FTCs of most (31/36, 86%) neurons elevated and decreased with increasing probe intensity but the inhibitory BF was not a¡ected (Fig. 6A,D). At very strong probe intensity, masking e¡ect was so weak that inhibitory FTCs were not measurable. However, inhibitory FTCs of ¢ve (14%) other neurons were hardly a¡ected by an increase in probe intensity (Fig. 7A shaded, B,C solid vs. dashed). We also determined the e¡ect of probe intensity on masking by calculating the percent masking as a function of probe intensity. The percent masking was obtained by dividing the change in number of impulses under masking conditions by the control number of impulses obtained at each probe intensity. Percent masking of 29 (81%) neurons typically decreased with increasing probe intensity but increased with increasing masker intensity (Fig. 6B,C). However,
HEARES 3935 2-7-02
146
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
Fig. 6. (A,D) Low- and high-£ank inhibitory FTCs (dashed lines, ¢lled circles, and ¢lled squares) of two IC neurons determined with a 4 ms BF probe at 10 dB increments (¢lled triangles) above the excitatory MT. Note that inhibitory FTCs decrease with increasing probe intensity. An inhibitory FTC was not measurable when the probe was delivered at 40 dB above the excitatory MT (A). (B,C,E,F) Variation of percent masking with inter-pulse intensity di¡erences. The intensity of the probe and the masker was increased at 10 dB increments above excitatory and inhibitory MTs (abbreviated as EMT and IMT). Note that percent masking decreased with increasing probe intensity but increased with increasing masker intensity in one neuron (B,C). In another neuron, percent masking decreased with increasing probe intensity only when the masker intensity was low (Ea). At stronger masker intensities, masking was either 100% (Eb^d,Fc) or only decreased at strong probe intensity (Fa,b) (see text for details).
at very strong probe intensity (Fig. 6B,C when the probe intensity was at 40 dB above the excitatory MT), masking became extremely ine¡ective. There were two (5%) neurons whose percent masking decreased with increasing probe intensity only when the masker intensity was low (Fig. 6Ea). At stronger masker intensities, the percent masking was either 100% (Fig. 6Eb^d,Fc) or decreased only at stronger probe Intensities (Fig. 6Fa,b). For those ¢ve (14%) neurons whose inhibitory FTCs were hardly a¡ected by probe intensity (e.g. Fig. 7A^C), the percent masking was always 100% (data not shown).
We plotted excitatory FTCs of 12 neurons with different masker intensities. Although the BF was not affected, the excitatory FTCs of these neurons systematically elevated and decreased with increasing masker intensity (Fig. 7D^F).
4. Discussion 4.1. Temporal window for masking In auditory physiology, processing of auditory infor-
Table 1 The temporal window for masking of responses of IC neurons determined with a 4 ms probe and di¡erent durations of low- and high-£ank maskers Temporal window (ms)
Masker^probe duration (ms) 8^4
4^4
1.5^4
Low-£ank masker (n = 18) NA 0^2 2^4 4^8 s8
0 0 1 6 11
0 1 4 7 6
8^4
4^4
1.5^4
High-£ank masker (n = 15) 5 4 5 2 2
0 0 0 3 12
0 1 0 6 8
n: number of IC neurons studied. NA: temporal window was not measurable because of extremely weak masking e¡ect.
HEARES 3935 2-7-02
4 3 0 6 2
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
147
Fig. 7. (A) Low- and high-£ank inhibitory FTCs (hatched) of an IC neuron determined with a 4 ms BF probe at 10 dB above the excitatory MT (¢lled triangle). The inhibitory FTC was hardly a¡ected by increasing probe intensity (¢lled circles in B; ¢lled squares in C). For comparison, inhibitory FTCs in A are re-plotted in B and C (dashed). (D) The excitatory FTC and inhibitory FTCs (hatched) of another IC neuron. The neuron’s excitatory FTC (dashed, ¢lled diamonds, un¢lled squares, and ¢lled circles) progressively decreased with increasing masker intensity (¢lled and un¢lled triangles in D) above the low- (E) and high-£ank (F) inhibitory MTs. For comparison, the excitatory FTC in D was replotted in E and F (dashed).
mation carried by sounds has traditionally been explained by neural interactions of divergent and convergent projections within the ascending auditory system (Suga, 1997). During these neural interactions, a neuron’s response properties are determined by the arrival time and intensity of excitatory and inhibitory inputs at the neuron. This is demonstrated in the present study in which masking of excitatory responses of IC neurons was characterized by variation in the number of impulses, response latency, and size of excitatory FTCs
with IPIs and was a¡ected by inter-pulse intensity difference (Figs. 1, 2, 6 and 7). We have shown that the start IPIs of IC neurons were distributed over a wide range of negative IPIs while the end IPIs were mainly distributed within positive 1^3 ms IPIs (Fig. 4). Because a masker with longer duration produced a greater decrease in a neuron’s number of impulses than a masker with shorter duration at each IPI, this duration-dependent masking increased the temporal window of masking by a left-ward
Fig. 8. Hypothetical sketches showing possible synaptic integration underlying time- and intensity-dependent masking of responses of IC neurons observed in this study. The probe (un¢lled) and the masker (¢lled) are delivered either at di¡erent IPIs (A^C) and/or at di¡erent durations and intensities (D^F). E and I: envelope of EPSPs and IPSPs (see text for details).
HEARES 3935 2-7-02
148
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
shift of the start IPI (Fig. 5). All these observations indicate that masking was most e¡ective when a masker was presented prior to or simultaneously with a probe, similar to a previous study (Litovsky and Yin, 1998). In this study, we did not examine the e¡ect of probe duration on the temporal window for masking. This was because most IC neurons behave as duration-tuned neurons such that they discharge maximally to a speci¢c sound duration or a range of sound durations but the discharge drops more than 50% at other sound durations (Brand et al., 2000; Casseday et al., 1994; Chen, 1998; Ehrlich et al., 1997; Fuzessery and Hall, 1999; Jen and Schlegel, 1982; Jen and Feng, 1999; Jen and Zhou, 1999; Pinheiro et al., 1991; Zhou and Jen, 2001). For this reason, it would be di⁄cult to determine if a change in temporal window for masking determined at di¡erent probe durations was due to interactions of probe and masker at di¡erent IPIs or simply due to a neuron’s selectivity to the probe duration. 4.2. Inter-pulse intensity di¡erences on masking In agreement with a previous study in frogs (Fuzessery and Feng, 1983), we observed that inhibitory FTCs of most IC neurons elevated and decreased with increasing probe intensity (Fig. 6A,D). We also observed that the excitatory FTCs of most IC neurons elevated and decreased with increasing masker pulse intensity (Fig. 7D^F). This intensity-dependent masking is conceivably a result of variation in excitation-to-inhibition ratio in which a masker needs to be stronger than a probe to produce a masking e¡ect. This is substantiated by the fact that percent masking decreased with increasing probe intensity but increased with increasing masker intensity (Fig. 6B,C). Under whole-cell patch-clamp recording conditions, a previous study showed that pulses with frequencies below and above the excitatory BF of bat IC neurons elicited short-latency and long-latency inhibitory postsynaptic potentials which inhibit the responses of IC neurons (Covey et al., 1996). The amplitude of these inhibitory synaptic potentials increased with increasing sound intensity. These synaptic mechanisms may explain our ¢nding that increasing masker intensity elevated and decreased excitatory FTCs (Fig. 7D^F). Masking responses of ¢ve IC neurons were intensityindependent such that increasing probe intensity did not a¡ect high- and low-£ank inhibitory FTCs (Fig. 7A^C). This intensity-independent masking has been previously reported for low-£ank inhibitory FTCs of anuran midbrain neurons (Fuzessery and Feng, 1983). This intensity-independent masking might be due to the fact that increasing excitation by the probe did not overcome the powerful masking e¡ect produced by the masker to these neurons.
