Response enhancement and reduction of the auditory brain-stem response in a forward-masking paradigm

Electroencephalography and chntcal Neurophyslology, 1987, 66. 427-439

427

Elsevaer Soenlaflc Pubhshers Ireland, Ltd EEG03157

Response enhancement and reduction of the auditory brain-stem response in a forward-masking paradigm A.K. Ananthanarayan

and G.M. Gerken

Untoerslty of Texas at Dallas, Calher Center for Commumcatmn Disorders, Dallas, T X 75235 (U S A ) (Accepted for publication 3 July, 1986)

Summary Alteratmns in the probe evoked auditory bra,n-stem response (ABR) were evaluated in 15 n o r m a l - h e a n n g subjects using several sttmulus configurations in a tone-on-tone forward-masking paradigm The stimulus parameters m a m p u l a t e d m the study included masker frequency; relative intensity of the masker, overall intensity of the masker-probe paar; and masker rise-fall tame Latency increases for waves III and V and an amphtude reduction for wave III were observed under some stamulus conditions These changes were interpreted m terms of paxtaal forward-masking effects. The masking effects were shown, to be maxamal for masker frequencaes in close proxarmty to the probe; to increase vath increasing level of masker; to be independent of the overall level of the masker-probe pant; and, to decrease with increasing nse-fall tame of the masker. Collectively, the forward-masking effects were interpreted as peripheral in origin, although, an adthtmnal bra,n-stem locus was not ruled out In contrast, the same stimuh which increased wave III and V iatencles and reduced wave Ill amphtude produced a robust a m p h t u d e Increment in wave V winch was termed enhancement Wave V enhancement was shown to be maxtmal for masker frequenoes m d o s e proxinuty to the probe; to decrease wath increasing masker level, and, to decrease vath faster nse-fall times of the masker The processes mediating wave V enhancement are not dear, however, it was concluded that wave V enhancement probably reflects the resultant of a complex central neuronal mteractaon, presumably m the Vlclrnty of the wave V generator(s)

Key words:

Auditory brain-stem response; Evoked response enhancement, Foreward-masking paradxgm, H u m a n subjects

The several waves that comprise the human auditory brain-stem response (ABR) are commonly described in terms of latency. Tins emphasis on response latency rather than amphtude m evoked response analysis is primarily due to the large variability associated with response amplitude. However, several studies have shown that response amplitude varies systematically with stimulus condition. For example, an increase in stimulus intensity produced an increase in response amplitude (Lev and Sohmer 1972; Hecox and Galambos 1974; Picton et al. 1974), while an increase m the stimulus repetition rate produced a

Correspondence to: G_M Gerken, Ph D , Umverslty of Texas at Dallas, Callier Center for Commumcataon Disorders, 1966 Inwood Road, Dallas, TX 75235, U S A

progressive decrease in response amphtude (Pratt and Sohmer 1976; Don et al. 1977). As further examples, an increase in the rise-fall time of the stimulus produced a decrease in the response amplitude (Suzuki and Horiuchi 1981; Hecox et al. 1976), and the presence of a continuous wide-band noise masker reduced the probe evoked response amplitude (Ananthanarayan 1983; Burkard and Hecox 1983). The measurement of amplitude has thus proved useful in evaluating auditory system function. A number of electrophysiological studies that employed an amphtude measure have used exther a simultaneous masking paradigm (Gumnit and Grossman 1961; Gerken 1971; Spoor et al. 1976; Ananthanarayan 1983; Burkard and Hecox 1983; N o m o t o 1985) or a forward-masking paradigm (Coats 1964; Kramer and Teas 1982; Laskv and

0013-4649/87/$03.50 © 1987 Elsevier Soentlfic Publishers Ireland, Ltd

428

Rupert 1982). In these studies, a reduction m auditory evoked response amplitude has been produced by the use of a broad-band noise as the masker or as the background signal. A few studies employing tonal maskers have shown that evoked response amplitude may be increased rather than decreased in the simultaneous and forward-masking paradigms (Gumnlt and Grossman 1961; Gerken 1971, 1973, Ananthanarayan and Gerken 1983; N o m o t o 1985). Enhancement of the loudness aspect of the auditory percept has been obtained in psychophysical experiments. When two bnef au&tory stimuli closely follow one another, the second can be enhanced m loudness (Buytendijk and Meester 1942; Irwin and Zwlslockl 1971; Sokolich and Zwlslocki 1972; Elmasian and Galambos 1973). The magmtude of this loudness enhancement has been shown to depend on several parameters, for example, loudness enhancement increases with increasing level of the first stimulus. Also, loudness enhancement decreases with increasing temporal separation between the first and the second stimulus. Other aspects of the psychophysical and the physiological experiments are not comparable, however, and there is no basis for concluding that the psychophysical phenomena are related to the presently available physiological data In a prevxous report we demonstrated two contrastmg alterations in the components of the auditory brain-stem evoked response (ABR) in a toneon-tone forward-masking para&gm (Ananthanarayan and Gerken 1983). One group of effects revolved a reduction m the amplitude of wave III and increased latency for waves III and V. These results were interpreted m terms of a peripheral forward-masking effect, the magnitude of which was shown to decrease w~th increase in the temporal separation between the masker and the probe (At). In contrast, the same stimulus conditions produced an amplitude increment in wave V relative to its unmasked amplitude. This effect was termed evoked response enhancement. Over the range of At values from 5 to 135 msec, wave V enhancement was maxamal for At values of 15 and 45 msec, and was interpreted as a central effect related to timing of sound sequences The purpose of this study is to evaluate the effects of other

A K ANANTHANARAYAN, G M GERKEN

stimulus parameters on evoked response amphtude The stimulus parameters evaluated include masker frequency, relative level of the masker, intensity level of the masker-probe pair, and masker rise-fall time.

