auditory steady-state responses and hearing device fitting · 2012-08-02 · tion of hearing...

241

CHAPTER 13

Auditory Steady-StateResponses and Hearing

Device Fitting

Part AThe Role of Auditory Steady-StateResponses in Fitting Hearing Aids

FRANZ ZENKER-CASTROJOSÉ JUAN BARAJAS DE PRAT

The advent of universal hearing screen-ing programs has ensured early detectionof hearing impairment at birth. Thisinvolves the fitting of hearing aids at theearliest stage possible. Hearing aid fittingsconstitute the most frequent method ofhearing impairment habilitation in new-borns and very young infants. Early hear-ing aid fitting in infants contributes toacquisition and development of oral lan-guage. For this reason, it is important tohave hearing aid fitting protocols specif-ically designed for very young infants.

These protocols will be dependent onelectrophysiological methods, becausebehavioural audiometry is not viableuntil the age of 5 to 6 months and, insome infants or young children withdevelopmental delay, not possible at all.

In infants and in the adult population,hearing aid fitting consists of three stages:the assessment of hearing sensitivity, theselection of adjustment parameters torestore the hearing perception, and theverification of the prescribed gain foreach patient (Scolie & Seewald, 2001).

13_Rance_241-264 1/13/08 10:09 AM Page 241

Copyright 2008Plural Publishing, Inc.

Assessing hearing implies the establish-ment of hearing thresholds and maximumcomfort and discomfort levels at differ-ent frequencies for each ear independ-ently. The prescription of adjustmentparameters requires the establishment ofthe gain in such a way that the speechspectrum (Cornelisee, Gagné, & Seewald,1991; Zenker, Delgado, & Barajas, 2003)is amplified within the dynamic hearingrange of the patient (Cornelisse, Seewald,& Jamieson, 1995; Zenker & Barajas,1999). Finally, the verification of amplifi-cation allows the clinician to check thatthe objectives set in the prescription of ad-justment parameters have been achieved(Stelmanchiwiccz, Kopun, Mace, Lewis,& Nittrouer, 1995).

For certain difficult-to-test populations,hearing thresholds can be obtained onlythrough electrophysiological measuresthat do not require any voluntary re-sponse from the individual (Goldstein &Aldrich, 1999). Moreover, electrophysio-logical tests can assist those involved inthe adaptation of hearing aids, becausethese tests can measure auditory functionobjectively (Picton et al., 2002). Assess-ment of auditory steady-state responses(ASSRs) provides a quick and objectiveway to establish electrophysiologicalhearing thresholds at different frequen-cies. The ASSRs have several advantagesin their application in hearing aid fitting:First, they provide assessment of hearingthresholds at different frequencies. Sec-ond, from these measurements it is pos-sible to infer the adjustment parametersof hearing aid devices. Third, the acous-tic characteristics of the ASSR’s stimuliallow verification that the hearing aid isfunctioning and that the subject perceivesand discriminates sounds at a brain level(Picton et al., 2002).

Recent studies (Cone-Wesson, Parker,Swiderski, & Rickards, 2002; Picton et al.,1998; Zenker, Fernández, & Barajas, 2006)have proposed that assessment of audi-tory evoked potentials, and specificallyASSRs, could serve as an useful tool in thefitting and verification of the function ofhearing aids. The application of electro-physiological techniques in hearing aidfitting requires measures of (1) the stim-uli, (2) the acoustic characteristics ofexternal auditory canal, and (3) the fre-quency-specific hearing threshold assess-ment. These aspects are reviewed in thefollowing sections.

Estimating Pure-Tone Audiometric Threshold

Through Auditory Evoked Potentials

Hearing aid amplification is based on theindividual characteristics of each patient.In this regard, individual amplification isestablished by gain prescription methods.Cornelisse and colleagues (1995) definedthis prescription process as “a functionthat prescribes hearing devices’ gain at different frequencies related to thepatient’s audiometric values.” The firstgain prescription methods were estab-lished by Watson and Knudsen (1940)and later by Lybarger (1944). At present,prescription methods can be providedeither by hearing aid manufacturers orthrough independent procedures suchas the “Desired Sensation Level” (Seewald,Ross, & Spiro, 1985) or the NationalAcoustic Laboratory approach (Byrne &Tonnisson, 1976).

The manufacturer modules for fittinghearing aids require knowledge of the

242 THE AUDITORY STEADY-STATE RESPONSE

13_Rance_241-264 1/13/08 10:09 AM 42


patient’s hearing sensitivity. Once thisinformation is obtained from electro-physiological procedures, an estimationof behavioural thresholds must be car-ried out. In the case of ASSRs, the estima-tion of hearing thresholds can beinflenced by certain measurement condi-tions such as the stimulus frequency, thedegree of hearing loss, the age of the sub-ject, and the duration of the electrophys-iological test. Some studies indicate thatthe physiological thresholds may bemore accurate in hearing-impaired indi-viduals than in normal-hearing subjects(Dimitrijevic et al., 2002; Rance & Briggs,2002). Two main procedures are used toestimate audiogram thresholds: The firstconsists of obtaining the average differ-ence between ASSRs and behaviouralthresholds; the second procedure con-sists of determining the regression func-tion between the physiological and thepsychoacoustic thresholds for differentvalues of hearing loss (Rance & Briggs,2002; Rance et al., 1995). The ways inwhich behavioural hearing thresholdscan be predicted from ASSR findings arediscussed in detail in Chapter 7.