In this study, we observed that presentation of a masker typically decreased a neuron’s excitatory response and the percent masking could be as large as 100% (i.e. complete masking). However, when the probe intensity was very strong (e.g. Fig. 6B,C,Ea), responses of some IC neurons were increased under masking conditions such that the percent masking became negative. Because the increased response of these neurons under masking conditions was always less than 20%, we do not know if this observation was due to response £uctuation or a true facilitation. 4.3. Possible mechanisms underlying time- and intensity-dependent masking In this study, we used a probe to elicit excitatory responses from a recorded IC neuron. We then demonstrated that the neuron’s excitatory response could be modi¢ed by the duration and intensity of a masker when delivered within a certain temporal window (Figs. 1^7). This modi¢cation of excitatory response of an IC neuron under di¡erent masking conditions was most likely a result of synaptic integration of excitation and inhibition at the IC neuron. Because the arrival time, amplitude and duration of excitatory and inhibitory postsynaptic potentials (i.e. EPSPs and IPSPs) determine the response of a postsynaptic neuron during synaptic transmission, the time- and intensitydependent masking of responses of IC neurons could conceivably be explained by integration of synaptic potentials as shown in Fig. 8. For example, when a masker was delivered prior to a probe at a large IPI, an IPSP would diminish before the arrival of an EPSP (Fig. 8A). As a result, a neuron’s excitatory response would not be a¡ected by the masker (e.g. Fig. 1B when IPI was between 310 and 37 ms). As both IPSPs and EPSPs overlapped at decreasing IPI (Fig. 8B), a neuron’s excitatory response would decrease (e.g. Fig. 1 when IPI was between 34 and +2 ms). However, when the masker was presented after the probe, the early arrival of an EPSP would not be a¡ected by the late arrival of an IPSP (Fig. 8C). Under this masking condition, the neuron’s response would not be a¡ected by the masker (e.g. Fig. 1 when IPI was larger than +3 ms). When a long-duration masker was presented prior to a probe at a speci¢c IPI, the long-duration IPSP would overlap with the EPSP to decrease excitatory response (Fig. 8D). However, when a short-duration masker was presented at the same IPI, overlapping of two PSPs would not occur (Fig. 8E). To produce masking, the short-duration masker would have to be presented prior to the probe at shorter IPIs. As a result, the temporal window for masking would become smaller for a shortduration masker (e.g. Fig. 5).
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
When a masker was presented prior to a probe at di¡erent inter-pulse intensity di¡erences, a neuron’s excitatory response would be determined by a summation of IPSPs and EPSPs (Fig. 8F). When the intensity of probe relative to that of masker progressively increased, the IPSP produced by the masker would become less e¡ective in reducing the EPSP. For this reason, the degree of masking would decrease with an increase in probe intensity (Fig. 6B,C). In summary, we have shown that masking of probeelicited responses of IC neurons occurred when a masker was presented within a certain temporal window. Masking increased with increasing masker duration and intensity but decreased with increasing probe intensity. The time- and intensity-dependent masking e¡ect may be explained on the basis of di¡erent arrival times and amplitudes of excitatory and inhibitory inputs at IC neurons. Future work using intracellular recording is needed to determine the synaptic mechanisms underlying these observations.
Acknowledgements We thank two anonymous reviewers for commenting on the earlier version of the manuscript. This work is supported by a research grant from the National Science Foundation (NSF IBN 9907610), a grant from the Research Council (URC 01-012 Jen) and a matching fund from the College of Arts and Sciences of the University of Missouri-Columbia.
References Aitkin, L., 1986. The Auditory Midbrain: Structure and Function in the Central Auditory Pathway. Humana Press, Clifton, NJ. Brand, A., Urban, R., Grothe, B., 2000. Duration tuning in the mouse auditory midbrain. J. Neurophysiol. 84, 1790^1799. Casseday, J.H., Covey, E., 1996. A neuroethological theory of the operation of the inferior colliculus. Brain Behav. Evol. 47, 311^ 336. Casseday, J.H., Ehrlich, D., Covey, E., 1994. Neural tuning for sound duration: role of inhibitory mechanisms in the inferior colliculus. Science 264, 847^850. Chen, G.D., 1998. E¡ects of stimulus duration on responses of neurons in the chinchilla inferior colliculus. Hear. Res. 122, 142^150. Covey, E., Julie, A.K., Casseday, J.H., 1996. Whole-cell patch-clamp recording reveals subthreshold sound-evoked postsynaptic currents in the inferior colliculus of awake bats. J. Neurosci. 16, 3009^3018. Ehret, G., Merzenich, MM., 1988. Complex sound analysis (frequency resolution, ¢ltering and spectral integration) by single units of the inferior colliculus of the cat. Brain Res. Rev. 13, 139^163. Ehrlich, D., Casseday, J.H., Covey, E., 1997. Neural tuning to sound duration in the inferior colliculus of the big brown bat, Eptesicus fuscus. J. Neurophysiol. 77, 2360^2372. Eysel, U.T., Worgotter, F., Pape, H.C., 1987. Local cortical lesions
149
abolish lateral inhibition at direction selective cells in cat visual cortex. Exp. Brain Res. 68, 606^612. Fuzessery, Z.M., Feng, A.S., 1983. Frequency selectivity in the anuran medulla: excitatory and inhibitory tuning properties of single neurons in the dorsal medullary and superior olivary nuclei. J. Comp. Physiol. A 150, 107^119. Fuzessery, Z.M., Hall, J.C., 1999. Sound duration selectivity in the pallid bat inferior colliculus. Hear. Res. 137, 137^154. Herbert, H., Ascho¡, A., Ostwald, J., 1991. Topography of projections from the auditory cortex to the inferior colliculus in the rat. J. Comp. Neurol. 304, 103^122. Hu¡man, R.F., Henson, O.W., Jr., 1990. The descending auditory pathways and acoustico-motor systems: connections with the inferior colliculus. Brain Res. Rev. 15, 295^323. Jen, P.H.S., Feng, R.B., 1999. Bicuculline application a¡ects discharge pattern and pulse-duration tuning characteristics of bat inferior collicular neurons. J. Comp. Physiol. A 184, 185^194. Jen, P.H.S., Schlegel, P.A., 1982. Auditory physiological properties of the neurons in the inferior colliculus of the big brown bat, Eptesicus fuscus. J. Comp. Physiol. A 147, 351^363. Jen, P.H.S., Zhang, J.P., 1999. Corticofugal regulation of excitatory and inhibitory frequency tuning curves of bat inferior collicular neurons. Brain Res. 841, 184^188. Jen, P.H.S., Zhou, X.M., 1999. Temporally patterned sound trains a¡ect duration tuning characteristics of bat inferior collicular neurons. J. Comp. Physiol. A 185, 471^478. Jen, P.H.S., Sun, X.D., Chen, D.M., Teng, H.B., 1987. Auditory space representation in the inferior colliculus of the FM bat, Eptesicus fuscus. Brain Res. 419, 7^18. Katsuki, Y., Sumi, T., Uchiyama, H., Watanabe, T., 1958. Electric responses of auditory neurons in cat to sound stimulation. J. Neurophysiol. 21, 569^588. Kelly, D.H., 1973. Lateral inhibition in human color mechanisms. J. Physiol. 228, 55^72. Litovsky, R.Y., Yin, T.C., 1998. Physiological studies of the precedence e¡ect in the inferior colliculus of the cat. II. Neural mechanisms. J. Neurophysiol. 