Methods

Subjects Fifteen female human subjects, ranging in age from 21 to 33 years, participated in the study, although not all of them participated in each of the 4 experiments. Hearing sensxtivlty in all subjects was 15 dB HL or better for octave frequencies from 0.5 to 8 0 kHz

Sttmulus generatton Two tone-burst stimuli, probe and masker, were utihzed in a forward-masking para&gm as shown m Fig. 1. In a forward-masking para&gm the probe (the sUmulus evoking the ABR) is preceded temporally by the masker The sdent interval between masker offset and probe onset is referred to as At. The probe was always a 4.0 kHz tone-burst of 10 msec overall duration including the 1 msec hnear rise and fall times. Probe level was set at 65 dB SL relative to the behaviorally measured threshold for the probe stimulus. The masking stimulus was 60 msec m overall duration including the 10 msec linear rise and fall times. Unless otherwise specified, the masker frequency was also 4.0 kHz and At was 15 msec. The intensities of the stimuli were independently controlled with two attenuators. The outputs of the two attenuators were rmxed, amplified, and then led into an electrostatlcally and magnetically shielded earphone The At interval as well as the repetition rate of masker-probe pairs was controlled by a rmcrocomputer (DEC 11/23). Masker intensity was controlled either manually or through the computer, whale masker frequency, duration, and rise-fall times were controlled manually.

Recording system Subjects rechned on a cot in an acoustically and electrically shielded booth The auditory brain-stem evoked responses were recorded dif-

ABR FORWARD-MASKING AND ENHANCEMENT

MASKER

PROBE

I.M ¢3 I,w ,.I a.

=E ISI

TIME Fig 1 Block diagram dlustratmg the temporal relationships between the masker and probe m a forward-masking paradigm At interval between masker offset and probe onset ISI mterstimulus interval

ferentlally between gold-plated surface electrodes on the vertex and the ipsilateral (right) mastoid. Another electrode placed on the forehead served as the common ground. The lnterelectrode impedances were maintained below 5000 ~2. The electrode leads were connected to an amplifier whose gain was 2 × 105 with bandpass filters set at 0 1 kHz and 3.0 kHz. The differentially amplified output was connected to an A / D converter which provided an additional × 2 gain prior to averaging by the 11/23 microcomputer. Each ABR wave form represents 1 500 stimulus presentations averaged using a 50/tsec dwell time and 320 points. Onset of averaging was synchronized with the probe onset and continued over a time epoch of 16 msec. A video graphics system was used with the DEC 11/23 to display the evoked potentials both as they were recorded onlane, and later for inspection and analysis

Response evaluation Two measures, latency and amplitude, were used to describe the alterations in waves III and V of the ABR Latency was measured from the onset of the probe to the peak of a given wave form. Amplitude was measured from the positive peak of a given wave form to the bottom of the following trough. For all experiments using the masking stimulus, latency changes were expressed as

429

amount of shift (in msec) relative to the latency in the unmasked condmon. The amplitude alteratlons in wave III and wave V for all experiments were normalized relative to the unmasked amplitude and were expressed as percentage change from the unmasked response Statistical treatment of the data involved the use of repeated measure analysas of variance (ANOVA) with latency and amplitude measures as the dependent variables and the masker parameters as the independent variables Separate oneway ANOVAs were performed for each dependent variable (latency, amplitude) and for each wave (III, V) Post-hoc analyses were performed using the Newman-Keuls multiple comparison test. The 0.05 level of sigmficance was selected a pnori as the alpha level For consistency with the definition of masking a n d / o r partial masking used in psychoacoustics (Scharf 1971) and in physiology (Davis and Derbyshare 1935), masking is taken to refer to the ellrmnation of a component of the ABR, and partial masking refers to an amplitude reduction. We will also interpret the increased latency of a component in terms of partial masking.

General procedures The common elements of the 4 experiments constituting this study are described in thas section The masker-probe pair was always presented at 4 / s e c to the right ear through a shielded earphone with a constant At of 15 msec. The slow presentation rate provided a sufficient time interval between probe offset and masker onset to mlnlnuze possible confounding effects from backward masking or from overlap by early and middle latency evoked potentials The random phase onset of the probe eliminated any possible contributaon of the cochlear microphonic to the ABR. In all experiments the order of forward-masking conditions was randomized across subjects, and an additional unmasked ABR was recorded at the end of each of the two recording sessions scheduled for each subject.

Spectfic procedures In experiment I, tone-bursts of different frequencies were used as maskers with 10 normal hearing subjects Evoked responses were first ob-

430

tamed for the probe alone (unmasked condition) and then in the presence of a masker (forwardmasking conditions). The masker frequencies used were 2.4, 3.2, 3.5, 3.75, 4.0, 4.3, 4.6, and 5.0 kHz. The masking stimulus was set at the same electrical amplitude as the 65 dB SL probe stimulus. Acoustically, all masking stimuli were within 3 dB of the intensity of the probe stimulus except for the 2.0 kHz masker which was 10 dB less intense. In experiment II, the level of the masker relative to the probe was varied. Responses were recorded from 10 normal hearing subjects. Responses were first obtained for the probe alone and then m the presence of the masker. The masker level was either 10 dB below the probe ( - 10 dB), equal in amplitude to the probe (0 dB), or 10 dB greater than the probe ( + 10 dB). In experiment III, the intensity level of the masker and the probe were jointly varied. Responses were recorded from 8 normal hearing subjects. Responses were first obtained to probe alone at 45, 55, and 65 dB SL and subsequently, responses to the probe were recorded in the presence of masker. For each probe level, the masker level was adjusted so that the peak equivalent level of the masker and the probe were the same. In experiment IV, the masker rise-fall time was the independent variable. Responses were recorded from 6 normal hearing subjects. Responses were first obtained for the probe alone and then in the presence of the masker. The masker rise-fall time was 0.05 msec, 1.0 msec, or 10 msec. As rise-fall time ts varied, the total energy also varies if the onset to offset duration of the masker is kept constant. With changes in rise-fall time, the plateau duration of the masker was adjusted to keep total energy constant using the formula given by Dallos and Olsen (1964). T = 2r/3 + P where T is the equivalent duration, r the rise-fall time, and P the duration of the plateau. Thus, all maskers had the same equivalent duration.