Frequency Specificity

The adjustment of hearing aids requiresfrequency specific information at hearingthreshold level. The thresholds obtainedfrom the ASSR are at least as accurate and have the same frequency-specificityas that for tone-burst evoked auditorybrainstem responses (ABRs) (Herdman,Picton, & Stapells, 2002). Moreover, theyoffer the advantage that thresholds forseveral carrier frequencies can can bemeasured simultaneously (Lins & Picton,

1995). Up to eight thresholds can be ob-tained in a significantly shorter time thanis typical with use of techniques basedon sequential testing using one stimulusat a time (John et al., 2002).

Stimulus Calibration

The prescription of amplification frombehavioural testing must take into accountthe calibration of the stimuli employed.In this regard, electrophysiological andbehavioural responses can significantly dif-fer for the same patient.These dissimilar-ities can be ascribed to differences in thetype of stimulus and physiological gener-ators involved in the detection of tone.

Modulated tones used during ASSRtesting are similar to warble tones usedin behavioural testing (Rance et al., 1998).This implies that calibration correctionsassociated with tone burst and clicks(used for ABR) are not required, and thestimuli can therefore be presented at lev-els that extend to 120 dB HL. Calibrationof ASSR stimuli can be carried out withthe same standards used for pure tonesemployed in audiometers.This proceduremakes it possible to obtain the thresh-olds in dB HL and thereby to directlyintroduce these values into hearing aidfitting modules. It should be noted thatthe values used in hearing aid fittingalgorithms are based on behaviouralthresholds. An additional step, therefore,is needed to convert the ASSR thresholds(in dB HL) to estimated behaviouralthresholds (also in dB HL). This can bedone by using regression formulas orcorrection factors. These methods andtheir benefits and limitations are reviewedin Chapter 7.

AUDITORY STEADY-STATE RESPONSES AND HEARING DEVICE FITTING 243

13_Rance_241-264 1/13/08 10:09 AM Page 243


Ear Canal Acoustics

Sininger and coworkers (1997) havedemonstrated that part of the differencesfound between the physiological andbehavioural thresholds in infants andadults may be due to the large ear canalresonances associated with small infantear canals.The effect of the ear canal res-onance on the gain prescription of thehearing aid can be minimized by the mea-surement of the real ear to couple differ-ence (RECD). Bagatto and associates(2005) proposed the following equationto obtain the finally estimated dB SPLthresholds at the ear canal level.

dB SPL threshold (ear canal level) =dB HL threshold + insert earphone

RETSPL + RECD

Table 13–1 provides an example of howhearing threshold, in dB SPL, is calcu-lated from ASSR threshold. In this case,the estimated behavioural thresholds atthe eardrum in dB HL can be convertedinto dB SPL by using the real-ear coupler-difference (RECD) plus the referenceequivalent threshold sound pressure levels

(RETSPLS) according to ANSI S3.6-1996.In the example case, values reflect theRECD for a newborn 6 months of age.

Applications of Auditory Evoked Potentials in

Hearing Device Fittings

Several studies have proposed the appli-cation of electrophysiological techniquesat the different stages of hearing aid fit-tings.An essential contribution of the elec-trophysiological technique consists of theverification of the prescribed gain andthe calculations of the adjustment param-eters of the hearing aid derived from theamplitude or latency of the AEPs.

The Amplitude ProjectionProcedure

Kiessling (1982, 1983) published severalarticles describing the clinical applica-tions of the ABRs in hearing aid fitting.He established an objective method basedon the amplitude of the ABR especiallyindicated for noncooperative subjects


Table 13–1. Example worksheet to calculate the hearing threshold, in dB SPL,at the ear canal from the ASSR thresholds in dB HL

500 Hz 1000 Hz 2000 Hz 4000 Hz

Thresholds (dB HL) 40.0 45.0 45.0 50.0(estimated from ASSR)

RECD 6.0 13.0 14.0 18.0

RETSPL correction 5.5 0.0 3.0 5.5

dB SPL threshold (ear canal) 51.5 58.0 62.0 73.5

13_Rance_241-264 1/13/08 10:09 AM Page 244


and infants with multiple disabilities.Theamplitude projection procedure (APP)is an attempt to establish the adjustmentparameters of hearing aids such as thedynamic range, gain, or compression fac-tor. These parameters are derived fromthe wave V amplitude intensity functionof the ABRs. The methodology is basedon the assumption proposed, and neverdemonstrated, by Kiessling, that wave Vamplitude of the ABR correlates with thesensation of loudness. The steepness ofthe amplitude intensity function of theABR described the amount of compres-sion needed for a listener with hearingloss. For example, a patient with a severe-to-profound hearing loss will have ele-vated ABR (or ASSR) thresholds, but theamplitude growth of the response abovethreshold will be minimal. This wouldcorrespond to the limited dynamic rangeof hearing, that is, the difference betweenthreshold and uncomfortably loud sounds,associated with severe-to-profound hear-ing loss. In such cases, greater compres-sion would be indicated, compared withthat for the patient who has a mild-to-moderate loss, with a considerably greaterdynamic range, and who demonstratessome growth of ABR (or ASSR ampli-tude) over a 30- to 40-dB stimulus range.In a subsequent study, Davidson, Wall,and Goodman (1990) studied the rela-tionship between loudness and the ABRwave V amplitude. In this study, greatindividual variation in the amplitude was found. As expected, variability wasreduced as soon as the number of ses-sions increased. This is because variabil-ity in background EEG noise causes greatvariability in the ABR amplitude. If back-ground noise levels are held constant(and low), such as by measuring the esti-mated noise from a statistic like Fsp (Don

et al., 1996), then ABR amplitude databecome less variable and more reliable.Davidson and colleagues reported no sig-nificant differences in the prescription ofthe hearing aid gain and compressionsettings for two out of three hearing-impaired subjects when the APP fittingprocedure was compared with conven-tional prescription methods. Given thathearing aid compression technologiestoday are much different than those exist-ing in 1990, it may be difficult to replicatethis result. Yet the underlying physiolog-ical principles are sound: The evokedpotential threshold can be used to esti-mate the amount of gain needed, and theamplitude growth function provides infor-mation relevant to the listener’s dynamicrange of hearing, for which compressioncan be used as compensation.