80, 1302^1316. Lu, Y., Jen, P.H.S., 2001. GABAergic and glycinergic neural inhibition in excitatory frequency tuning of bat inferior collicular neurons. Exp. Brain Res. 14, 331^339. Lu, Y., Jen, P.H.S., Zheng, Q.Y., 1997. GABAergic disinhibition changes the recovery cycle of bat inferior collicular neurons. J. Comp. Physiol. A 181, 331^341. Mountcastle, V.B., Powell, T.P.S., 1959. Neural mechanisms subserving cutaneous sensitivity, with special reference to the role of afferent inhibition in sensory perception and discrimination. Bull. Johns Hopkins Hosp. 105, 201^232. Pinheiro, A.D., Wu, M., Jen, P.H.S., 1991. Encoding repetition rate and duration in the inferior colliculus of the big brown bat, Eptesicus fuscus. J. Comp. Physiol. A 169, 69^85. Pollak, G.D., Casseday, J.H., 1989. The Neural Basis of Echolocation in Bats. Springer-Verlag, Berlin. Saldan‹a, E., Feliciano, M., Mugnaini, E., 1996. Distribution of descending projections from primary auditory neocortex to inferior colliculus mimics the topography of intracollicular projections. J. Comp. Neurol. 371, 15^40. Singer, W., Creutzfeldt, O.D., 1970. Reciprocal lateral inhibition of on- and o¡-center neurons in the lateral geniculate body of the cat. Exp. Brain Res. 10, 311^330. Suga, N., 1995. Sharpening of frequency tuning by inhibition in the central auditory system: tribute to Yasuji Katsuki. Neurosci. Res. 21, 287^299. Suga, N., 1997. Parallel-hierarchical processing of complex sounds for specialized auditory function. In: Crocker, M.J. (Ed.), Encyclopedia of Acoustics. Wiley, New York, pp. 1409^1418.
HEARES 3935 2-7-02
150
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
Suga, N., Tsuzuki, K., 1985. Inhibition and level-tolerant frequency tuning in the auditory cortex of the mustached bat. J. Neurophysiol. 53, 1109^1145. Suga, N., Zhang, Y., Yan, J., 1997. Sharpening of frequency tuning by inhibition in the thalamic auditory nucleus of the mustached bat. J. Neurophysiol. 77, 2098^2114. Winer, J.A., Larue, D.T., Diehl, J.J., Hefti, B.J., 1998. Auditory cortical projections to the cat inferior colliculus. J. Comp. Neurol. 400, 147^174.
Zhou, X.M., Jen, P.H.S., 2000. Neural inhibition sharpens auditory spatial sensitivity of bat inferior collicular neurons. J. Comp. Physiol. A 186, 389^398. Zhou, X.M., Jen, P.H.S., 2001. The e¡ect of sound intensity on duration tuning characteristics of bat inferior collicular neurons. J. Comp. Physiol. A 187, 63^73.
HEARES 3935 2-7-02
Interaction of excitation and inhibition in inferior collicular neurons of the big brown bat, Eptesicus fuscus Yong Lu 1 , Philip H.-S. Jen
Division of Biological Sciences, University of Missouri-Columbia, Columbia, MO 65211, USA Received 22 January 2002; accepted 9 April 2002
Abstract Neurons in the inferior colliculus (IC) of the midbrain receive convergent excitatory and inhibitory inputs from both lower and higher auditory nuclei. Interaction of these two opposing inputs shapes different response properties of IC neurons. In this study, we examined this interaction of excitation and inhibition in IC neurons using a probe (excitatory pulse) and a masker (inhibitory pulse) under different stimulation conditions. Inhibition of probe-elicited responses by a masker, i.e. masking, occurred when the masker was presented at certain inter-pulse intervals (the temporal window) in relation to the probe. At the best inter-pulse interval, masking was maximal such that a neuron had the minimal number of impulses, the longest response latency, and the smallest excitatory frequency tuning curve. The temporal window for masking expanded with increasing masker duration. The inhibition decreased with increasing probe intensity but increased with increasing masker intensity. Increasing masker intensity also produced progressive shrinkage in excitatory frequency tuning curves. Similarly, increasing probe intensity produced progressive shrinkage of inhibitory frequency tuning curves. Possible mechanisms underlying the time and intensity dependence of inhibition are discussed. / 2002 Elsevier Science B.V. All rights reserved. Key words: Bat; Inferior colliculus; Intensity; Temporal window; Masking
1. Introduction In sensory signal processing, neural inhibition plays an important role in shaping response properties of sensory neurons. For example, neural inhibition contributes to color contrast and edge contrast in the retina, the lateral geniculate nucleus and the visual cortex (Eysel et al., 1987; Kelly, 1973; Singer and Creutzfeldt, 1970). In the somatosensory system, neural inhibition de¢nes the excitatory areas of peripheral and central somatosensory neurons (Mountcastle and Powell, * Corresponding author. Tel.: +1 (573) 882-7479; Fax: +1 (573) 884-5020. E-mail address: [email protected] (P.H.-S. Jen). 1 Present address: Department of Otolaryngology-HNS, University of Washington, Seattle, WA 98195, USA.
Abbreviations: BF, best frequency; EPSP, excitatory postsynaptic potential; FTC, frequency tuning curve; IC, inferior colliculus; IPI, inter-pulse interval; IPSP, inhibitory postsynaptic potential; MT, minimum threshold; PST, peristimulus time
1959). In the auditory system, Katsuki et al. (1958) suggested that increasing sharpness of excitatory frequency tuning curves (FTCs) along the central auditory pathway was due to synaptic inhibition from neighboring auditory neurons. Since then, many studies have examined the contribution of neural inhibition in auditory signal processing in the frequency domain using a forward masking paradigm. These studies showed that many central auditory neurons have inhibitory areas on one or both frequency £anks of their excitatory FTCs (Ehret and Merzenich, 1988; Fuzessery and Feng, 1983; Jen and Zhang, 1999; Lu and Jen, 2001; Suga and Tsuzuki, 1985; Suga et al., 1997; Zhou and Jen, 2000). These inhibitory areas contribute to sharpening of excitatory FTCs (see review by Suga, 1995). All these examples are evidence that sensory signal processing is determined by the interplay between excitatory and inhibitory inputs regardless of sensory modality. In the ascending auditory pathway, the central nucleus of the inferior colliculus (IC) receives and integrates excitatory and inhibitory inputs from many lower
0378-5955 / 02 / $ ^ see front matter / 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 2 ) 0 0 4 5 7 - 4
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
auditory nuclei (Aitkin, 1986; Casseday and Covey, 1996; Pollak and Casseday, 1989) and from the auditory cortex (Herbert et al., 1991; Hu¡man and Henson, 1990; Saldan‹a et al., 1996; Winer et al., 1998). For this reason, temporal and spectral signal processing of IC neurons is a¡ected by the timing and intensity of excitatory and inhibitory inputs. Presumably, inhibition only occurs when both excitatory and inhibitory inputs arrive at IC neurons within a certain time window. Furthermore, inhibitory inputs with stronger intensity and longer duration should be more e¡ective in producing inhibition of auditory response of IC neurons than inhibitory inputs with weaker intensity and shorter duration. The main objective of this study was to determine how the interaction of excitation and inhibition might shape di¡erent response properties of IC neurons. To achieve this objective, we examined how the response of IC neurons elicited by a probe (an excitatory pulse) might be inhibited by a masker (an inhibitory pulse) under di¡erent stimulation conditions. Speci¢cally, we examined variation in the number of impulses, response latency, and excitatory FTC of IC neurons with interpulse interval (IPI). We then studied the e¡ect of masker duration on rate^IPI functions. We also studied how inter-pulse intensity di¡erence might a¡ect excitatory and inhibitory FTCs of IC neurons.