Results

Experiment 1. the effects of masker frequency The ABRs obtained from one subject in the

A.K A N A N T H A N A R A Y A N , G M G E R K E N

presence of the various maskers are shown in Fig. 2. It can be seen that waves III and V vary with masker frequency. The mean latency and mean amplitude functions for waves III and V are shown in Fig. 3 It can be seen in Fig. 3 (top hal0 that the latency shifts for wave III tended to be maximal in the vicinity of 4.3 kHz, while the shifts for wave V tended to be maxamal in the vicinity of 3.75 kHz. For both components, maskers with frequencies proximal to the probe produced appreciably greater latency shifts than the more remote maskers, suggesting a frequency dependent forward-masking effect. The mean amphtude changes for waves III and V are plotted in the bottom half of Fig. 3. Wave III showed greater amplitude decrements for masker frequencies at or below 4.0 kHz than for

UM 2.4. 5.2 5.5 5 75 4.0 4_5 46 5_0

Fig 2_ Effect of masker frequency on the ABRs recorded from one subject Waves I, III and V are labeled for the unmasked response (UM) Masker frequency (m kHz) is identified to the right of each trace The calibration markers are 2 msec and 500 nV

ABR FORWARD-MASKING AND ENHANCEMENT

431

~06 I1-h

-'-04 O9 )¢.) Z LIJ I-~[

6O

£ 4O IM

(.9 Z

,~ 2 0 I o I,I

rl~

0

-

~ -2o _J

-40 i

24

!

32 35 MASKER

i

375

|

i

4.3

40

FREQUENCY

!

5.0

utilized to test the significance o f the effects of masker frequency on latency and amphtude. The effect of m a s k e r f r e q u e n c y was tested against each d e p e n d e n t v a r i a b l e (latency, a m p l i t u d e ) a n d for each wave (III, V) i n d e p e n d e n t l y . Results of these analyses i n d i c a t e d that for all maskers, the shifts in l a t e n c y for b o t h waves I I I a n d V were signific a n t ( P < 0.0003). Also, the a m p l i t u d e alterations m waves I I I a n d V were slgmficant ( P < 0.0002). I n o r d e r to d e t e r r m n e if there was a significant difference b e t w e e n the a l t e r a t i o n s p r o d u c e d b y each m a s k e r in e~ther l a t e n c y or a m p h t u d e , the N e w m a n - K e u l s m u l t i p l e c o m p a r i s o n test was utilized. A l l p o s s i b l e pairwlse c o m p a r i s o n s were m a d e for b o t h the d e p e n d e n t variables, a n d for each wave, separately. T h e results i n d i c a t e d that for b o t h wave I I I a n d wave V m a s k e r frequencies b e t w e e n 3.5 a n d 5.0 k H z p r o d u c e d significantly larger shifts in l a t e n c y t h a n the lower m a s k e r f r e q u e n o e s ( 2 . 4 - 3 . 2 kHz). T h e m a g m t u d e of amp h t u d e reductxon for wave I I I was n o t sigmfic a n t l y different across m a s k e r frequency, and the m a g n i t u d e of wave V e n h a n c e m e n t was slgrufic a n t l y greater for m a s k e r frequencies b e t w e e n 3.75 a n d 5.0 k H z than for the lower frequencies.

46 (kHz)

Fig 3 Effect of masker frequency for all subjects Wave III data are plotted with circles, and wave V data, with squares Top mean latency shift (msec) re, the unmasked condition as a function of masker frequency: a pos, twe difference indicates a latency increase Bottom mean amplitude change re the unmasked con&tlon as a funct]on of masker frequency a pos,twe change ln&cates an amphtude increase Vertical bars represent the S_E M m a s k e r frequencies greater t h a n 4.0 kHz. H o w ever, It should b e n o t e d that wave I I I e x h i b i t e d an a m p l i t u d e i n c r e m e n t in s o m e subjects for m a s k e r s a b o v e 4.0 kHz. In s h a r p contrast, the a m p h t u d e b e h a v i o r of wave V (Fig. 3, b o t t o m half) showed a f r e q u e n c y - d e p e n d e n t e n h a n c e m e n t . It c a n be seen that the m e a n m a g m t u d e of enh a n c e m e n t i n c r e a s e d g r a d u a l l y f r o m virtually zero at 2.4 k H z to m a x t m a l values rangrng b e t w e e n 50 a n d 63% for m a s k e r frequencies falling b e t w e e n 3.75 a n d 5.0 kHz. R e p e a t e d m e a s u r e analys~s of v a r i a n c e was

Experiment II." the effects of relattve level of masker A B R s o b t a i n e d in the several test c o n & t l o n s from one subject are shown in Fig. 4. By respecti o n it m a y b e seen that there were l a t e n c y shifts in wave V as the m a s k e r level was increased a n d wave I I I showed a m p l i t u d e r e d u c t i o n at the tugher m a s k e r intensities (0 a n d + 10 dB). W a v e V exh i b i t e d a m p h t u d e e n h a n c e m e n t at all m a s k e r levels. T h e top half o f Fig. 5 depicts the m e a n l a t e n c y shift for b o t h wave I I I a n d wave V as a function of relative m a s k e r level. It can b e seen that the l a t e n c y functions for waves I I I a n d V are essentially parallel. I n b o t h cases, the m a g n i t u d e of l a t e n c y shift increased with increase m m a s k e r level T h e a m p l i t u d e function of wave I I I (Fig. 5, b o t t o m half) is c o m p a t i b l e with its l a t e n c y b e h a v ,or m that there was a progressive d e c r e m e n t in a m p h t u d e as the m a s k e r level was increased O n c e again, a n d in contrast, wave V a m p l i t u d e ( F i g 5, b o t t o m half) e x h i b i t e d e n h a n c e m e n t . T h e m a g m -

432

A K ANANTHANARAYAN,

Ill

G M GERKEN

08 E

UM

I-h

06 ¸

¢.D

-I0

>(..) z lad I-,,~ _1

04'

02

ILl

40

Z "1" (..)

÷L0

Fig 4 Effect of m a s k e r intensity on the A B R s recorded from onc subject W a v e s I, I I I and V are labeled for the u n m a s k e d response ( U M ) The relative level in dB of the m a s k e r r e p r o b e lntens~ty is identified to the right of each trace The c a h b r a h o n m a r k e r s are 2 msec and 500 nV

tude of wave V enhancement is described by a non-monotonic function characterized by maximum enhancement at the 0 dB relative masker level Repeated measure analysis of variance was used to test the significance of the effects of masker intensity on the two dependent variables Independent analyses were done by testing masker intensity against each dependent variable and against data for each wave In summary, the results indicated that the alterations m latency and amplitude for both waves (re., the unmasked condition) were significant across masker intensity ( P < 0.0005) The Newman-Keuls multiple comparison procedure was used to test for significant differences between the effects produced by different masker levels. The results indicated that the differences in

lad t~ X) I'-" ...I n

!

