Inferior colliculus neuronal responses to masking-level-difference stimuli

HBalllr, IESBICH ELSEVIER

Heating Research 99 (1996) 79-84

Inferior colliculus neuronal responses to masking-level-difference stimuli Pitchaiah Mandava *, Allen L. Rupert, George Moushegian Callier Center for Communication Disorders, University of Texas at Dallas, 1966 lnwood Road, Dallas, TX 75235, USA

Received 20 February 1996; revised 8 May 1996; accepted 11 May 1996

Abstract

Seventy-one inferior colliculus neurons, with best frequencies below 1.5 kHz, were studied in a binaural, forward-masking paradigm in chinchilla. Masker and signal frequencies were presented at neuronal best frequency. Masker level was set 10-15 dB above neuronal threshold and varied to include a range of signal-to-masker ratios and overall intensities. Without the masker, 33 of the neurons preferred an in-phase signal (SO), 29 an out-of-phase (STr) signal, and the remaining 9 had 'no-preference' (NP), responding equally well to SO and S'rr. Complete protocols from 53 of the 71 neurons were obtained with and without maskers over a range of levels. With an in-phase masker (NO), some neurons responded better to dichotic (NOS'tr) than to diotic (NOSO) sounds. Generally, they maintained a particular phase preference with and without masker. Some neurons, however, altered phase preference and responsivity when binaural maskers were added to signal. Signal-to-noise ratios between 0 and 30 dB were sufficient to differentiate neuronal responsiveness to NOSO and NOS~. The results suggest that identical neural mechanisms are not involved in processing unmasked (SO or S-n) and masked binaural sounds (NOSO, NOS~r). Furthermore, changes in neuronal sensitivity favor the NOSIr condition upon addition of noise (NO) to the signal (SO or S'rr). We conclude that greater neural activity is generated with stimuli which produce masking-level difference than stimuli that do not. Keywords: Binaural masking-level difference; Inferior colliculus neuron

1. I n t r o d u c t i o n

Inferior colliculus (IC) and superior olivary complex (SOC) are sites where binaural stimuli are processed (Rose et al., 1966; Geisler et al., 1969; Goldberg and Brown, 1969; Moushegian et al., 1971, 1975). Bilaterally innervated neurons within these aggregates are differentially responsive to time and level differences of sounds, the stimulus cues involved in lateralization and localization as well as binaural masking (Jeffress, 1971; McFadden, 1975; see review by Irvine, 1992). In binaural masking studies, signal and masker are presented to both ears. If the binaural signal or masker phase is reversed, masked thresholds are lowered (Hirsh, 1948); the effect is known as a masking-level difference (MLD). Numerous psychophysical experimental strategies have been employed to understand how the auditory system works in various detection and MLD tasks (Jeffress,

* Corresponding author. Present address: Box 143529, UTMB, Galveston, TX 77555, USA. E-mail: [email protected]

1971; Jeffress and McFadden, 1971; Small et al., 1972; McFadden, 1975; Yama, 1991). It has been reported that the M L D condition N O S ~ evokes cortical responses at a lower sound level than N O S O (Tunis and Teas, 1974; Yonovitz et al., 1979). Such differences in cortical evoked responses may be affected by binaural mechanisms at midbrain and medullary regions of the auditory neuraxis (Efron, 1990). A number of neuronal studies of the SOC and IC have attempted to identify the neural substrate and mechanisms responsible for M L D (Roth and Sinex, 1981; Langford, 1984; Caird et al., 1989, 1991). It was first reported that, at the level of IC, no correlation exists between M L D tasks and either discharge rates or synchronization indices (Caird et al., 1989). A later study suggested that if a signal is presented at a neuron's best delay and the masked threshold determined at various masker delays, the population responses of a subset of IC neurons correlate with the psychophysical M L D function (Caird et al., 1991). In the current study a forward-masking paradigm that minimizes level differences was used because binaural neurons, particularly those with low best frequencies, may

0378-5955/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved. PII S0378-5955(96)00085- 8