Electrophysiological GainVerification

Functional gain is one of the most usedprocedures in the verification of the pre-scribed parameters of hearing aids. Mea-surement of the amount of functional gainis calculated as the difference betweenaided and unaided thresholds at eachspecific frequency obtained through freefield testing (Mueller,Hawkins,& Northen,1992) and is defined as the relative deci-bel difference between the aided andunaided thresholds. Because this tech-nique is based on voluntary behaviouralprocedures, the inherent degree of vari-ability to which behavioural thresholdsmeasurements are subjected will alsoinfluence functional gain measurements(Mueller et al., 1992). Interaction be-tween the test stimuli, the transduceremployed, the hearing aid and the test


13_Rance_241-264 1/13/08 10:09 AM Page 245


room acoustics are well known limita-tions that interfere with the accuratedetermination of hearing thresholds.Moreover, verifying the hearing aid inthe free field does not provide ear spe-cific information. Several attempts havebeen made in order to reproduce thefunctional gain test through electrophys-iological procedures. Electrophysiologytesting has been seen as a procedure thatmay provide an objective method to ver-ify the adjustment of a hearing aid.

Mokotoff and Krebs (1976) publisheda pioneer study in which ABR measure-ments were obtained from adult hearingaid users. Results indicated that aidedABR measurement compared favorablywith aided audiological data. Severalauthors have used the ABR with click ortone burst in order to obtain an objectivemeasurement of the hearing aid response.As described by Mahoney (1985), how-ever, obtaining ABR in the sound fieldwith amplification is a complex issue. Ingeneral, the brief nature of the stimulusemployed during ABR testing showshigh susceptibility to distortion in both:the sound field speaker and the hearingaid amplifier (Hall & Ruth, 1985).

Picton and colleagues (1998) demon-strated that ASSRs can be recorded whenmultiple stimuli are presented simultane-ously through a sound field speaker andamplified with a hearing aid. Therefore,this procedure seems more useful thanprocedures using transient stimuli. Its appli-cation is limited, however, because ASSRcannot provide information regarding howwell the nonlinear processing of the hear-ing aid is benefiting the listener (Picton etal., 2002). Even though ASSR can be usedto assess suprathreshold hearing, it doesnot provide information regarding to howthe aided sound is perceived, followingprocessing in the brain (Picton et al.,2002).

Fitting Hearing Aids from the Auditory Steady-State

Response Testing

Loudness and Auditory Steady-State Responses

Loudness measurement serves two impor-tant clinical functions in audiologicalpractice: to determine the adjustment ofhearing aids (Fabry & Schum, 1994) andto distinguish the site-of-lesion in sensori-neural hearing loss (Hall, 1991). Subjectivejudgements of loudness are often obtainedto define the most comfortable level(MCL) or the loudness discomfort level(LDL). Another method often employedis to obtain loudness judgement over arange of stimulus levels providing infor-mation on how listeners perceive thegrowth of loudness. Finally, a loudnessgrowth function can be derived fromthe intensity values of the stimulus bythe loudness magnitude.

Nonlinear circuits found in digital orwide dynamic range compression (WDRC)hearing aids have incorporated new fit-ting strategies that provide informationabout the loudness growth function overthe range of intensities amplified (Ser-panos, O’Malley, & Gravel, 1997). Theadjustment of non linear hearing aidsinvolves the concept of loudness growthnormalization, where hearing aid featureswould be adjusted for a particular hear-ing loss in order to normalize the percep-tion of loudness (Byrne et al., 2001).From this kind of loudness judgment it ispossible to derive electroacoustic char-acteristics of hearing aids such as aver-age gain, maximum output, compressionratio and onset level (Kiessking, 1982).Much effort has been made in order todevelop adequate methods to measure


13_Rance_241-264 1/13/08 10:09 AM Page 246


loudness, and a series of different loud-ness scaling procedures have been espe-cially proposed for hearing aid fitting,rather than for diagnostic purposes.

Objective methods for estimatingloudness growth have been proposedusing electrophysiological measures. Sev-eral studies have revealed that loudnessgrowth could be estimated using click-evoked ABR (Davidson et al., 1990; Galam-bos & Hecox, 1977; Picton et al., 1977;Thornton, 1987). In these studies, thewave V latency (Rosenhamer et al., 1981),the slope of the latency-intensity function(Galambos & Hecox, 1977; Picton et al.,1977; Thorton, 1987), interaural latencydifferences (Rosenhamer et al., 1981), andthresholds of the ABR (Conjin, Brocaar,& van Zanten, 1990) have been used asindicators of loudness. ABR amplitudehas not been used as latency since it hasgreater intersubject variability (Schwartz,Morris, & Jacobson, 1994).The major dis-advantage of the procedure based on ABRmeasurements is the lack in frequencyspecificity of these responses.