2. Materials and methods Seven big brown bats (four males, three females, 24^30 g body weight, b.w.) were used for this study. The surgical procedures have been described in previous studies (Jen et al., 1987). Brie£y, the £at head of a 1.8 cm nail was glued onto the exposed skull of each Nembutal-anesthetized bat (45^50 mg/kg b.w.) with acrylic glue and dental cement 1 or 2 days before the recording session. Exposed tissue was treated with an antibiotic (Neosporin) to prevent in£ammation. During recording, the bat was administered the neuroleptanalgesic, Innovar-Vet (fentanyl 0.08 mg/kg b.w., droperidol 4 mg/kg b.w.) and was placed inside a bat holder (made of wire mesh) which was suspended in an elastic sling. After the bat’s head was immobilized, small holes were bored in the skull above the IC for insertion of micropipette electrodes ¢lled with 3 M KCl to record sound-activated IC neurons. The bat’s head was oriented such that its eye^snout line was pointed to 0‡ in azimuth and 0‡ in elevation relative to the position of the loudspeaker. Recordings were conducted inside a double-walled sound-proof room (temperature 28^30‡C) with its ceiling and inside walls covered with 3 inch convoluted polyurethane foam to reduce echoes. A local anesthetic
141
(lidocaine) was applied to the open wound area. The recording depth was read from the scale of a microdrive (David-Kopf). A common indi¡erent electrode (silver wire) was placed at the nearby temporal muscles. The experiments were conducted in compliance with NIH publication No. 85-23, ‘Principles of Laboratory Animal Care’ and with the approval of the Institutional Animal Care and Use Committee (#1438) of the University of Missouri-Columbia. As in previous studies (Jen et al., 1987), continuous sine waves from an oscillator (KH model 1200) were formed into 4 ms pulses (0.5 ms rise^decay times, at 2 pps) by a homemade pulse generator (electronic switch) driven by a stimulator (Grass S88). These pulses were ampli¢ed after passing through a decade attenuator (HP 350D) before being fed to a small condenser loudspeaker (AKG model CK 50, 1.5 cm diameter, 1.2 g). The loudspeaker was placed 23 cm away from the bat and 30‡ contralateral to the recording site. Calibration of the loudspeaker was conducted with a Bru«el and Kjaer 1/4 inch microphone (4135) placed at the position where the bat’s head would be during recording. The output was expressed in dB SPL referred to 20 WPa root mean square. A frequency^response curve of the loudspeaker was plotted to determine the maximal available stimulus intensity at each frequency. Frequency-modulated (FM) pulses were generated by means of a ramp generator. Upon isolation of an acoustically evoked IC neuron with 4 ms pulses, its excitatory best frequency (BF) and minimum threshold (MT) were determined by changing the sound frequency and intensity. At the MT, the neuron, on average, responded with 50% probability to excitatory BF pulses. A probe was then delivered at the excitatory BF and 10 dB above the excitatory MT to elicit excitatory response of the isolated neuron. An inhibitory FTC was then plotted with combinations of frequency and intensity of a masker that decreased the neuron’s excitatory response by at least 20%. Because big brown bats use downward-sweep FM pulses for orientation, we determined the temporal window of masking using FM pulses sweeping downward across the 10 dB bandwidth of excitatory and inhibitory FTCs as a probe and a masker, respectively (i.e. the bandwidth of the excitatory and inhibitory FTCs at 10 dB above the MT, see Fig. 1A). However, if a neuron did not respond or responded poorly to FM pulses, BF pulses delivered at 10 dB above excitatory and inhibitory MTs were used as the probe and masker. The masker was delivered 10 ms before (expressed as negative IPIs), simultaneously (expressed as zero IPI), and 4 ms after (expressed as positive IPIs) the probe at 1 ms increments. When a neuron had low- and high-£ank inhibitory FTCs, BF or FM pulses of both inhibitory FTCs were sequentially used as the masker. The neu-
HEARES 3935 2-7-02
142
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
ron’s number of impulses and latency in response to excitatory FM or BF pulses at 10 dB above the MT were obtained at each IPI. These data were then plotted against the IPI to obtain the neuron’s rate^IPI and latency^IPI functions. The neuron’s temporal window of masking was de¢ned as the range of IPIs within which a neuron’s number of impulses was decreased by a masker by at least 50% of the maximal response. The two IPIs corresponding to the beginning and end of the temporal window were arbitrarily de¢ned as the start IPI and end IPI of the temporal window. Within the temporal window, there was a best IPI or a range of IPIs at which masking was maximal (see Fig. 1B). We then studied the e¡ect of masker duration on the temporal window of masking with a 4 ms probe and three di¡erent masker durations (1.5, 4, and 8 ms). To determine the e¡ect of inter-pulse intensity di¡erences on masking of responses of IC neurons, we plotted the inhibitory FTCs at di¡erent probe intensities. Speci¢cally, a series of inhibitory FTCs was plotted when 4 ms BF probe was delivered at the best IPI and at 10 dB increments above the excitatory MT. The number of impulses was also obtained when the intensity of probe and masker was increased in 10 dB increments above the excitatory and inhibitory MTs. Using the number of impulses elicited by the probe alone as a control, the percent masking was then calculated by dividing the change in the number of impulses under masking conditions by the control. Whenever possible, a series of excitatory FTCs was plotted with the masker delivered at the best IPI and at 10 dB increments above the inhibitory MT. Recorded action potentials were ampli¢ed, band-pass ¢ltered (Krohn-Hite 3500), and fed through a window discriminator (WPI 121) before being sent to an oscilloscope (Tektronix 5111) and an audio monitor (Grass AM6). They were then sent to a computer (Gateway 2000, 486) for acquisition of peristimulus time (PST) histograms (bin width: 500 Ws, sampling period : 300 ms) to 16 stimulus presentations. The total number of impulses in each PST histogram was used to quantify a neuron’s response under each stimulus condition.