80

P. Mandava et al. / Hearing Research 99 (1996) 79-84

be sensitive to both level and time differences. The findings, therefore, are a consequence only of the phase difference between NOSO and NOSw. Furthermore, forward masking is an appropriate design for study of nervous system processes related to masking phenomena (Harris and Dallos, 1979), since simultaneous and forward masking are presumably based on shared or identical mechanisms (Small et al., 1972; Dolan and Trahoitis, 1972; Punch and Carhart, 1973), and also because they give rise to similar MLD values (Yama, 1991). 2. Methods The data were obtained from 53 chinchillas weighing between 250 and 350 g. They were initially anesthetized intramuscularly with ketalar (45 m g / k g ) and nembutal (45 mg/kg). A supplemental dose of ketalar (10 mg), and occasionally nembutal (5 mg), was administered approximately every hour to maintain an appropriate anesthetic state. In preparation for the experiment, the animal was placed in a stereotaxic instrument and held firmly with hollow earbars and a bite bar. A coiled wire, to monitor cochlear microphonic and auditory nerve activity, was placed near each round window and sealed onto the bulla wall with dental acrylic. Skin and bone around the foramen magnum were removed to visualize the cerebellum, fourth ventricle, and medulla. A tungsten microelectrode, was inserted into the cerebellum and advanced rostrally at an angle to enter the central nucleus of the left IC, where neuronal recordings were obtained. At the conclusion of each experiment, the animal was perfused with saline and 10% formalin. The brain was removed from the skull, cut serially and stained with hematoxylin and eosin to ascertain the location of neurons from which recordings were obtained. This study includes results only from those neurons which were unequivocally established to be within the central nucleus of the IC. Signals and maskers were pure tones delivered at a neuron's best frequency. The duration, delay, and phase of signal and masker were controlled separately. The signal, when presented alone or with masker, was 40 ms and the masker 100 ms in duration; both had rise-fall times of 5 ms. The signal was delayed 10 ms from the off of the masker. The stimuli were presented 50 times at a rate of 2 / s through earphones (TDH-39) attached to the hollow earbars inserted into the ears. Having determined a neuron's phase preference (SO vs. S-rr), the signal and masker level were set at 10-15 dB above a neuron's threshold for the 0 dB signal/masker ratio condition. Neuronal spikes were monitored by an amplitude discriminator whose output, a square pulse for each discharge, was fed into a computer (PDP-11/20) programmed to display on-line post-stimulus time histograms (PSTH). The data were transferred to a PDP-11/45 for storage and further analyses. The PSTHs were viewed and

selectively windowed to sum the number of neuronal discharges to the masker and the signal. Throughout the text NO denotes the masker, SO and S'rr the signals, in- and out-of-phase, respectively, and NOSO or NOS'rr as the combined stimuli. In all of the figures, neuronal responses to the masked conditions (NOSO and N O S ~ ) were normalized to the best response obtained in the unmasked condition. For example, if a neuron was driven better by SO, the data are presented as ratios of N O S O / S O and N O S ~ / S O ; if on the other hand, a better response was obtained to S~, the ratios are N O S O / S w and NOS-rr/Sw.

3. Results Recordings were obtained from 71 IC neurons having best frequencies below 1.5 kHz. The majority were categorized as excitatory-excitatory (EE) because monaurally they could be driven by sound at either ear (Goldberg and Brown, 1969). At best frequency, 47 neurons, mostly EE, were driven better to contralateral than ipsilateral sounds; only 12 (EE) responded better to an ipsilateral sound; 7 were driven equally well by either ear; 2 required binaural sounds to be activated and 3 could not be classified. Most of the 71 neurons were driven tonically to the 40 ms binaural tones at best frequency. Thirty-three responded best to in-phase signals (SO), 29 to out-of-phase signals (Sw), and 9 responded similarly to SO and S~r. The functions displayed in Fig. 1 are from 4 EE neurons which in the presence of an in-phase masker were driven better when the signal was out-of-phase (NOSav)

IOO

I00

~ 50

~o

~

e=NOS~ o:NOSO

B/

50

31-T

0

g ~;100 x

,~5o

ZT-S

~-4

0

-30 -25-20

-i5

-Io

-s

0

-zo

-15 -I0

-5

0

SIGNAL / M A S K E R RATJO(dB)

Fig. 1. Neuronal rate functions to signal/masker ratio changes of NOSO and NOS~. A: Best frequency, 500 Hz, EE onset neuron; B: Best frequency, 300 Hz, EE onset neuron; C: Best frequency, 500 Hz, EE onset neuron. D: Best frequency, 370 Hz, EE tonic neuron. Ordinate: discharge rates expressed in percentages of N O S O / S f r and NOS ~r/S ~. Abscissa: Ratio of signal and masker intensities.