Assessment of ASSRs partly overcomesthe limitations of ABR testing. Becausethe amplitude of the ASSR decreases asthe intensity of the stimulus decreases,ASSRs can be used as an indicator ofloudness. The major limitation of thisprocedure is the variability of the ampli-tude of the ASSRs, because recordings ata given intensity vary from subject tosubject. The recorded amplitude of theASSRs depends on several parameters.Among those the most important are theamount of synchronized current in thegenerators, the orientation of these gen-erators in relation to the recording elec-trodes, and the impedance of the volumeconductor.

An electrophysiological measure ofloudness growth could assist audiolo-

gists in estimating discomfort levels anddetermining hearing aid features. Objec-tive measurement of loudness could beincluded in the prescription of gain inorder to fit hearing aids within the firstfew month of age.

In a recent contribution of our group(Zenker, Barajas, & Fernández, 2005), anattempt was made to prove whether it is possible to establish a relationshipbetween subjective loudness growthderived from the contour test and thephysiological responses obtained fromthe ASSR. The contour test is a clinicalmethod to quantify loudness perception(Cox, Alexander, Taylor, & Gray, 1997).This test was designed to develop a func-tion that describes the growth of loud-ness perception as input levels increasefrom near threshold to uncomfortablyloud levels. In the contour test, tonalstimuli were 5% warble tones presentedat .5, 1, 2, and 4 kHz. Verbal judgementfrom the subject of the perceived loud-ness was required by rating the loudnessin seven categories ranging from verysoft to uncomfortably loud.

The amplitude of the ASSR for eachlevel as a function of frequency is illus-trated in Figure 13–1.

There are differences in the slope ofthe amplitude growth function as a func-tion of frequency.The slopes of the func-tions vary from 0.002 µV/dB (right ear,4000 Hz) to 0.0005 µV/dB (left ear, 1000Hz). (Right-versus-left ear ASSR amplitudedifferences are observed among normalhearing listeners [John, personal com-munication, 2007], but the application ofthis knowledge to clinical diagnosis isyet unclear.) Not only do the slopes varywith frequency, but overall amplitudedoes as well. The largest amplituderesponses are found for 500 and 4000 Hz(at 80 dB HL). The variability between


13_Rance_241-264 1/13/08 10:09 AM Page 247


subjects was fairly similar for a givenstimulus level across carrier test frequen-cies. The variability increased as theintensity level increased. Table 13–2 pro-vides the mean amplitude (and standarddeviation) of ASSR amplitude as a func-tion of frequency, ear, and level.

Figure 13–2a shows the amplitudespectra of ASSRs and the loudness growthfunctions (Figure 13–2b) (from the con-tour test) obtained in a 29-year-old subjectwith normal hearing. The amplitudegrowth functions as a function of fre-quency and the loudness growth func-tions are correlated.

A multiple regression analysis was per-formed using the data from the contourtest (sensation of loudness), level of thestimulus and amplitudes of the ASSR.Themultiple regression analysis resulted in aprediction of loudness estimated from theASSR amplitude and amplitude growthfunctions (Figure 13–3a). This relation-ship can be defined by the equation

Y(f) = B0 + B1 * level + B2 * AMPLITUDE

where B0 = y-interceptB1 = level (dB HL) coefficientB2 = ASSR amplitude coefficient


Figure 13–1. ASSR amplitude as a function of level in normal-hearing subjects.Amplitude growth functions are derived by means of linear regression. As expected,the amplitude of the responses increases with increasing level above threshold.

13_Rance_241-264 1/13/08 10:09 AM Page 248


249

Tab

le 1

3–2.

Mea

n am

plitu

de a

nd s

tand

ard

devi

atio

ns o

f A

SS

Rs

for

each

car

rier

freq

uenc

y an

d ea

ch e

ar (

R a

nd L

).T

heco

rres

pond

ing

mod

ulat

ion

rate

for

each

car

rier

freq

uenc

y al

so is

indi

cate

d.R

espo

nse

ampl

itude

incr

ease

s as

the

stim

ulus

inte

nsity

incr

ease

s fo

r al

l car

rier

freq

uenc

ies.

AS

SR

Mo

du

late

d T

on

es (

Hz)

500

1000

2000

4000

Inte

nsi

ty(d

B H

L)

R(8

1 H

z)L

(77

Hz)

R(8

9 H

z)L

(85

Hz)

R(9

7 H

z)L

(93

Hz)

R(1

05 H

z)L

(101

Hz)

800.

120

(0.0

89)

0.75

9(0

.065

)0.

074

(0.0

40)

0.05

6(0

.034

)0.

055

(0.0

31)

0.07

7(0

.049

)0.

124

(0.0

66)

0.12

2(0

.066

)

700.

110

(0.0

81)

0.84

2(0

.068

)0.

064

(0.0

28)

0.05

6(0

.030

)0.

053

(0.0

24)

0.05

0(0

.035

)0.

084

(0.0

45)

0.06

3(0

.036

)

600.

093

(0.0

90)

0.06

5(0

.070

)0.

056

(0.0

23)

0.05

0(0

.027

)0.

047

(0.0

25)

0.05

4(0

.031

)0.

067

(0.0

37)

0.05

9(0

.037

)

500.

070

(0.0

67)

0.05

2(0

.064

)0.

047

(0.0

25)

0.03

9(0

.025

)0.

053

(0.0

22)

0.04

3(0

.018

)0.

041

(0.0

20)

0.04

3(0

.025

)

400.

044

(0.0

39)

0.02

9(0

.030

)0.

035

(0.0

20)

0.02

6(0

.019

)0.

037

(0.0

17)

0.03

8(0

.017

)0.

030

(0.0

18)

0.03

1(0

.019

)

300.

036

(0.0

30)

0.02

0(0

.015

)0.

026

(0.0

19)

0.02

4(0

.015

)0.