3. Results Masking of auditory response was studied in 78 IC neurons (48 neurons studied with FM pulses and 30 neurons with BF pulses) which were isolated at depths of 346^1992 Wm. The excitatory BF ranged from 17.1 to 91.3 kHz and excitatory MT ranged from 10 to 85 dB SPL. When stimulated with 4 ms BF or FM pulses, these neurons either discharged one to three impulses (phasic responders, 43, 55%), four to seven impulses (phasic bursters, 26, 33%) or discharged
impulses throughout the entire 4 ms pulses (tonic responders, 9, 12%), similar to what we have reported earlier (Jen and Schlegel, 1982; Jen and Feng, 1999; Lu et al., 1997). Inhibitory FTCs were obtained at both the low and the high £ank (49, 63%) (Fig. 1A), or only at the low £ank (17, 22%) or high £ank (12, 15%) of the excitatory FTCs. Because the e¡ect of IPI, pulse duration and intensity on masking of response of IC neurons determined with FM pulses was similar to that determined with BF pulses, all data are combined together in the following presentation. 3.1. The e¡ect of IPI on masking Fig. 1A shows an IC neuron which had inhibitory FTCs (hatched) at the low and high £ank of its V-shaped excitatory FTC (un¢lled circles). Both inhibitory MTs were higher than excitatory MT but low£ank inhibitory MT was lower than high-£ank inhibitory MT. As shown in the right panel, this neuron discharged 42 impulses to a probe sweeping across the 10 dB bandwidth of excitatory FTC (used as a control, con). When a masker sweeping across the 10 dB bandwidth of low-£ank inhibitory FTC was delivered prior to the probe, the number of impulses progressively decreased to seven as the IPI decreased from 310 to 0 ms. When the masker was delivered after the probe, the number of impulses progressively increased up to 45 at an IPI of +4 ms. The neuron’s rate^IPI function is shown in Fig. 1B. This neuron’s temporal window of masking is 2.1 ms with a start IPI at 31.1 ms and an end IPI at +1 ms (indicated by two un¢lled arrows in Fig. 1B). At the best IPI of 0 ms, masking was maximal such that the number of impulses was minimal (indicated by the ¢lled arrow in Fig. 1B). Fig. 2 shows variations in the number of impulses and response latency of another IC neuron determined at di¡erent IPIs with a masker delivered at the BF of low- (left panel) and high-£ank (right panel) inhibitory FTC. When a low-£ank masker was used, the neuron’s number of impulses decreased with decreasing IPI and reached a minimum at an IPI of 34 ms. It then increased to a maximum at an IPI of +2 ms (Fig. 2A, ¢lled circles refer to left ordinate). However, the neuron’s latency did not change when the IPI varied between 310 and 36 ms but increased to the longest latency at IPIs of 34 and 32 ms. The latency then decreased to the shortest at an IPI of 0 ms and remained unchanged with further variation in IPI (Fig. 2A, un¢lled circles refer to right ordinate). The temporal window for this masking e¡ect was 4.2 ms (between the two un¢lled arrows in Fig. 2A) and the best IPI was at 34 ms (solid arrow in Fig. 2A). The average temporal window and the best IPI for masking of responses of
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
143
Fig. 1. (A) The excitatory (un¢lled circles) and two inhibitory (hatched) FTCs of an IC neuron of the big brown bat, Eptesicus fuscus. Inhibitory FTCs were obtained using a forward masking paradigm. For this neuron, 4 ms FM pulses sweeping across the frequency bandwidth at 10 dB (un¢lled and ¢lled triangles) above the MTs of excitatory and inhibitory FTCs were used as a probe and a masker, respectively (probe: ¢lled arrow sweeping from 45 to 39 kHz; maskers: un¢lled arrows sweeping from 37 to 23 kHz across the low-£ank inhibitory FTC and from 64 to 49 kHz across the high-£ank inhibitory FTC). (B) A rate^IPI function showing variation in the neuron’s number of impulses under masking conditions when determined with a low-£ank masker at di¡erent IPIs. The masker was delivered prior to (negative IPI values), simultaneously (zero IPI), or after (positive IPI values) the probe. The horizontal dashed line indicates 50% of maximal response. The range of IPIs indicated between two un¢lled arrows is the temporal window for masking within which the number of impulses discharged to the probe is reduced at least 50%. The ¢lled arrow indicates the IPI of maximal masking at which the number of impulses was minimal. The response pattern (in peristimulus time histogram, PST) and the number of impulses (N) of this neuron obtained with the probe alone (used as a control, con) and under masking conditions at di¡erent IPIs are shown in the right panel. Envelopes of the probe and masker are shown at the bottom of each PST histogram.
53 IC neurons were 6.3 ms and 31.1 ms when determined with a low-£ank masker. When a high-£ank masker was used, the neuron’s number of impulses decreased to none at IPIs between 31 and 32 ms (Fig. 2B, solid arrows) before increasing again to a maximum at an IPI of +3 ms (Fig. 2B, ¢lled circles refer to left ordinate). The latency of this neuron lengthened with decreasing IPI. The latency, however, was not measurable at IPIs between 32 and 31 ms because of complete inhibition. The latency then decreased to the shortest at an IPI of 1 ms and remained unchanged with further variation in IPI (Fig. 2B, un¢lled circles refer to right ordinate). The temporal window for this masking e¡ect was 6.0 ms (between the two un¢lled arrows in Fig. 2B) and the best IPIs were between 32 and 31 ms (Fig. 2B, ¢lled arrows). The average temporal window and the best IPI for masking responses of 46 IC neurons were 7 ms and 31.8 ms when determined with a high-£ank masker. Fig. 2C,D show the neuron’s excitatory FTCs obtained with a probe alone (Fig. 2Ca) and under di¡erent masking conditions. The excitatory FTC elevated
and decreased with systematic decrease in IPI (Fig. 2Ca^c, Dg^i). The excitatory FTC was either the smallest (Fig. 2Cd) or not measurable (Fig. 2Dj) at the best IPI. The excitatory FTC then increased and shifted downward with further variation in IPI (Fig. 2Ce,f, Dk,l). While the above V- or U-shaped rate^IPI functions were observed for most (71%) IC neurons (e.g. Figs. 1B and 2A,B), three other types of rate^IPI functions were observed for the remaining neurons (Fig. 3). The rate^ IPI functions of 17% neurons only crossed the 50% maximal masking line (the dashed line in Fig. 3) once such that only the end IPI (un¢lled arrow) and the best IPI (¢lled arrow) were measurable (Fig. 3A). The rate^ IPI functions of 10% of IC neurons crossed the 50% maximal masking line three times so that they had two temporal windows and one best IPI for masking (Fig. 3B). The rate^IPI functions of 2% of IC neurons were always above the 50% maximal masking line such that a temporal window for masking could not be determined (Fig. 3C). Fig. 4 shows the distributions of start IPIs and end
HEARES 3935 2-7-02
144
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
Fig. 2. Rate^IPI and latency^IPI functions showing variation in the number of impulses (solid circles refer to left ordinate) and response latency (un¢lled circles refer to right ordinate) of an IC neuron determined under masking conditions using low-£ank (A) and high-£ank BF pulses as maskers (B) at di¡erent IPIs (abscissa). The neuron’s excitatory FTCs plotted at di¡erent IPIs are shown below in C and D (see Fig. 1 for legends and text for details).