81

P. Mandaua et al. / Hearing Research 99 (1996) 79-84 I00

I

I

I

I

A

100

B

NOSO /~

• : NOSII o:

/o.

5o

50 '~

-~

3 S - 3 "

c

D 50

8N

~a-,

0

I L I - 2 0 -I,5 - I 0 SIGNAL / MASKER RATIO (dB)

-I15 -I~) -5

0'

I -5

I 0

Fig. 2. Neuronal rate functions to signal/masker ratio changes of NOSO and NOS,rr. A: Best frequency, 300 Hz, EE tonic neuron; B: Best frequency, 1014 Hz, EE tonic neuron; C: Best frequency, 835 Hz, EE tonic neuron; In D: Best frequency, 633 Hz, EE tonic neuron; Ordinate and abscissa same as in Fig. l except that ordinate values are expressed as NOSO/SO and NOS'n"/SO.

than when it was in-phase (NOSO). The responsiveness of these neurons increased for N O S w as the signal to masker ratio was raised; whereas for the N O S O combination, the discharges remained minimal. For the N O S O condition, neuronal activity was not present throughout the intensity range (Fig. 1A,D), or increased only minimally at the highest s i g n a l / m a s k e r ratio (Fig. 1B,C). Thus all 4 neurons are significantly suppressed by N O S O sounds. The highest response, for example, to N O S O is about 5% for the neuron whose results appear in Fig. lB. To lower the response in the NOS-rr condition to 5%, the intensity of the signal had to be reduced 10 dB, suggesting that NOS'tr was an improvement of 10 dB over NOSO. For the other neuron (Fig. 1C), the response to N O S O at 0 dB s i g n a l / m a s k e r ratio is about 10%; to bring the NOS~r function to 10% required more than 15 dB reduction in the s i g n a l / m a s k e r ratio. In Fig. IA,D, the signal (SAT) had to be lowered by 30 and 15 dB to bring the discharge rates to values which were obtained when the sound was NOSO. Such differences o f neural response to N O S O and NOS~r could be the basis for the neurophysiological mechanism in 'release from m a s k i n g ' (McFadden, 1975). All of these EE neurons were only minimally responsive to NOSO, an indication that the extent of excitation was selectively and effectively increased by a simple phase reversal in the signal. The functions illustrated in Fig. 2 are also from 4 EE neurons. In contrast, however, these showed maximal activity to N O S O rather than to N O S w . F o r the signal to masker ratios used, the discharge rates o f 3 of the neurons were not altered when the stimulus was N O S r r (Fig. 2A,B,D); to N O S O , however, their discharge rates in the presence of the masker stimulus, increased as signal levels were elevated. In Fig. 2C discharge rates were raised

slightly when the stimulus was NOSax; but to NOSO, neuronal activity was e l e v a t e d c o n s i d e r a b l y with s i g n a l / m a s k e r ratio increases. The results of Figs. 1 and 2 indicate that most binaural neurons of the IC respond better either to N O S O or NOSTr depending on their SO or S~r preference; and that as the level of the signal is raised the discharge rates are increased to the preferred, whether it is N O S O or NOS-rr, but not the non-preferred sound. The results in Fig. 3 were obtained at three intensities o f the combined signal and masker stimulus; they illustrate how overall levels near and beyond threshold influence or modify discharge rates. The data represent the extremes observed from a population of neurons where overall level was a parameter. Both neurons were more active to NOS'rr than to N O S O at 10, 20, and 30 dB. One displayed steeper rate functions to NOS'rr and was affected minimally by overall levels and s i g n a l / m a s k e r ratios of N O S O (Fig. 3A). The 30 dB overall level produced the most and the 10 dB level the least ' M L D ' . Response values exceeding 100% indicate that the combination o f signal and masker produced a greater discharge rate than when the signal alone was sounded. In Fig. 3B, the effect of overall levels is not as dramatic; but the general trend of N O S r r is a higher discharge rate for each overall level. At 10 dB overall level (Fig. 3B), discharge rates to N O S O and NOS-rr were equal when the s i g n a l / m a s k e r ratio was - 10 dB, whereas at 20 dB overall level, equivalence in discharge rates occurred at a - 2 0 dB ratio; and at 30 dB overall level, the rates of discharge were identical when the s i g n a l / m a s k e r was - 3 0 dB. These extreme differences consequent to overall levels are evidence of considerable diversity of neuronal function in binaural masking. These data also show that neurons reach equivalent discharge rates for signal i n - a n d out-of-phase anywhere from - 10 to - 3 0 dB depending on overall level of the sound. This range does not parallel psychophysical observations that the largest M L D s occur to signals close to threshold and smaller ones to signals well above threshold (Jeffress and McFadden, 1971).