!

!

-I0 0 I0 RELATIVE MASKER LEVEL (dB) Fig 5 Effect of m a s k e r intensity for all subjects W a v e III d a t a are plotted with circles, and wave V data, wath squares Top m e a n latency shift (msec) r e the u n m a s k e d c o n d i t i o n as a function of the relative level of the m a s k e r a positive difference indicates a l a t e n c y increase B o t t o m m e a n a m p h t u d e c h a n g e r e the u n m a s k e d c o n d i t i o n as a function of the relative level of the masker: a posatxve change indicates an a m p h t u d e increase N o t e that at the higher m a s k e r level ( + 10), wave III was Ldentlfied m only 5 of the 10 subj'ects tested Vertical bars represent the S E M

latency and amphtude for each wave were slgmftcant for all possible palrwlse comparisons across masker levels In other words, the magnitude of the change in latency and amphtude for waves III and V is dependent on the relative level of the masker

Experiment III the effects of the lntenstty level of the masker-probe parr The ABRs recorded in one subject at several intensity levels of the masker-probe pair are shown

ABR FORWARD-MASKING AND ENHANCEMENT

433

i n Fig. 6. C o n s i s t e n t w i t h t h e o b s e r v a t i o n s i n t h e preceding experiments, the striking contrast in the b e h a v i o r o f w a v e s I I I a n d V is r e a d i l y a p p a r e n t . A l s o , b y i n s p e c t i o n it a p p e a r s t h a t t h e w a v e V a m p h t u d e e n h a n c e m e n t is m o r e r o b u s t f o r t h e 55 a n d 65 dB, t h a n f o r t h e 45 d B c o n d i t i o n . W a v e I I I w a s n o t a l w a y s d i s c e r n i b l e f o r t h e 45 d B u n m a s k e d c o n d i t i o n , as i n Fig. 6. P l o t t e d i n t h e t o p h a l f o f Fig. 7 a r e t h e m e a n l a t e n c y s h i f t s f o r w a v e s I I I a n d V as a f u n c t i o n o f m a s k e r - p r o b e level. It is e v i d e n t t h a t w a v e I I I a n d wave V latency functions were essentially parallel a n d e a c h s h o w e d a s m a l l i n c r e a s e i n l a t e n c y as t h e

[= ILl.... -r O9

0.6

)-

z ILl I'--

0.4

.--I

A

ta 4 0 t,.9

Z ,,~ "r (,.)

"' r',,

0

I..--I

45

-40

s's MASKER ~ PROBE LEVEL (dB SL)

55

Fig 7 Effect of joint variation of masker and probe miens]ties for all subjects Wave Ill data are plotted with circles, and wave V data, with squares Top mean latency shift (msec) r e the unmasked condxtlon as a function of masker-probe level a positive difference indicates a latency increase Bottom mean amplitude change r e the unmasked con&tlon as a function of masker-probe level a poslave change m&cates an amphtude increase Note that wave III was identified m only a few subjects for the 45 dB SL condmon Vertical bars represent the SEM

levels of the masker creased.

65

Fig 6 Effect of joint variation of masker and probe intensities

on the ABRs recorded from one subject. The probe level m dB SL is identified to the fight of each pair of traces Note that the masker was always the same amphtude as the probe In each pa~r of traces, the top trace represents the unmasked conditton and the bottom trace the masked condmon Waves I, III and V are labeled on one unmasked trace The cahbratlon markers are 2 msec and 500 nV

and probe

were jointly in-

T h e b o t t o m h a l f o f Fig. 7 d e p i c t s t h e m e a n a m p l i t u d e c h a n g e , f o r w a v e s I I I a n d V, as a f u n c t i o n o f m a s k e r - p r o b e level. A g a i n w a v e s I I I and V exhibit dissirmlar behavior. As can be seen i n t h e b o t t o m h a l f o f Fig. 7 t h e a m o u n t o f w a v e V e n h a n c e m e n t i n c r e a s e d f r o m a b o u t 28% a t 45 d B S L t o a m e a n v a l u e b e t w e e n 65 a n d 70% f o r t h e 55 a n d 65 d B S L c o n d i t i o n s . I n s u m m a r y , t h e s e res u l t s s e e m to s u g g e s t t h a t t h e b o t h 55 a n d t h e 65 dB SL conditions produced relatively larger wave V e n h a n c e m e n t t h a n & d t h e 45 d B S L c o n d i t i o n . Analysis of variance measures to test the effect o f j o i n t l y v a r y i n g m a s k e r a n d p r o b e levels o n t h e dependent variables for each wave did not reach

434

A K ANANTHANARAYAN, G M GERKEN

significance at the 0 0 5 level ( P > 0.4701) Since the m a i n effect was not significant, further comparisons b e t w e e n levels was not a p p r o p r i a t e The results of this e x p e r i m e n t are c o m p a t i b l e with the c o n c l u s i o n that the overall level of the maskerp r o b e pair does n o t significantly alter the latency a n d a m p l i t u d e functions for waves III a n d V.

E

0.6

I'= LI.