IPIs of temporal windows for masking responses of IC neurons determined with a 4 ms probe and a low- or high-£ank masker. While the start IPIs distributed over a wide range of IPIs without showing a large number at any speci¢c IPI, the distribution of end IPIs peaked between IPIs of 1 and 3 ms (Fig. 4A,B, un¢lled vs. ¢lled bars). The start IPIs of most (69%) IC neurons were between 37 and 31 ms (average : 35.5 U 3.2 ms for a low-£ank masker, 36.7 U 2.9 ms for a high-£ank
masker, Fig. 4, un¢lled bars). The end IPIs of most (76%) neurons were between 0 and 3 ms (average: 0.8 U 1.8 ms for a low-£ank masker, 0.3 U 1.5 ms for a high-£ank masker, Fig. 4, ¢lled bars). 3.2. The e¡ect of masker duration on rate^IPI functions Rate^IPI functions of IC neurons varied with masker duration. Fig. 5 shows rate^IPI functions of a represen-
Fig. 3. Three other types of rate^IPI functions obtained from IC neurons. (A) The rate^IPI function crosses the 50% maximal response dashed line at one point only. (B) The rate^IPI function crosses the 50% maximal response dashed line at three di¡erent points. (C) The rate^IPI function never crosses the 50% maximal response dashed line. Un¢lled arrow: crossing points. Filled arrow: IPI of maximal masking.
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
145
tative IC neuron determined with a 4 ms probe and three di¡erent durations of low- and high-£ank maskers. The start IPI of the neuron’s rate^IPI function shifted right-ward with decreasing masker duration. As a result, the temporal window for masking became small at shorter masker duration (Fig. 5, 8 ms, between ¢lled arrows; 4 ms, between un¢lled arrows; 1.5 ms, between un¢lled arrows). When determined with a 1.5 ms low-£ank masker, inhibition was so weak that the number of impulses was never below the 50% maximal response line (e.g. Fig. 5A, un¢lled squares). Under this stimulation condition, the neuron’s temporal window for masking was not measurable. Table 1 shows the distribution of temporal windows for masking of responses of IC neurons determined with a 4 ms probe and three masker durations. It is clear that the temporal window for masking responses of most IC neurons was longer than 4 ms when determined with a 4 or 8 ms masker but was either shorter than 4 ms or not measurable when determined with a 1.5 ms masker. Fig. 5. Rate^IPI functions of an IC neuron obtained with a 4 ms BF probe (excitatory pulse, EP) and three di¡erent durations of low- (A) and high-£ank (B) BF maskers (inhibitory pulse, IP; 8 ms, un¢lled circles, curve a; 4 ms, ¢lled circles, curve b; 1.5 ms; un¢lled squares, curve c). The interval between each set of identical arrows represents the temporal window of masking. Note that the temporal window for masking increases with increasing masker duration. A temporal window was not available for curve Ac which did not cross the 50% maximal response dashed line.
3.3. The e¡ect of inter-pulse intensity di¡erence on masking
Fig. 4. Distribution of the start and end IPIs of temporal windows of IC neurons determined under masking conditions using a low(A) and a high-£ank (B) masker. Note that most start IPIs (un¢lled bars) were between 37 and 31 ms and most end IPIs (¢lled bars) were between 0 and 3 ms. The average start and end IPI and size of temporal window are indicated by the double-headed arrow. NA: temporal windows were not measurable. N: total number of neurons studied (see text for details).
We studied the e¡ect of inter-pulse intensity di¡erence on masking by plotting inhibitory FTCs of 36 neurons with di¡erent probe intensities. Inhibitory FTCs of most (31/36, 86%) neurons elevated and decreased with increasing probe intensity but the inhibitory BF was not a¡ected (Fig. 6A,D). At very strong probe intensity, masking e¡ect was so weak that inhibitory FTCs were not measurable. However, inhibitory FTCs of ¢ve (14%) other neurons were hardly a¡ected by an increase in probe intensity (Fig. 7A shaded, B,C solid vs. dashed). We also determined the e¡ect of probe intensity on masking by calculating the percent masking as a function of probe intensity. The percent masking was obtained by dividing the change in number of impulses under masking conditions by the control number of impulses obtained at each probe intensity. Percent masking of 29 (81%) neurons typically decreased with increasing probe intensity but increased with increasing masker intensity (Fig. 6B,C). However,
HEARES 3935 2-7-02
146
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
Fig. 6. (A,D) Low- and high-£ank inhibitory FTCs (dashed lines, ¢lled circles, and ¢lled squares) of two IC neurons determined with a 4 ms BF probe at 10 dB increments (¢lled triangles) above the excitatory MT. Note that inhibitory FTCs decrease with increasing probe intensity. An inhibitory FTC was not measurable when the probe was delivered at 40 dB above the excitatory MT (A). (B,C,E,F) Variation of percent masking with inter-pulse intensity di¡erences. The intensity of the probe and the masker was increased at 10 dB increments above excitatory and inhibitory MTs (abbreviated as EMT and IMT). Note that percent masking decreased with increasing probe intensity but increased with increasing masker intensity in one neuron (B,C). In another neuron, percent masking decreased with increasing probe intensity only when the masker intensity was low (Ea). At stronger masker intensities, masking was either 100% (Eb^d,Fc) or only decreased at strong probe intensity (Fa,b) (see text for details).
at very strong probe intensity (Fig. 6B,C when the probe intensity was at 40 dB above the excitatory MT), masking became extremely ine¡ective. There were two (5%) neurons whose percent masking decreased with increasing probe intensity only when the masker intensity was low (Fig. 6Ea). At stronger masker intensities, the percent masking was either 100% (Fig. 6Eb^d,Fc) or decreased only at stronger probe Intensities (Fig. 6Fa,b). For those ¢ve (14%) neurons whose inhibitory FTCs were hardly a¡ected by probe intensity (e.g. Fig. 7A^C), the percent masking was always 100% (data not shown).
We plotted excitatory FTCs of 12 neurons with different masker intensities. Although the BF was not affected, the excitatory FTCs of these neurons systematically elevated and decreased with increasing masker intensity (Fig. 7D^F).
4. Discussion 4.1. Temporal window for masking In auditory physiology, processing of auditory infor-
Table 1 The temporal window for masking of responses of IC neurons determined with a 4 ms probe and di¡erent durations of low- and high-£ank maskers Temporal window (ms)
Masker^probe duration (ms) 8^4
4^4
1.5^4
Low-£ank masker (n = 18) NA 0^2 2^4 4^8 s8
0 0 1 6 11
0 1 4 7 6
8^4
4^4
1.5^4
High-£ank masker (n = 15) 5 4 5 2 2
0 0 0 3 12
0 1 0 6 8
n: number of IC neurons studied. NA: temporal window was not measurable because of extremely weak masking e¡ect.