!

|

150

w

!

!

i

|

|

150

A too

~E 50

30dB 20dB IOdB • l m = NOS1T 0 & m = NOSO

/

IO0

50

u0 0

-~b

-~'o

-,o

o

-~

-2o

-,b

"

SIGNAL / MASKER RATIO (d8)

Fig. 3. Rate functions for two neurons to signal/masker ratio changes and overall levels of NOSO and NOS~r. A: Best frequency, 400 Hz, EE Onset neuron; B: Best frequency, 650 Hz, EE tonic neuron. Parameter is overall level of sound relative to neuron's threshold at best frequency.

P. Mandaua et al. / Hearing Research 99 (1996) 79-84

82

IOO

;

,

e = sll"

,

, e=NOSI[ o • NO SO

O - SO

i10 0

A

50

~-Ie,

o

.

I00

i

.

.

.

0 IO0

I

c

D

.~ 5o

5O

41o|

C

S.,

I00

50

-~o-,;-,;-;

i

i

i

0

i

I00

o

'

-15

'

-tO

-

~

o'

S I G N A L / M A S K E R (dB)

Fig. 4. Responses of three neurons to signals in- and-out-of-phase in masked and unmasked stimulus paradigms as a function of intensity. A,B: Best frequency, 900 Hz, EE tonic neuron; C,D: Best frequency: 683 Hz, EE onset neuron; E,F: Best frequency, 300 Hz, EE tonic neuron.

I

i

l

i

!

i

I

i

i

A I00 -

z~

5O

~

o

Table 1 Phase difference of binaural neurons with and without masker

-=ST[

, -,

O0

// NOS'I~

o=NO$O

0 :

:

!

e

I00

~ 5c

5O

I

I

I

I

I

-I0 -5 0 RELATIVE I N T E N S I T Y (dB)

-20

-15

I

I

I

i

-~5 - I 0 -5 0 SIGNAL / MASKEff(dB)

Fig. 5. Responses of two neurons to signals in- out-of-phase in masked and unmasked stimulus paradigms as a function of intensity. A,B: Best frequency, 500 Hz, EE onset neuron: C,D: Best frequency, 250 Hz, El

onset neuron.

The functions illustrated in Figs. 4 and 5 were obtained by utilizing the same experimental paradigms as those described earlier except that discharge rate functions were from various signal levels in the absence o f a masker as

NOS"rr NOSO NP

Sw (n = 17) *

SO (n = 27)

NP (n = 9)

Total (n = 53)

16 1 0

4 23 0

4 1 4

24 25 4

* Six discharged at higher rates to NOS'rr than to S'rr. SO, signal in phase; S'rr, signal out of phase; NP, equally well to SO, ST; NOSO, noise in phase, signal in phase; NOSIr, noise in phase, signal out of phase.

well as in its presence. The three neurons whose unmasked responses are shown in Fig. 4A,C,E increased their discharge rates as signal intensities were raised. This was the case whether the signal was in- or out-of-phase. For the neuron whose responses to level are shown in Fig. 4A, the discharge rates to SO and S'rr are identical whereas for the others they are not (Fig. 4C,E). One has a lower threshold to S Tr than to SO (Fig. 4C) and although the other has similar thresholds to S'rr and SO, its responsiveness was more vigorous to S ~ than to SO when the signal level was raised (Fig. 4E). Fig. 4 also displays the responsiveness of these neurons when the masker was added to the signal. In all instances, the masker dramatically affected discharge rate in a manner unpredictable from the unmasked condition. In Fig. 4B, responses to NOS'rr are better than to NOSO; in Fig. 4D they are the same; and in Fig. 4F, NOS-rr is almost identical to its unmasked function whereas N O S O is greatly depressed. Response functions from two neurons which were driven more by S'rr than to SO in the absence of noise are illustrated in Fig. 5. The discharge rates of both neurons exceeded 100% to the preferred stimulus ( N O S w ) and were greatly depressed to N O S O throughout 1 5 - 2 0 dB of s i g n a l / m a s k e r ratios. These results (Figs. 4 and 5) indicate the presence of a mechanism in the lower brainstem which selectively differentiates binaural signals in the presence of a masker. Table 1 lists findings on 53 neurons from which recordings were obtained to the entire set of stimuli in the M L D repertoire. The tabulation displays the preference these neurons showed to S w, SO, or NP in the presence and absence of a masker. There were 17, 27 and 9 neurons in the S ~ , SO, and NP categories, respectively, when the masker was not sounded. With masker, however, the distribution changed. Four SO neurons became more sensitive to NOS'rr than to NOSO. Of the 9 NP neurons, 4 changed to a preference o f NOS'rr and one to N O S O when masker was introduced. The most invariant group of neurons was the S'rr since only one from this category moved to N O S O with masker. The S'rr neurons were further distinguished by the fact that only they responded at higher discharge rates with masker than in its absence; none of the SO or NP exhibited this characteristic.