-r O3

)0 Z

04

ILl

FJ

Experiment I V

the effects of masker rise-fall time

T h e A B R wave form traces o b t a i n e d from one subject at several masker rise-fall times are shown in Fig. 8. Two observations m a y be m a d e from these traces First, It m a y be seen that there was a general r e d u c t i o n In wave III a m p l i t u d e in the masked conditions. Second, there was an enhancem e n t of wave V which seemed to be largest for the 10 msec masker rise-fall time C o m p a r i s o n of the latency behavior of wave III a n d wave V (top half of Fig 9) indicates a parallel

80

4O b.I

(.~ Z 1"

(..) bJ O

0

I._J Q_

-40 I

III i

UM

0,05

~

I00

Fig 8 Effect of m a s k e r rise-fall time on the A B R s recorded from one subject The m a s k e r rise-fall t,me (msec) is identified to the right of each trace W a v e s I, III and V are labeled for the u n m a s k e d trace ( U M ) The c a h b r a t l o n m a r k e r s are 2 msec and 500 nV

i

i

005 I_0 I0 MASKER RISE-FALL TIME (ms)

Fig 9 Effect of masker nse-fall time for all subjects Wave III data are plotted wdh circles, and wave V data, with squares Top mean latency shift (msec) re the unmasked condition as a funcUon of masker rise-fall time a positive difference indicates a latency increase Bottom mean amphtude change re the unmasked condition as a function of masker rise-fall time a positive change indicates an amphtude increase Vertical bars represent the S E M

decrease in the m a g n i t u d e s of the latency shifts for waves III and V as the masker rise-fall time was increased from 0.05 msec to 10.0 msec. The a m p l i t u d e decrement for wave III (Fig. 9, b o t t o m half) is r e a s o n a b l y consistent with the increased latency with respect to a p a r t i a l - m a s k i n g framework. U n l i k e the results in the preceding experiments, the direction of the wave V amplitude change (Fig 9, b o t t o m half) is c o m p a t i b l e with its latency change, in that its m a g n i t u d e increased with decreasing latency shift. Wave V was still enhanced, however, in all conditions. The m a g n i t u d e of e n h a n c e m e n t increased from a b o u t 28% for the 0 0 5 msec masker rise-fall time to

ABR FORWARD-MASKINGAND ENHANCEMENT about 78% for the 10.0 msec masker rise-fall time The results of the repeated measure analysis of variance indicated that changing the masker risefall time produced significant latency and amplitude alterations for both, waves III and V ( P < 0.0007). The Newman-Keuls multiple comparison procedure showed that the latency and amplitude alterauons produced by the 100 msec rise-fall time condition were sxgmficantly different from the alterations produced by the 0.05 and 1 0 msec conditions. However, there was no significant difference between the 0 05 and 1.0 msec conditions. From the above analyses it is reasonable to conclude that increasing the masker rise-fall time results m significantly smaller shifts in latency for waves III and V, lesser amplitude reduction for wave III, and greater enhancement for wave V

Discussion The results of this study relate primarily to 3 alterations in the ABR: alterations that are seen m the comparison of responses recorded in a toneon-tone forward-masking paradigm and of responses recorded in quiet. The alteratxons are" (i) the parallel latency slufts for waves III and V; (i0 the amplitude reduction of wave III, and (m) the amplitude enhancement of wave V In a previous study the latency shifts for waves III and V and the amplitude reduction of wave III were interpreted in terms of partial forward-masking, which involved an increase in latency and an associated amplitude reduction or elimination of a given wave of the ABR. In contrast, wave V amplitude showed an enhancement phenomenon which was incompatible with a forward-masking explanation (Ananthanarayan and Gerken 1983). We will first consider the data for (1) and (n) above that were previously interpreted in terms of partial forward-masking and we will then dxscuss wave W enhancement

Forward-masking effects In experiment I, a restricted range of masker frequencies on either side of the probe produced the larger latency stufts for waves III and V. These frequency-dependent latency shafts for waves III

435 and V cannot be explained m terms of an interaction of the probe and masker on the basdar membrane because there was a temporal separation between the masker and the probe stimuli. Physiologically, forward-masking reflects both receptor adaptation and post-stimulus recovery from this adaptation (Harris and Dallos 1979). Thus, adaptation occurs when the masker 'adapts' the same set or overlapping sets of neural elements that would also respond to the probe The magmtude of adaptation, then, would depend on both the time interval between the masker and the probe, and the extent of overlap in the frequency domain of the neural elements responding to both the masker and the probe Tins viewpoint is supported by Prijs and Eggermont (1981), who concluded from their analysis of narrow-band contributions to the eighth nerve action potenttal that adaptation (as reflected by latency shaft, amplitude reduction, and broader wave forms) was maximal for narrow bands of actwlty that were near the frequency of the tone-burst stimuli Thus, magnitude of forward-masking as reflected by latency shift is dependent on the masker frequency The discrepancy in our data between the frequency of maximal latency shift for waves III and V (Fig. 3 : 4 3 and 3.75 kHz) may be due simply to small sample size and large variability m the data Recall that, for some subjects, wave III was completely masked, hence, data for those subjects could not be presented in Fig. 3. The amphtude reduction for wave III may also be attributed to changes related to partial forward-masking These changes include a reduction in the number of responding elements a n d / o r poor synchrony in the responding elements (Thornton and Coleman 1975; Kevanishvdh and Lagldze 1979). Since the previously noted variability m wave Ili also affects the amplitude measurements, the form of the wave III amphtude function in Fig. 3 is not analyzed further except to note that, in general, wave 1II amplitude was reduced. Increasing the intensity of the masker relatwe to the probe (experiment II) increased the magnitude of the forward-masking effect Similar results have been obtained in several other forward-masking experiments aimed at evaluating the VIIlth

436 nerve action potential a n d / o r the brain-stem evoked responses (Harris 1977; Kramer and Teas 1982; Lasky and Rupert 1982). Harris (1977) found that increasing the masker level relative to the probe not only produced a greater reduction in the magnitude of the probe response, but also prolonged the recovery time His interpretation of the effect of masker level is that probe response magnitude is directly related to the level of excitation evoked by the masker. The higher the masker level, the greater the masker-related activity via recruitment of more basal fibers, and consequently there is not only greater reduction m excltablhty immediately following masker offset, but also the post-stimulus recovery process is lengthened. It is possible that the latency shift and amplitude reduction with increasing masker intensity reflects such an intensification of the adaptation process which m turn elevates the threshold of the hypothetical excitatory process (Coats 1971) Another explanation would be to view the masker intensity effect as a simple attenuation of the probe intensity by the masker. However, differences in the behavior of VIIIth nerve action potential under conditions of stimulus attenuation and a d a p t a t i o n / m a s k i n g have been reported (Coats 1971; Prljs 1980; Prils and Eggermont 1981) These mvesUgators found that the latency shafts produced by masking were smaller than those produced by signal attenuation. Thus, an explanation of the forward-maskang effect in terms of the attenuation of the probe is not acceptable. In the third experiment, it was found that when the intensities of the masker and probe were vaned together, thus maintaining a constant probe-tomasker intensity ratio, the magnitude of masking, as reflected by the latency of wave III and wave V was not significantly different across the 3 levels used. This finding is consistent with other studies winch suggest that it ~s the s~gnal-to-noise ratio which deterrmnes the magnitude of masking and not the overall level. For example, Spoor et al. (1976) investigated the effects of broad-band noise on the tone burst evoked human whole nerve action potentml. Using amplitude as the dependent variable, the effects of the noise were best predicted by signal-to-noise ratio rather than the absolute level. In another study, Lasky and Rupert