HEARES 3935 2-7-02
4 3 0 6 2
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
147
Fig. 7. (A) Low- and high-£ank inhibitory FTCs (hatched) of an IC neuron determined with a 4 ms BF probe at 10 dB above the excitatory MT (¢lled triangle). The inhibitory FTC was hardly a¡ected by increasing probe intensity (¢lled circles in B; ¢lled squares in C). For comparison, inhibitory FTCs in A are re-plotted in B and C (dashed). (D) The excitatory FTC and inhibitory FTCs (hatched) of another IC neuron. The neuron’s excitatory FTC (dashed, ¢lled diamonds, un¢lled squares, and ¢lled circles) progressively decreased with increasing masker intensity (¢lled and un¢lled triangles in D) above the low- (E) and high-£ank (F) inhibitory MTs. For comparison, the excitatory FTC in D was replotted in E and F (dashed).
mation carried by sounds has traditionally been explained by neural interactions of divergent and convergent projections within the ascending auditory system (Suga, 1997). During these neural interactions, a neuron’s response properties are determined by the arrival time and intensity of excitatory and inhibitory inputs at the neuron. This is demonstrated in the present study in which masking of excitatory responses of IC neurons was characterized by variation in the number of impulses, response latency, and size of excitatory FTCs
with IPIs and was a¡ected by inter-pulse intensity difference (Figs. 1, 2, 6 and 7). We have shown that the start IPIs of IC neurons were distributed over a wide range of negative IPIs while the end IPIs were mainly distributed within positive 1^3 ms IPIs (Fig. 4). Because a masker with longer duration produced a greater decrease in a neuron’s number of impulses than a masker with shorter duration at each IPI, this duration-dependent masking increased the temporal window of masking by a left-ward
Fig. 8. Hypothetical sketches showing possible synaptic integration underlying time- and intensity-dependent masking of responses of IC neurons observed in this study. The probe (un¢lled) and the masker (¢lled) are delivered either at di¡erent IPIs (A^C) and/or at di¡erent durations and intensities (D^F). E and I: envelope of EPSPs and IPSPs (see text for details).
HEARES 3935 2-7-02
148
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
shift of the start IPI (Fig. 5). All these observations indicate that masking was most e¡ective when a masker was presented prior to or simultaneously with a probe, similar to a previous study (Litovsky and Yin, 1998). In this study, we did not examine the e¡ect of probe duration on the temporal window for masking. This was because most IC neurons behave as duration-tuned neurons such that they discharge maximally to a speci¢c sound duration or a range of sound durations but the discharge drops more than 50% at other sound durations (Brand et al., 2000; Casseday et al., 1994; Chen, 1998; Ehrlich et al., 1997; Fuzessery and Hall, 1999; Jen and Schlegel, 1982; Jen and Feng, 1999; Jen and Zhou, 1999; Pinheiro et al., 1991; Zhou and Jen, 2001). For this reason, it would be di⁄cult to determine if a change in temporal window for masking determined at di¡erent probe durations was due to interactions of probe and masker at di¡erent IPIs or simply due to a neuron’s selectivity to the probe duration. 4.2. Inter-pulse intensity di¡erences on masking In agreement with a previous study in frogs (Fuzessery and Feng, 1983), we observed that inhibitory FTCs of most IC neurons elevated and decreased with increasing probe intensity (Fig. 6A,D). We also observed that the excitatory FTCs of most IC neurons elevated and decreased with increasing masker pulse intensity (Fig. 7D^F). This intensity-dependent masking is conceivably a result of variation in excitation-to-inhibition ratio in which a masker needs to be stronger than a probe to produce a masking e¡ect. This is substantiated by the fact that percent masking decreased with increasing probe intensity but increased with increasing masker intensity (Fig. 6B,C). Under whole-cell patch-clamp recording conditions, a previous study showed that pulses with frequencies below and above the excitatory BF of bat IC neurons elicited short-latency and long-latency inhibitory postsynaptic potentials which inhibit the responses of IC neurons (Covey et al., 1996). The amplitude of these inhibitory synaptic potentials increased with increasing sound intensity. These synaptic mechanisms may explain our ¢nding that increasing masker intensity elevated and decreased excitatory FTCs (Fig. 7D^F). Masking responses of ¢ve IC neurons were intensityindependent such that increasing probe intensity did not a¡ect high- and low-£ank inhibitory FTCs (Fig. 7A^C). This intensity-independent masking has been previously reported for low-£ank inhibitory FTCs of anuran midbrain neurons (Fuzessery and Feng, 1983). This intensity-independent masking might be due to the fact that increasing excitation by the probe did not overcome the powerful masking e¡ect produced by the masker to these neurons.
In this study, we observed that presentation of a masker typically decreased a neuron’s excitatory response and the percent masking could be as large as 100% (i.e. complete masking). However, when the probe intensity was very strong (e.g. Fig. 6B,C,Ea), responses of some IC neurons were increased under masking conditions such that the percent masking became negative. Because the increased response of these neurons under masking conditions was always less than 20%, we do not know if this observation was due to response £uctuation or a true facilitation. 4.3. Possible mechanisms underlying time- and intensity-dependent masking In this study, we used a probe to elicit excitatory responses from a recorded IC neuron. We then demonstrated that the neuron’s excitatory response could be modi¢ed by the duration and intensity of a masker when delivered within a certain temporal window (Figs. 1^7). This modi¢cation of excitatory response of an IC neuron under di¡erent masking conditions was most likely a result of synaptic integration of excitation and inhibition at the IC neuron. Because the arrival time, amplitude and duration of excitatory and inhibitory postsynaptic potentials (i.e. EPSPs and IPSPs) determine the response of a postsynaptic neuron during synaptic transmission, the time- and intensitydependent masking of responses of IC neurons could conceivably be explained by integration of synaptic potentials as shown in Fig. 8. For example, when a masker was delivered prior to a probe at a large IPI, an IPSP would diminish before the arrival of an EPSP (Fig. 8A). As a result, a neuron’s excitatory response would not be a¡ected by the masker (e.g. Fig. 1B when IPI was between 310 and 37 ms). As both IPSPs and EPSPs overlapped at decreasing IPI (Fig. 8B), a neuron’s excitatory response would decrease (e.g. Fig. 1 when IPI was between 34 and +2 ms). However, when the masker was presented after the probe, the early arrival of an EPSP would not be a¡ected by the late arrival of an IPSP (Fig. 8C). Under this masking condition, the neuron’s response would not be a¡ected by the masker (e.g. Fig. 1 when IPI was larger than +3 ms). When a long-duration masker was presented prior to a probe at a speci¢c IPI, the long-duration IPSP would overlap with the EPSP to decrease excitatory response (Fig. 8D). However, when a short-duration masker was presented at the same IPI, overlapping of two PSPs would not occur (Fig. 8E). To produce masking, the short-duration masker would have to be presented prior to the probe at shorter IPIs. As a result, the temporal window for masking would become smaller for a shortduration masker (e.g. Fig. 5).
HEARES 3935 2-7-02
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
When a masker was presented prior to a probe at di¡erent inter-pulse intensity di¡erences, a neuron’s excitatory response would be determined by a summation of IPSPs and EPSPs (Fig. 8F). When the intensity of probe relative to that of masker progressively increased, the IPSP produced by the masker would become less e¡ective in reducing the EPSP. For this reason, the degree of masking would decrease with an increase in probe intensity (Fig. 6B,C). In summary, we have shown that masking of probeelicited responses of IC neurons occurred when a masker was presented within a certain temporal window. Masking increased with increasing masker duration and intensity but decreased with increasing probe intensity. The time- and intensity-dependent masking e¡ect may be explained on the basis of di¡erent arrival times and amplitudes of excitatory and inhibitory inputs at IC neurons. Future work using intracellular recording is needed to determine the synaptic mechanisms underlying these observations.