P. Mandava et al. / Hearing Research 99 (1996) 79-84

Although the phase preference changes with masker did not approach statistical significance, a substantial shift of preferences occurred with masker, all of which interestingly are toward the NOS w group.

4. Discussion

Discovered and described by Hirsh (1948) and extensively pursued subsequently by Jeffress (1971) and others (Small et al., 1972; McFadden, 1975; Yama, 1991), a MLD means that the threshold to a low-frequency binaural signal, in the presence of a masker, is lowered appreciably when the phase of either the signal or masker is reversed. Time and level differences of sounds at the ears are the physical cues for spatial localization employed by twoeared animals. Since they are also parameters involved in "release from masking" (Jeffress and McFadden, 1971), any of the several binaural aggregates of the caudal auditory neuraxis, e.g., medial superior olive (MSO), or S-segment of the SOC, the lateral lemniscal nuclei, and IC are sites where binaural stimuli are known to be processed (see review by Irvine, 1992). The study of any one of these aggregates may provide information regarding the neural basis of the MLD phenomenon. We chose to study neurons of the IC for two reasons. First, the IC receives and integrates inputs from all the lower monaural and binaural auditory nuclei; secondly, most IC neurons are responsive to binaural sounds. The MLD phenomenon has been studied in the MSO and IC (Langford, 1984; Caird et al., 1989, 1991). Langford reported that in the MSO 47% of the neurons with BFs below 1.8 kHz were "eeei", meaning no response to sounds monaurally in either ear; however, when both ears were stimulated this type of neuron was activated in a facilitatory manner. The data from these e c e i neurons were interpreted as evidence for the existence literally of a coincidence-detection-place model as the basis for binaural hearing and MLD. Findings from earlier research differed in the sense that very few, if any, ece i neurons were identified in recordings of MSO neurons (Goldberg and Brown, 1969; Moushegian and Rupert, 1974). More recent research on the MSO has failed to record from any significant number of this neuronal type (Yin and Chan, 1990; Spitzer and Semple, 1995). These studies, however, have presented findings from EE neurons which support a coincidence detection model. Neuronal studies of the IC have also not uncovered any binaural eee i neurons (Caird et al., 1989, 1991). In the present investigation, also, the majority were EE neurons. One study of MLD phenomenon in the IC classified cells in accordance to whether they responded better to in-phase (O cells) or out-of-phase (w cells) signals (Caird et al., 1989). More than half (n = 92) of the binaural neurons were depicted as O-cells (n = 49) and relatively few as w-cells (n = 10); and about one-third of the cells