A K ANANTHANARAYAN,G M GERKEN (1982) reported that the magmtude of wave V shift in a forward-masking paradigm is 'near-constant' across click intensity for a constant clickto-masker intensity level. Finally, Don et al (1977) found that the magmtude of wave V latency shift with increasing rate of stimulation was constant across chck intensity. The lack of significant statistical differences in the results of this experiment, in contrast with highly sigmficant results m other experiments, would suggest that the forwardmasking effect is largely independent of the overall level of the masker-probe pair. In experiment IV, the latency sinfts for waves III and V decreased as the masker rise-fall time was .increased from 0.05 to 10 msec Also, the amplitude of wave III exhibited partial recovery in the presence of a masker with 10 msec rise-fall time. An upward spread of the forward-masking effects due to the increased spread of acoustic energy in the spectra of the 4.0 kHz masker for the faster rise-fall times may account for the greater shifts m latency for both waves III and V. In other words a 'broad-band masker' would reduce or elirmnate the neural contribution to the probe evoked response over a broader cochlear region than a 'narrow-band masker'. Tins intensification of adaptation at faster rise-fall times is also reflected in the reduced amplitude for wave III Another consistent observation m all our experiments was the essentially parallel latency functions for waves III and V. In keeping with these results, several other investigators have demonstrated greater shifts m latency for waves III and V relative to wave I, using either a repetition rate paradigm (Thornton and Coleman 1975; Martin 1976; Don et al. 1977, Harkins et al. 1979, Gerling 1983) or a forward-masking paradigm (Kramer and Teas 1982; Lasky and Rupert 1982) Harkins et al. (1979) evaluated the effects of stimulus rate on the interwave latencies of various A B R components. They found that the I - I I I interval showed the largest increase with increase in stimulus rate. Martin (1976) noted that wave III and wave V latency shifted in a parallel fashion with increasing stimulus repetition rate. Lasky and Rupert (1982) using a noise-on-chck forwardmasking paradigm also demonstrated parallel

ABR FORWARD-MASKINGAND ENHANCEMENT latency functions for waves III and V These studies regard the greater latency shifts for waves III and V (re., wave I) as reflecting reduced output from the penphery and suggest that adaptation is primarily a peripheral phenomenon. However, to the extent that I - I I I and I - V lnterpeak intervals reflect central changes, the greater latency shifts for waves III and V relative to wave I shift may be interpreted as reflecting additional adaptation within the brain-stem, probably at or caudal to the wave III generator(s).

Enhancement of waoe V In the preceding section, wave III amphtude reduction along with the latency shifts for waves III and V were described primarily in terms of a peripheral forward-masking phenomenon. In this section we will consider the contrasting behavior of wave V amplitude which revealed a robust amplitude enhancement under conditions in which decrements would be predicted. In a previous report (Ananthanarayan and Gerken 1983) we demonstrated that wave V enhancement followed a non-monotonic time course For the several values of At used, maximal enhancement of wave V was obtained at 15 and 45 msec. These results suggested that a time-dependent process mediated enhancement. That this process is also frequency sensitive ~s suggested by the results of experiment I which showed that on a masker frequency continuum, maximum wave V enhancement occurred for a restricted range of frequencies proximal to the probe frequency. Also, wave V enhancement was shown to decrease at higher masker intensity (experiment II) and to increase with longer masker nse-fall times (experiment IV). In addition, the fact that Lasky and Rupert (1982) and Kramer and Teas (1982) faded to demonstrate sigmficant wave V enhancement using noise maskers m their forward-masking expenments suggests the importance of the stimulus spectrum in wave V enhancement. These results taken together describe an enhancement phenomenon which ~s dependent upon the collective effects of several stimulus parameters namely; the value of At, the relative frequencies of masker and probe, the relative intensities of masker and probe, and the masker rise-fall time.

437 Other physiological experiments have also demonstrated enhancement in the evoked responses of the cortex and the thalamo-cortex (Nomoto 1985; Gerken et al. 1986). Typically m these experiments, the brief stimulus which is used to evoke the response is presented during a sustained background (Gummt and Grossman 1961; Gerken 1971, 1973). As with wave V enhancement the magmtude of the evoked potential enhancement is dependent on particular combinations of stimulus parameters (Gerken 1973). While there is little doubt about the presence of response enhancement in the auditory system, ~t ~s not clear if enhancement at different levels of the auditory system revolves a common underlying mechanism. Tins lack of closure is due to factors such as differences in the stimulus configurauons that produce enhancement, differences m the origin of the responses evaluated, species differences, and the differences in the characteristics of the enhancement demonstrated at several levels of the auditory system. In addition, most of the earlier evoked potential studies used anesthetised animals thereby altering a n d / o r obscuring certain normal modes of operation m the nervous system. Thus, assumptions suggesting fundamental similarities in the underlying mechamsm(s) mediating the wave V enhancement, the cortical and thalamo-cortical evoked potential enhancement, and the loudness enhancement are at best tenous. The specific mechanism(s) mediating wave V enhancement is not clear. This is in part due to the fact that there is not only uncertainty concerning the locus of wave V generator(s), but also there remains a lack of certainty concerning the nature of the underlying activity producing wave V. As discussed elsewhere, wave V enhancement does not seem to reflect the simple summation or interference of waves within the ABR (Ananthanarayan and Gerken 1983). As a speculatwe explananon, we propose that wave V enhancement reflects an alteranon in the neuronal interaction between excitation and inhibition in the neural elements that are involved in the generation of wave V. Wave V enhancement can occur if the masker can selectively alter the balance of exotation and inhibition in the direction of less inhibition or more exotation, thereby