033

(0.0

19)

0.02

8(0

.018

)0.

047

(0.0

86)

0.02

5(0

.014

)

200.

013

(0.0

11)

0.01

6(0

.010

)0.

025

(0.0

13)

0.01

8(0

.011

)0.

022

(0.0

15)

0.02

1(0

.012

)0.

020

(0.0

09)

0.02

4(0

.013

)

100.

020

(0.0

12)

0.01

7(0

.014

)0.

0218

(0.0

14)

0.02

0(0

.009

)0.

015

(0.0

09)

0.01

4(0

.009

)0.

022

(0.0

12)

0.01

4(0

.009

)

13_Rance_241-264 1/13/08 10:09 AM Page 249


250

Fig

ure

13–

2.A

, AS

SR

am

plitu

de s

pect

ra r

ecor

ded

at le

vels

ran

ging

from

80

to 2

0 dB

HL

for

the

four

car

rier

freq

uenc

ies

pres

ente

d to

eac

hea

r.A

rrow

sin

dica

te a

sig

nific

ant

resp

onse

.As

leve

l inc

reas

es,

so d

oes

the

ampl

itude

of

the

AS

SR

.B,

Loud

ness

Con

tour

Tes

t re

sults

are

for

the

sam

e su

bjec

t.

13_Rance_241-264 1/13/08 10:09 AM Page 250


B0, B1, and B2 are coefficient valuesobtained from the multiple regressionanalyses (Figure 13–3b). The sensation of loudness showed a significant correla-tion with ASSR amplitude growth ofbetween 0.82 and 0.85 regardless of thecarrier frequency. The regression line forall frequencies has a slope of −0.43 loud-ness/amplitude * level.

In summary, these data suggest thatloudness growth can be reasonably wellpredicted from ASSR amplitude (at leastin subjects with normal hearing).

Hearing Aid Selection

Regardless of the fitting formula employedto adjust a hearing aid, all procedures

must give information about some criticalparameters. First, hearing dynamic rangemust be established from the pure tonehearing loss and loudness discomfortlevel; second, the hearing aid is supposedto amplify the whole range of speechinto the dynamic range of a particular hear-ing loss; third, the difference betweenthe hearing loss and the lower limit ofthe speech dynamic range provide theamount of gain required by the hearingaid; fourth, the compression factor willbe determined by the degree of hearingloss to relative to the long-term averagespeech spectrum (LTASS) and finally, themaximum power output (MPO), thehearing aid power may be establishedfrom the amplitude growth function ofthe electrophysiological response.


Figure 13–3. A, Loudness judgements plotted as a function of predicted loudnessfrom the equation derived from the multiple regression analysis of ASSR amplitudewith loudness contour. (The regression equation accounts for 70% of total variance.)B, The table shows the result of the multiple regression analysis predicting loudness,with a high correlation index regardless of carrier frequency and ear.

13_Rance_241-264 1/13/08 10:09 AM Page 251


In a recent study, Zenker and col-leagues (2006) proposed an ASSR pre-scription method based on the APP pre-sented earlier in this chapter. Dynamicrange, gain, compression, and MPO ofthe hearing aid were established fromthe level amplitude function of the ASSRs.ASSRs were obtained in a 27-month-oldinfant diagnosed with bilateral moderatesensorineural hearing impairment. ASSRthresholds were 50 dB SPL for 0.5Hz and1.0 kHz; 60 dB SPL for 2 kHz; and 70 dBSPL for 4 kHz. (Converted to dB HL, thethresholds were 55, 50, 63, and 75 dB for0.5, 1.0, 2.0, and 4.0 kHz, respectively.)

Dynamic Range

Figure 13–4 shows the APP from whichthe dynamic range is obtained. In this fig-ure, the previously established amplitudelevel function from a group of normal-hearing subjects (continuous line) is presented with the amplitude level func-tion obtained from a group of hearing-impaired children (dashed line). In thesame figure, the dynamic range of speech(40 to 80 dB) for the 500 Hz is projectedupward from the abscissa to the normalamplitude intensity function. Then, hori-zontally, it is projected to the functionobtained from the hearing-impaired indi-vidual and, finally, projected verticallyupward to yield the output dynamicrange. Hence, the equivalent dynamicrange for this child with a moderate hear-ing loss is 27 dB (84 − 57 dB) for a 40-dBinput range.

Gain and Compression Factor

The difference between the initial pointof the output dynamic range (in thiscase, 57 dB) and the initial point of the

input dynamic range (40 dB) define thegain (57 − 40 = 17 dB) of the hearing aid.The width of the output dynamic rangeis determined from the output range (in this case, 57 to 84 dB, or 27 dB).The need for compression is determinedby the ratio of the output dynamic range(27 dB) to the input dynamic range(40 dB); that is, 27/40 = 0.67.

Maximum Power Output

From the ASSRs obtained for this patient,the MPO can be derived from the loud-ness sensation levels estimated from theamplitude level function. The MPO ofthe hearing aid is determined from theformula

MPO = [Loud − B0 − (B2 * amplitude)]/B1

The values of the regression coefficients(determined from the multiple regres-sion analysis described previously) areshown in Figure 13–3b. The MPO mustbe fitted to the category of uncomfort-ably loud, that is, contour test category 7.The amplitude data (Table 13–3) used inthe formula are those obtained at thehighest test level (80 dB SPL) for thispatient.The MPO settings for this patientare 103, 110, 123, and 121 dB SPL for 0.5,1.0, 2.0, and 4.0 kHz, respectively (seeTable 13–3).