83

had no preference. In our population (n = 71), 33 preferred in-phase (O-cells), 29 out-of-phase stimuli at BF (w-cells) and only 9 responded equally well to either phase. The methods and stimuli of the present study were different from the other MLD studies. A forward-masking design was used to simulate a psychophysical MLD paradigm. All IC neurons were differentially responsive to a MLD stimulus. Since neither time nor intensity differences were present in the sounds we employed, the results are a consequence simply of the phase differences between NOSO and NOSw. Only a few more neurons were better driven by an in-phase (SO) than an out-of-phase (S,rr) signal between ears; and with masker, the shift of preferences by neurons is almost entirely to NOSw. For instance, 4 of the SO and 4 of the NP neurons shift preference to NOSw. This observation, and the fact that only Sw neurons discharged at rates greater than 100% at certain signal masker ratios (Fig. 1C, 3A and 5B), suggests that more neural activity is generated from the IC for NOSw than NOSO sounds. In this study neurons exhibited easily distinguishable responses between NOSO and NOS w functions for signalto-noise ratios as little as - 2 5 dB and as great as 0 dB. In psychophysical studies, MLDs between subjects may vary from a few dB to as much as 15 dB. Thus, neurons in the inferior colliculus are more sensitive to stimuli of the MLD paradigm than is the psychophysical observer. Overall level, as would be expected, is a variable over which diotic (NOSO) and dichotic (NOS-rr) sounds differentially affect the neuronal discharges (Fig. 3). However, these neural data do not correlate closely to the psychophysical finding that MLDs are greatest near threshold. Over a two decades ago Butler and Kluskens (1971) wrote that the amplitude of the auditory evoked response (N1P2) was significantly greater to Sw than to SO signals at 200 Hz but equal amplitudes were obtained when the frequency was 2000 Hz. They wrote that the S rr sound was perceived to be dispersed throughout the head whereas the SO is heard punctately. They speculated that an outof-phase sound might excite more neurons than an in-phase one. Their findings were extended to show that the N1P2 potentials are also larger for NOSw than NOSO at near threshold and suprathreshold levels (Tanis and Teas, 1974; Yonovitz et al., 1979). These evoked response studies have concluded that significantly more neurons are activated by dichotic than diotic sounds. Based on different evoked response measures, all intermediate substrates starting at the SOC and ending at the cortex have been suggested as processing centers wholly or partly involved in the psychophysical MLD phenomenon (Jerger et al., 1982; Kevanishvilli and Lagidze, 1987; Fowler and Mikami, 1992; Galambos and Makeig, 1992). Caird et al. (1989) wrote that the inferior collicular neuronal discharge rates or synchronization indices did not show any correlation with MLD. A later study used tones

84

P. Mandava et al. / Hearing Research 99 (1996) 79-84

a n d v o w e l s at a n e u r o n ' s b e s t d e l a y ( C a i r d et al., 1991). M a s k e d t h r e s h o l d s w e r e d e t e r m i n e d as a f u n c t i o n o f m a s k e r delay. T h e p o p u l a t i o n r e s p o n s e o f cells e x h i b i t e d f u n c t i o n s s i m i l a r to p e r f o r m a n c e in h u m a n M L D e x p e r i m e n t s ; t h e y c o n c l u d e d t h a t all l e v e l s o f a u d i t o r y n e u r a x i s u p to the IC are i n v o l v e d in p r o c e s s i n g M L D s . A s w a s t h e c a s e for the a u d i t o r y e v o k e d cortical res p o n s e , the p r e s e n t s t u d y h a s s h o w n t h a t m o r e n e u r a l a c t i v i t y is e v o k e d w i t h s t i m u l i w h i c h p r o d u c e a M L D t h a n s t i m u l i t h a t d o not. All o f t h e s e f i n d i n g s are e v i d e n c e that the I C p l a y s a n i n t e g r a l role in b i n a u r a l p r o c e s s i n g i n c l u d ing the r e l a t e d p h e n o m e n o n o f M L D . T h i s is s u p p o r t e d b y b e h a v i o r a l r e s e a r c h s h o w i n g that less t h a n 3 0 % o f the c e n t r a l n u c l e u s o f the IC is s u f f i c i e n t to p e r m i t an a n i m a l to l o c a l i z e a s o u n d in the h e m i f i e l d c o n t r a l a t e r a l to the l e s i o n ( J e n k i n s a n d M a s t e r t o n , 1982).

Acknowledgements This research was supported by M.F. Jonsson Professorship.

References Butler, R.A. and Kluskens, L. (1971) The influence of phase inversion on the auditory evoked response. Audiology 10, 353-357. Caird, D.M., Pillman, F. and Klinke, R. (1989) Processing of binaural masking level difference signals in the cat inferior colliculus. Hear. Res. 43, 1-24. Caird, D.M., Palmer, A.R. and Rees, A. (1991) Binaural masking level difference effects in single units of the guinea pig inferior colliculus. Hear. Res. 57, 91-106. Dolan, T.R. and Trahoitis, C. (1972) Binaural interaction in backward masking. Percept. Psychophys. 11, 92-94. Efron, R. (1990) The decline and fall of hemispheric specialization. Lawrence Erlbaum, New Jersey. Fowler, C.G. and Mikami, C.M. (1992) Effects of noise bandwidth on the late-potential masking level difference. Electroencephalogr. Clin. Neurophysiol. 84, 157-163. Galambos, R. and Makeig, S. (1992) Physiological studies of central masking in man. II. Tonepip SSRs and the masking level difference. J. Acoust. Soc, Am. 92, 2691-2697. Geisler, D.C., Rhode, W.S. and Hazelton, D.W. (1969) Responses of inferior colliculus neurons in the cat to binaural acoustic stimuli having wide-band spectra. J. Neurophysiol. 32, 960-974. Goldberg, J.M. and Brown, P.B. (1969) Reponse of binaural neurons of dog superior olivary complex: some physiological mechanisms of sound localization. J. Neurophysiol. 32, 613-636. Harris, D.M. and Dallos, P. (1979) Forward masking of auditory nerve fibre responses. J. Neurophysiol. 43, 1083-1107.