438 p r o d u c i n g a state o f e n h a n c e d excitation through 'd~smhJbtnon' O n e p o s s i b l e w a y this d i s l n h i b l t l o n could be p r o d u c e d IS if the a d a p t i v e p r o p e r t i e s of the e x o t a t o r y and i n h i b i t o r y n e u r o n s are different, for example, the e x c i t a t o r y c o m p o n e n t could recover m u c h m o r e r a p i d l y than the i n t u b l t o r y c o m p o n e n t Thus, e n h a n c e m e n t will occur d u n n g the time interval in which the i n h i b i t o r y c o m p o nent is still diminished. Also, such an enhancem e n t will also be influenced b y those stimulus p a r a m e t e r s that alter this interaction. W h t l e there ~s evidence for r a p i d l y recovering excitatory neurons (recovery times ranging f r o m 10 to 50 msec) m the inferior colliculus of rats ( M a r u s e v a 1971), there are no p u b l i s h e d data, to o u r knowledge, on recovery times for i n h i b i t o r y neurons. M a r u s e v a (1971) has, however, suggested that m h i b l t o r y events develop m o r e slowly t h a n e x c i t a t o r y events. W h i l e deflmtlve f o r m u l a t i o n s c o n c e r n i n g the site of wave V e n h a n c e m e n t c a n n o t be d e d u c e d from our data, it would seem that certain p e r i p h eral sites can be ruled out. It is possible that the wave V generator(s) recetves i n p u t via a p a t h w a y not reflected m wave | I I However, the increased l a t e n c y of the wave V w o u l d seem to i n & c a t e that the wave V generator(s) is also driven b y a source affected b y f o r w a r d - m a s k i n g , hence, the enhancem e n t of wave V w o u l d be a central effect T h e c o n t r a s t i n g a m p l i t u d e b e h a v i o r of waves I I I and V further s u p p o r t s the suggestion that e n h a n c e m e n t ts local to the neural acttwty u n d e r l y i n g wave V g e n e r a t i o n and, hence, a central effect. W e c o n c l u d e that wave V e n h a n c e m e n t is media t e d b y a central process, p r e s u m a b l y p r o x i m a l to the wave V generator(s) F u r t h e r m o r e , the enh a n c e m e n t process is s u f f i o e n t l y powerful that it can reverse the p e r i p h e r a l effects of forwardmasking.

R6sum6 Augmentation et r~ductlon de la rJponse audltlve du tronc c~r~bral dans un paradtgme de masquage proactlf Des a l t 6 r a n o n s de la r6ponse 6voqu6e auditlve d u t r o n c c6r6bral ( A B R ) o n t 6t6 e s t i m & s chez 15

A K ANANTHANARAYAN, G_M GERKEN sujets e n t e n d a n t n o r m a l e m e n t en u t i h s a n t phisieurs c o n f i g u r a t i o n s de stimulus dans un p a r a d i g m e de m a s q u a g e p r o a c t l f son sur son. Les p a r a m & r e s de s t i m u l a t i o n manipul6s d a n s cette 6tude c o m p r e n a i e n t , la fr6quence m a s q u a n t e et son xntensit6 relative; l'lntensit6 totale du couple m a s q u a n t et test; et le t e m p s de m o n t & et de redescente du son m a s q u a n t La latence a u g m e n t a l t p o u r les ondes III et V et l ' a m p h t u d e de l ' o n d e I I I d l m m u a l t p o u r certaines c o n d i t i o n s de s n m u l a t l o n . Ces c h a n g e m e n t s o n t 6t6 interpr&6s c o m m e des effets de m a s q u a g e proactff partlel. Les effets de m a s q u a g e ont 6t6. m a x l m a u x p o u r des fr6quences d u son m a s q u a n t tr+s proches de celles d u test, ds a u g m e n t m e n t avec I ' a u g m e n t a t i o n du nlveau du son m a s q u a n t , ds &aient l n d 6 p e n d a n t s du nlveau total du c o u p l e m a s q u a n t - t e s t ; ils d&ro~ssment avec l ' a u g m e n t a tion du t e m p s de m o n t & / d e s c e n t e du son m a s q u a n t . En s o m m e , les effets de m a s q u a g e p r o actlf ont ~t~ interpr6t~s c o m m e & a n t d ' o r i g m e p & i p h 6 r i q u e , bien q u ' u n niveau c o m p l 6 m e n t a l r e au nlveau d u t r o n c c6r6bral ne soit pas h 6carter. En revanche, les mSmes s n m u l u s qul a u g m e n talent les latences des ondes I I I et V e t r6duisaient l ' a m p h t u d e de l ' o n d e I I I o n t p r o d u i t une a u g m e n t a t i o n i m p o r t a n t e de l ' a m p l l t u d e de l ' o n d e V, d6sxgn& c o m m e facihtation. Cette f a c l h t a t i o n de l ' o n d e V a 6t6 m a x i m a l e p o u r des fr6quences de m a s q u e tr6s p r o c h e du test; elle d & r o l s s a i t avec l ' a u g m e n t a t l o n du m v e a u du masque; et, 6galement, avec des temps de m o n t 6 e / d e s c e n t e plus c o u r t du masque. Les processus e n t r a ] n a n t l'augm e n t a t l o n de l ' o n d e V ne sont pas clalrs; cepend a n t on p e u t conclure que cette a u g m e n t a t i o n r6sulte p r o b a b l e m e n t d ' i n t e r a c t i o n s n e u r o n a l e s centrales complexes, a p r o x l m l t 6 semble-t-il d u g6n6rateur(s) de l ' o n d e V. Th~s work was supported m part by Grants 16411 and 19512 from the Natxonal Insntute of Neurological and Commumcatlve D~sorders and Stroke

References Ananthanarayan, A K and Gerken, G M Post-shmulatory effects on the auditory brain stem response parttal-masklng and enhancement Electroenceph chn_ Neurophyslol, 1983, 55 223-226