Finally, curves describing hearing aidoutput as a function of input are shownfor each carrier frequency in Figure 13–5.These curves are shown with respect tothreshold (estimated from ASSR thresh-old) and the maximum output (estimatedfrom ASSR amplitude). As shown in thefigure, using these fitting procedures,speech is amplified within the dynamicrange of this patient’s hearing.


13_Rance_241-264 1/13/08 10:09 AM Page 252


253

Fig

ure

13–

4.A

, T

he a

mpl

itude

pro

ject

ion

proc

edur

e (A

PP

).S

olid

line

indi

cate

s th

e am

plitu

de le

vel f

unct

ion

for

norm

al-h

earin

g su

bjec

ts;

dash

ed li

nein

dica

tes

the

ampl

itude

leve

l fun

ctio

n fo

r in

fant

s an

d ch

ildre

n w

ith s

enso

ry h

earin

g lo

ss.T

he in

put s

peec

h dy

nam

ic r

ange

of 4

0to

80

dB is

pro

ject

ed u

pwar

d on

the

nor

mal

cur

ve.T

hen

the

proj

ectio

n fr

om t

he c

urve

for

ears

with

hea

ring

loss

yie

lds

the

outp

ut d

ynam

icra

nge

for

this

pat

ient

.B, T

he g

ain

requ

irem

ent i

s es

timat

ed a

s th

e di

ffere

nce

betw

een

hear

ing

loss

(57

dB

) an

d th

e lo

wer

lim

it of

the

LTA

SS

(40

dB),

or

57 –

40

= 1

7 dB

.The

com

pres

sion

fact

or (

C)

is g

iven

by

the

ratio

of

the

dyna

mic

ran

ge (

84 –

57

= 2

7 dB

) of

the

pat

ient

to

the

norm

al s

peec

h dy

nam

ic r

ange

(80

– 4

0 =

40

dB),

so

C =

0.6

7.

13_Rance_241-264 1/13/08 10:09 AM Page 253


254

Table 13–3. MPO values prescribed from the ASSR for a 27-month-old infantdiagnosed with bilateral moderate sensorineural hearing impairment (see text)

ASSR Modulated Tones (Hz)

500 1000 2000 4000Feature (81 Hz) (89 Hz) (97 Hz) (105 Hz)

Amplitude (µV) 0.07 0.066 0.061 0.058

Loudness category 7 7 7 7

B0 −0.36 −0.26 −0.38 −0.29

B1 4.8 4 −6.18 5

B2 −1.89 −0.51 −1.96 −5.01

B2 × amplitude −0.13 −0.03 −0.12 −0.29

MPO 103 110 123 121

Figure 13–5. Input-output curves prescribed for each carrier frequencystudied. The patient’s hearing threshold, average speech for each carrierfrequency, MPO, and final insertion gain calculated are shown.

13_Rance_241-264 1/13/08 10:09 AM Page 254


Summary

The treatment of those with hearing loss involves the selection and fitting ofamplification devices. In difficult-to-testindividuals, such as young infants, sub-jective and objective measures such asfunctional gain and real-ear probe mea-surements, are not always possible. Forthose subjects who do not provide reli-able responses to behavioural audiometry,the appropriate selection and fitting ofhearing aids requires the establishmentof accurate hearing thresholds by othermeans. ASSR can be used in the charac-terization of hearing loss to estimate theauditory threshold. In addition, the ASSRcan provide information at threshold andalso at suprathreshold levels. ASSRs canbe used to verify and select the adjust-ment of hearing aids.

The role of ASSR in hearing aid fittingsoutlined in this chapter are: the estima-tion of hearing thresholds that can beintroduced in the fitting software andthe prescription of hearing aid featuressuch as MPO, gain and the compressionratio. The procedures described in thischapter need further empirical demon-stration of their effectiveness and accu-racy. As well, the the estimation error of hearing threshold from the ASSR canbe unacceptably large when applied tohearing aid fittings, but in cases whereno other information is available, theASSR thresholds provide a valuable start-ing point.

The APP method described providesestimates of loudness sensation derivedfrom the amplitude level function of theASSR. This procedure provides fre-quency specific information about fea-tures of hearing aids, such as averagegain, compression factor, onset level, and

output limitation. This procedure, baseddirectly on the ASSR amplitudes, coulddeliver an initial adjustment of the hear-ing aid until other behavioural responsescan be obtained.

References

Aoyagi, M., Kiren, T., Furuse, H., Fuse, T.,Suzuki, Y., Yokota, M., et al. (1994). Pure-tone threshold prediction by 80-Hz ampli-tude-modulation following response. ActaOtolaryngologica Suppl, 511, 7–14.

Bagatto, M., Moodie, S., Scollie, S., Seewald, R.,Moodie, S., Pumford, J., et al. (2005). Clini-cal protocols for hearing instrument fittingin the Desired Sensation Level method.Trends in Amplification, 9(4), 199–226.

Burkard, R. F., & Secor, C. (2002). Overviewof auditory evoked potentials. In J. Katz(Ed.), Handbook of clinical audiology(pp. 233–248). New York: LippincottWilliams & Wilkins.

Byrne, D., Dillon, H., Ching, T., Katsch, R., &Keidser, G. (2001). NAL-NL1 procedure forfitting nonlinear hearing-aids: Characteris-tics and comparisons with other proce-dures. Journal of the American Academyof Audiology, 12(1), 37–51.

Cone-Wesson, B., Parker, J., Swiderski, N., &Rickards, F. (2002). Full-term and prema-ture neonates. Journal of the AmericanAcademy of Audiology, 13, 260–269.