Hirsh, I.J. (1948) The influence of interaural phase on interaural summation and inhibition. J. Acoust. Soc. Am. 20, 536-544. Irvine, D.R.F. (1992) Physiology of the auditory brainstem. In: A.N. Popper and R.R. Fay (Eds.), The mammalian auditory patwhay: Neurophysiology, Springer, New York, pp. 153-231. Jeffress, L.A. (1971) Detection and lateralization of binaural signals. Audiology 10, 77-84. Jeffress, LA. and McFadden, D. (1971) Differences of interaural phase and level in detection and lateralization. J. Acoust. Soc. Am. 49,

1169-1179. Jenkins, W.M., Masterton, R.B. (1982) Sound Localization: effects of unilateral lesions in central auditory system. J. Neurophysiol. 47, 987-1016. Jerger, J., Hanley, M. and Rivera, V. (1982) Auditory brainstem response and the masking level difference. Ann. NY Acad. Sci. 388, 466-470. Kevanishvilli, Z. and Lagidze, Z. (1987) Masking level difference: an electrophysiological approach. Scand. Audiol. 16, 3-11. Langford, T.L. (1984) Responses elicited from medial superior olivary neurons by stimuli associated with binaural masking and unmasking. Hear. Res. 15, 39-50. McFadden, D. (1975) Masking and the binaural system. In: D.B. Tower (Ed.), The Nervous System: Human Communications and its Disorders, Raven Press, New York, pp. 137-146. Moushegian, G., Stillman, R.D. and Rupert, A.L. (1971) Characteristic delays in the superior olive and inferior colliculus. In: M.B. Sachs (Ed.), Physiology of the Auditory System, Educational Consultants, Baltimore, MD, pp. 245-254. Moushegian, G. and Rupert, A.L. (1974) Relations between the psycbophysics and the neurophysiology of sound localization. Fed. Proc. 33, 1924-1927. Moushegian, G., Rupert, A.L. and Gidda, J. (1975) Functional characteristics of superior olivary neurons to binaural stimuli. J. Neurophysiol. 38, 1037-1048. Punch, J. and Carhart, R. (1973) Influence of interaural phase on forward masking. J. Acoust. Soc. Am. 54, 897-904. Rose, J.E., Gross, N., Geisler, C. and Hind, J.E. (1966) Some neural mechanisms in the inferior colliculus of the cat which may be relevant to localization of a sound source. J. Neurophysiol. 29, 288-314. Roth, G.L. and Sinex, D.G. (1981) Responses of inferior colliculus neurons in a masking level difference paradigm. J. Acoust. Soc. Am. 69, S12. Small, A.M., Boggess, J., Klich, R., Kuehn, D., Thelin, J. and Wiley, T. (1972) MLDs in forward and backward masking. J. Acoust. Soc. Am. 51, 1365-1367. Spitzer, M.W. and Semple, M.N. (1995) Neurons sensitive to interaural phase disparity in gerbil superior olive: diverse monaural and temporal response properties. J. Neurophysiol. 73, 1668-1690. Tanis, D.C. and Teas, D.C. (1974)Evoked potential correlates of interaural phase reversals. Audiology 13, 357-365. Yarna, M.F. (1991) Effects of tempopral sensation and masker level on binaural analysis in forward masking. J. Acoust. Soc. Am. 91, 327335. Yin, T.C.T. and Cban, J.C.K. (1990) Interaural time sensitivity in medial superior olive of cat. J. Neurophysiol. 64, 465-488. Yonovitz, A., Thompson, C.L. and Lozar, J. (1979) Masking level differences: auditory evoked responses with homophasic and antiphasic signal and noise. J. Speech Hear. Res. 22, 403-411.