ABR' FORWARD-MASKING AND ENHANCEMENT Ananthanarayan, A K Effects of lspllateral and contralateral noise on the auditory bramstem evoked response In E J Moore (Ed), Bases of Audttory Brain-stem Evoked Responses Grune and Stratton, New York, 1983" 243-251 Burkard, R and Hecox, K The effect of broadband noise on the h u m a n braanstem audnory evoked response I Rate and intensity effects J. acoust Soc A m e r , 1983, 74 1214-1223 Buytendljk, F J_J and Meester, A Duration and course of auditory sensation Pontif Acad S o , 1942, 6" 557 Coats, A C Physlologacal observations of auditory masking If Effects of masker intensity J Neurophysiol, 1964, 27 1002-1010 Coats, A C Depression of chck action potentials by attenuatxon, cooling and masking Acta oto-laryng (Stockh), 1971, 284 (Suppl). 1-19 Dallos, P and Olsen, W Integration of energy at threshold with gradual rise-fall tone pips J acoust Soc A m e r , 1964, 36 743-751 Davis, H and Derbyshare, A J The m e c h a m s m of auditory masking_ Amer J P h y s m l , 1935, 113 34-42 Don, M , Allen, A R and Starr, A Effect of chck rate on the latency of auditory brmnstem responses m h u m a n s A n n Otol (St Lores), 1977, 86 186-195 Elmaslan, R O and Galambos, R Loudness enhancement with monoaural, binaural, and dlchotic stimulation J acoust Soc A m e r , 1973, 53 335A Gerken, G M Electrophyslologacal observations relevant to maslong Audiology, 1971, 10 97-109 Gerken, G M Enhancement of the medial gemculate evoked response m conscious cat Electroenceph chn Neurophyslol, 1973, 34. 509-519 Gerken, G M , Slmhadri-Sunuthra, R and Bhat, K H V Increase in central auditory responsiveness d u n n g continuous tone stimulation or following h e a n n g loss In R Salvi, D Henderson, R P_ H a m e r m k and V Colletti (Eds), Applied and Basic Aspects of Noise Induced H e a n n g Loss Plenum Press, New York, 1986 195-211 Gerhng, I J Auditory Nerve and Braanstem Response Stimulus Rate Effects from Persons wxth Normal and Impaared Auditory Sensmvaty Doctoral Dissertation, U m v of Texas, Dallas, TX, 1983 G u m m t , R J and Grossman, R_G Potentials evoked by sound m the auditory cortex Amer J Physiol, 1961, 200 1219-1225 Harkms, S W , McEvoy, T M and Scott, M L Effects of mterstimulus interval on latency of the bramstem auditory evoked potential Int J Neuroscl, 1979, 10 7-14. H a m s , D M Fol-ward-Masking and Recovery from Short Term Adaptation m Single Auditory Nerve Fibers_ Research Report, Northwestern O m v , Chtcago, IL, 1977 H a m s , D M and Dallos, P Forward-masking of auditory nerve fiber responses J Neurophysiol, 1979, 42 10831107 Heeox, K and Galambos, R. Brmnstem auditory evoked responses in h u m a n infants and adults Arch_ Otolaryng, 1974, 9 9 : 3 0 - 3 3 Hecox, K , Squires, N and Galambos, R Braanstem auditory

439 evoked responses in m a n I_ Effect of stimulus nse-fall time and duration J acoust Soc A m e r , 1976, 60 1187-1192 Irwin, R J and Zw~slocka, J J Loudness effects in pmrs of tone bursts Percept Psychophys, 1971, 10 189-192 Kevamshvilh, A and Lagldze, Z Recove~ry functmn of the h u m a n bra,ustem auditory evoked potential Audiology, 1979, 18. 472-484 Kramer, S J and Teas, D C Forward masking of auditory nerve (N1) and bramstem (wave V) response m h u m a n s J acoust Soc A m e r , 1982, 72" 795-803 Lasky, R E and Rupert, A_L Temporal m a s k m g of auditory evoked braznstem responses m h u m a n newborns and adults Hear R e s , 1982, 6 315-334 Lev, A and Sohmer, H Sources of the averaged neural responses recorded in ammals and h u m a n subjects d u n n g cochlear auchometry Arch O h r , N a s , u Kehlk Hellk, 1972, 201 79-90 Martin, J I Short Latency Auditory Evoked Potentials m H u m a n s Doctoral Dissertation, Baylor U m v , 1976 Maruseva, A_M Temporal charactenstlcs of the auditory neurons in the inferior colhculus In. G.V Gersuni (Ed), Sensory Processes at the Neuronal and Behavioral Levels Academic Press, New York, 1971 181-200 Nomoto, M Local auditory evoked potentials and effects of pure tones on click evoked potentials m the pigeon Hear R e s , 1985, 17 13-25 Plcton, T W , Hillyard, S A , Krausz, H I and Galambos, R H u m a n auditory evoked potentials_ I Evaluation of components Electroenceph chn Neurophysiol, 1974, 36 179-190 Pratt, H and Sohmer, H Intensity and rate functions of cochlear and bralnstem evoked responses to click stimuli m m a n Arch Otolaryng, 1976, 212. 85-92 Pnjs, V F. On penpheral auditory adaptation II C o m p a n s o n of electrically and acoustically evoked action potentials m the gmnea pig Acustlca, 1980, 45 1 - 8 Prijs, V_F. and Eggermont, J_J Narrow-band analysis of compound action potentials for several stimulus conditions in the guinea pig. Hear R e s , 1981, 4 23-41 Scharf, B , Fundamentals of auditory maslong_ Audiology, 1971, 10: 30-40. Sokohch, W G and Zwlslocka, J J_ Oil loudness enhancement of a tone burst by a preceding tone burst J acoust Soc A m e r , 1972, 52 141A-142A Spoor, A., Eggermont, J_J and Odenthal, D W C o m p a n s o n of h u m a n and a m m a l data concerning a d a p t a u o n and masking of eighth nerve compound action potentials In R Ruben, C Elberhng and G Salomon (Eds), Electrocochleography. Umversity Park Press, Baltimore, MD, 1976 183-198 Suzuki, T and Honuclu, K Rise-time of pure-tone sUmnh in brmnstem response audlometry Audiology, 1981, 20 101-112 Thornton, A R_D and Coleman, M J The adaptation of cochlear and bra.m stem auditory evoked potentials m h u m a n s Electroenceph chn Neurophyslol, 1975, 39 399-406