Conjin, E. A. J. G., Brocaar, M. P., & van Zan-ten, G. A. (1990). Frequency specificity ofthe auditory brainstem response elicitedby 1000 Hz filter clicks. Audiology, 29,181–195.

Cornelisse, L. E., Seewald, R. C., & Jamieson,D. G. (1995). The input output formula:A theoretical approach to the fitting ofpersonal amplification devices. Journal ofthe Acoustical Society of America, 97,1854–1864.

Cornelisse, L. E., Gagné, J. P., & Seewald, R. C.(1991). Ear level recordings of the long-


13_Rance_241-264 1/13/08 10:09 AM Page 255


term average spectrum of speech. Ear andHearing, 12, 47–54.

Cornelisse, L. E., Seewald, R. C., & Jamieson,D. G. (1995). The input/output formula:A theoretical approach to the fitting ofpersonal amplification devices. Journal of the Acoustical Society of America, 97,1854–1864.

Cox, R. M., Alexander, G. C., Taylor, I. M., &Gray, G. A. (1997). The contour test ofloudness perception. Ear and Hearing,18(5), 388–400.

Davidson, S. A., Wall, L. G., & Goodman, C. M.(1990). Preliminary studies on the use ofan ABR projection procedure for hearing-aid selection. Ear and Hearing, 11,332–339.

Dimitrijevic, A., John, M. S., van Roon, P., Pur-cell, D. W., Adamonis, J., Ostroff, J., et al.(2002). Estimating the audiogram usingmultiple auditory steady-state responses.Journal of the American Academy ofAudiology, 13(4), 205–224.

Don, M. & Elberling, C. (1996). Use of quan-titative measures of auditory brainstemresponse peak amplitude and residualbackground noise in the decision to stopaveraging. Journal of the Acoustical Soci-ety of America, 99(1), 491-499.

Fabry, D. A., & Schum, D. J. (1994). The roleof subjective measurement techniques inhearing-aid fittings. In M. Valente (Ed.),Strategies for selecting and verifyinghearing-aid fittings (pp. 136–155). NewYork: Thieme.

Galambos, R., & Hecox, K. (1977). Clinicalapplications of the brain stem auditoryevoked potentials. In J. E. Desmedt (Ed.),Auditory evoked potentials in man. Psy-chopharmacology correlates of EPs.Progress in Clinical Neurophysiology, 2,1–19. Basel, Switzerland: S. Karger.

Goldstein, R., & Aldrich,W. M. (1999). Evokedpotential audiometry: Fundamentalsand applications. Boston: Allyn & Bacon.

Hall, J. W., & Ruth, R. A. (1985). Acousticreflexs and auditory evoked responses inhearing-aid selection. Seminars in Hear-ing, 6, 251–277.

Hayes, D., & Jerger, J. (1982). Auditory brain-stem response (ABR) to tone-pips: Resultsin normal and hearingimpaired subjects.Scandinavian Audiology, 11, 133–142.

Herdman, A. T., Picton, T. W., & Stapells, D. R.(2002). Place specificity of multiple audi-tory steady-state responses. Journal of the Acoustical Society of America, 112,1569–1582.

John, M. S., Purcell, D. W., Dimitrijevic, A., &Picton,T.W. (2002).Advantages and caveatswhen recording steady-state responses tomultiple simultaneous stimuli. Journal ofthe American Academy of Audiology,13(5), 246–259.

Kiessling, J. (1982). Hearing-aid selection bybrainstem audiometry. Scandinavian Audi-ology, 11(4), 269–275.

Kiessling, J. (1983). Clinical experience inhearing-aid adjustment by means of BERamplitudes. Archives of Otorhinolaryn-gology, 238(3), 233–240.

Lins, O. G., Picton, T. W., Boucher, B. L.,Durieux-Smith,A.,Champagne, S.C.,Moran,L. M., et al. (1996). Frequency-specificaudiometry using steady-state responses.Ear and Hearing, 17, 81–96.

Lybarger, S. (1944). Method of fitting hear-ing-aids. U.S. Patent Applications S.N. 543,278, July 3, 1944.

Mahoney, T. M. (1985). Auditory brainstemresponse hearing-aid applications. In J. T.Jacobson (Ed.), The auditory brainstemresponse (pp. 349–370). San Diego, CA:College-Hill Press.

Mokotoff, B., & Krebs, D. F. (1976). Brainstemauditory-evoked responses with amplifica-tion [Abstract]. Journal of the AcousticalSociety of America, 60, S16.

Muller, H. G., Hawkins, D. B., & Northern, J.L. (1992). Probe microphone measure-ments: Hearing-aid selection and assess-ment. San Diego, CA: Singular.

Picton,T.W., Dimitrijevic,A., van Roon, P., John,M. S., Reed, M., & Finkelstein, H. (2002).Possible roles for the auditory steady stateresponses in fitting hearing-aids. In R. C.Seewald & J. S. Gravel (Eds.), A soundfoundation through early amplification


13_Rance_241-264 1/13/08 10:09 AM Page 256


2001. Proceedings of the 2nd Inter-national Conference (pp. 63–73). Basel:Phonak.

Picton, T. W., Durieux-Smith, A., Champagne,S., Whittingham, J., Moran, L., Giguére, C.,et al. (1998). Objective evaluation of aidedthresholds using auditory steady stateresponses. Journal of the American Acad-emy of Audiology, 9, 315–331.

Picton,T.W., Ouellette, J., & Hamel, G. (1979).Brainstem evoked potentials to tone pipsin notched noise. Journal of Otolaryngol-ogy, 8, 289–314.