Acknowledgements We thank two anonymous reviewers for commenting on the earlier version of the manuscript. This work is supported by a research grant from the National Science Foundation (NSF IBN 9907610), a grant from the Research Council (URC 01-012 Jen) and a matching fund from the College of Arts and Sciences of the University of Missouri-Columbia.
References Aitkin, L., 1986. The Auditory Midbrain: Structure and Function in the Central Auditory Pathway. Humana Press, Clifton, NJ. Brand, A., Urban, R., Grothe, B., 2000. Duration tuning in the mouse auditory midbrain. J. Neurophysiol. 84, 1790^1799. Casseday, J.H., Covey, E., 1996. A neuroethological theory of the operation of the inferior colliculus. Brain Behav. Evol. 47, 311^ 336. Casseday, J.H., Ehrlich, D., Covey, E., 1994. Neural tuning for sound duration: role of inhibitory mechanisms in the inferior colliculus. Science 264, 847^850. Chen, G.D., 1998. E¡ects of stimulus duration on responses of neurons in the chinchilla inferior colliculus. Hear. Res. 122, 142^150. Covey, E., Julie, A.K., Casseday, J.H., 1996. Whole-cell patch-clamp recording reveals subthreshold sound-evoked postsynaptic currents in the inferior colliculus of awake bats. J. Neurosci. 16, 3009^3018. Ehret, G., Merzenich, MM., 1988. Complex sound analysis (frequency resolution, ¢ltering and spectral integration) by single units of the inferior colliculus of the cat. Brain Res. Rev. 13, 139^163. Ehrlich, D., Casseday, J.H., Covey, E., 1997. Neural tuning to sound duration in the inferior colliculus of the big brown bat, Eptesicus fuscus. J. Neurophysiol. 77, 2360^2372. Eysel, U.T., Worgotter, F., Pape, H.C., 1987. Local cortical lesions
149
abolish lateral inhibition at direction selective cells in cat visual cortex. Exp. Brain Res. 68, 606^612. Fuzessery, Z.M., Feng, A.S., 1983. Frequency selectivity in the anuran medulla: excitatory and inhibitory tuning properties of single neurons in the dorsal medullary and superior olivary nuclei. J. Comp. Physiol. A 150, 107^119. Fuzessery, Z.M., Hall, J.C., 1999. Sound duration selectivity in the pallid bat inferior colliculus. Hear. Res. 137, 137^154. Herbert, H., Ascho¡, A., Ostwald, J., 1991. Topography of projections from the auditory cortex to the inferior colliculus in the rat. J. Comp. Neurol. 304, 103^122. Hu¡man, R.F., Henson, O.W., Jr., 1990. The descending auditory pathways and acoustico-motor systems: connections with the inferior colliculus. Brain Res. Rev. 15, 295^323. Jen, P.H.S., Feng, R.B., 1999. Bicuculline application a¡ects discharge pattern and pulse-duration tuning characteristics of bat inferior collicular neurons. J. Comp. Physiol. A 184, 185^194. Jen, P.H.S., Schlegel, P.A., 1982. Auditory physiological properties of the neurons in the inferior colliculus of the big brown bat, Eptesicus fuscus. J. Comp. Physiol. A 147, 351^363. Jen, P.H.S., Zhang, J.P., 1999. Corticofugal regulation of excitatory and inhibitory frequency tuning curves of bat inferior collicular neurons. Brain Res. 841, 184^188. Jen, P.H.S., Zhou, X.M., 1999. Temporally patterned sound trains a¡ect duration tuning characteristics of bat inferior collicular neurons. J. Comp. Physiol. A 185, 471^478. Jen, P.H.S., Sun, X.D., Chen, D.M., Teng, H.B., 1987. Auditory space representation in the inferior colliculus of the FM bat, Eptesicus fuscus. Brain Res. 419, 7^18. Katsuki, Y., Sumi, T., Uchiyama, H., Watanabe, T., 1958. Electric responses of auditory neurons in cat to sound stimulation. J. Neurophysiol. 21, 569^588. Kelly, D.H., 1973. Lateral inhibition in human color mechanisms. J. Physiol. 228, 55^72. Litovsky, R.Y., Yin, T.C., 1998. Physiological studies of the precedence e¡ect in the inferior colliculus of the cat. II. Neural mechanisms. J. Neurophysiol. 80, 1302^1316. Lu, Y., Jen, P.H.S., 2001. GABAergic and glycinergic neural inhibition in excitatory frequency tuning of bat inferior collicular neurons. Exp. Brain Res. 14, 331^339. Lu, Y., Jen, P.H.S., Zheng, Q.Y., 1997. GABAergic disinhibition changes the recovery cycle of bat inferior collicular neurons. J. Comp. Physiol. A 181, 331^341. Mountcastle, V.B., Powell, T.P.S., 1959. Neural mechanisms subserving cutaneous sensitivity, with special reference to the role of afferent inhibition in sensory perception and discrimination. Bull. Johns Hopkins Hosp. 105, 201^232. Pinheiro, A.D., Wu, M., Jen, P.H.S., 1991. Encoding repetition rate and duration in the inferior colliculus of the big brown bat, Eptesicus fuscus. J. Comp. Physiol. A 169, 69^85. Pollak, G.D., Casseday, J.H., 1989. The Neural Basis of Echolocation in Bats. Springer-Verlag, Berlin. Saldan‹a, E., Feliciano, M., Mugnaini, E., 1996. Distribution of descending projections from primary auditory neocortex to inferior colliculus mimics the topography of intracollicular projections. J. Comp. Neurol. 371, 15^40. Singer, W., Creutzfeldt, O.D., 1970. Reciprocal lateral inhibition of on- and o¡-center neurons in the lateral geniculate body of the cat. Exp. Brain Res. 10, 311^330. Suga, N., 1995. Sharpening of frequency tuning by inhibition in the central auditory system: tribute to Yasuji Katsuki. Neurosci. Res. 21, 287^299. Suga, N., 1997. Parallel-hierarchical processing of complex sounds for specialized auditory function. In: Crocker, M.J. (Ed.), Encyclopedia of Acoustics. Wiley, New York, pp. 1409^1418.
HEARES 3935 2-7-02
150
Y. Lu, P.H.S. Jen / Hearing Research 169 (2002) 140^150
Suga, N., Tsuzuki, K., 1985. Inhibition and level-tolerant frequency tuning in the auditory cortex of the mustached bat. J. Neurophysiol. 53, 1109^1145. Suga, N., Zhang, Y., Yan, J., 1997. Sharpening of frequency tuning by inhibition in the thalamic auditory nucleus of the mustached bat. J. Neurophysiol. 77, 2098^2114. Winer, J.A., Larue, D.T., Diehl, J.J., Hefti, B.J., 1998. Auditory cortical projections to the cat inferior colliculus. J. Comp. Neurol. 400, 147^174.
Zhou, X.M., Jen, P.H.S., 2000. Neural inhibition sharpens auditory spatial sensitivity of bat inferior collicular neurons. J. Comp. Physiol. A 186, 389^398. Zhou, X.M., Jen, P.H.S., 2001. The e¡ect of sound intensity on duration tuning characteristics of bat inferior collicular neurons. J. Comp. Physiol. A 187, 63^73.
HEARES 3935 2-7-02