Picton, T. W., Woods, D. L., Barbeau-Braun, J.,& Healey,T. M. G. (1977). Evoked potentialaudiometry. Journal of Otolaryngology,6, 90–119.

Purdy, S. C., & Abbas, P. J. (1989). Auditorybrainstem response audiometry using lin-early and Blackman gated tonebursts.American Speech-Language-Hearing Asso-ciation, 31, 115–116.

Rance, G., & Briggs, R. J. (2002). Assessmentof hearing in infants with moderate to pro-found impairment:The Melbourne experi-ence with auditory steady-state evokedpotential testing. Annals of Otology, Rhi-nology, and Laryngology Supplement,189, 22–28.

Rance, G., Rickards, F. W., Cohen, L. T., DeVidi, S., & Clark, G. M. (1995).The auto-mated prediction of hearing thresholds insleeping subjects using auditory steady-state evoked potentials. Ear and Hearing,16, 499–507.

Rance, G., Rickards, F. W., Beer, D. E., & Clark,G. M. (1998). Steady state evoked potentialand behavioural hearing thresholds in agroup of children with absent click evokedauditory brainstem response. Ear andHearing, 19, 48–61.

Rosenhamer, H. J., Lindstrom, B., & Lund-borg,T. (1981). On the use of click-evokedelectric brainstem responses in audiologi-cal diagnosis. IV: Interaural latency differ-ences (wave V) in cochlear hearing loss.Scandinavian Audiology, 10, 67–73.

Savio, G., Cardenas, J., Perez-Abalo, M., Gon-zalez, A., & Valdes, J. (2001). The low and

high frequency auditory steady state re-sponses mature at different rates. Audiol-ogy and Neuro-Otology, 6, 279–287.

Schwartz, D. M., Morris, M. D., & Jacobson, J.T. (1994). The normal auditory brainstemresponse and its variants. In J. T. Jacobson(Ed.), Principles and applications inauditory evoked potentials (pp. 123–154).Needham Heights, MA: Allyn & Bacon.

Seewald, R. C., Ross, M., & Spiro, M. K.(1985). Selecting amplifications character-istics for young hearing-impaired children.Ear and Hearing, 6, 48–51.

Serpanos, Y. C., O’Malley, H., & Gravel, J. S.(1997).The relationship between loudnessintensity functions and the Clic-ABR wave Vlatency. Ear and Hearing, 18(5), 409–419.

Sininger, Y. A., Abdala, C., & Cone-Wesson, B.(1997). Auditory threshold sensitivity ofthe human neonate as measured by theauditory brainstem response. HearingResearch, 104, 27–38.

Stapells, D. R. (2000). Threshold estimation of the tone evoked auditory brainstemresponse. A literature meta-analysis. Jour-nal of Speech-Language Pathology andAudiology, 24(2), 74–82.

Stapells, D. R., Gravel, J. S., & Martin, B. E.(1995).Thresholds for auditory brain stemresponses to tones in notched noise frominfants and young children with normalhearing or sensorineural hearing loss. Earand Hearing, 16, 361–371.

Stapells, D. R., Picton, T. W., & Durieux-Smith,A. (1994). Electrophysiological measuresof frequency-specific auditory function.In J. T. Jacobson (Ed.), Principles andapplications in auditory evoked poten-tials (pp. 251–282). Needham Heights,MA: Allyn & Bacon.

Stelmachowicz, P. G., Kopun, J., Mace, A.,Lewis, D. E., & Nittrouer, S. (1995). Theperception of amplified speech by listen-ers with hearing loss: Acoustic correlates.Journal of the Acoustical Society ofAmerica, 98, 1388–1399.

Thorton, A. R. D. (1987). Electrophysiologicalmeasures of hearing disorders. BritishMedical Bulletin, 43, 926–939.


13_Rance_241-264 1/13/08 10:09 AM Page 257


Vander Werff, K. R., & Brown, C. J. (2005).Effect of audiometric configuration onthreshold and suprathreshold auditorysteady-state responses. Ear and Hearing,26(3), 310–326.

Watson, N., & Knudsen, V. (1940). Selectiveamplification in hearing-aids. Journal ofthe Acoustical Society of America, 18,401–408.

Zenker, F., Barajas, J. J., Meco, G., & Espinosa,S. (1996). Auditory brainstem responseto 1000 Hz filtered tone burst in normalhearing subjects. Presented at the XXIIIInternational Congress of Audiology, Bari,Italy.

Zenker, F., Fernández, R., & Barajas, J. J.(2005). The contour test electrified. Pre-sented at the VII European Federation ofAudiology Societies (EFAS) Congress,Goteborg, Sweden.

Zenker, F., Delgado, J., & Barajas, J. J. (2003).Acoustic characteristics and audiologicalapplications of the Long-Term SpeechSpectrum. Revista de Logopedia, Fonia-tría y Audiología, 23(2), 13–20.

Zenker, F., Fernández, R., & Barajas, J. J.(2006). Fitting hearing-aids in early child-hood based on auditory evoked potentialsin steady states. Acta Otorrinolaringológ-ica Española, 57(9), 388–393.


13_Rance_241-264 1/13/08 10:09 AM Page 258


259

Part BFitting Cochlear Implants Using

Electrically Evoked Auditory Steady-State Responses

BARBARA CONE-WESSON

Infants with severe to profound congen-ital hearing loss are now undergoingcochlear implant surgery at the age of12 months or younger.This sophisticatedhearing technology requires a fair amountof “input data” about the user’s hearingsensitivity, comfortable and uncomfort-able listening levels, and, ideally, speechperception abilities. The limitations ofbehavioural techniques in the infantpopulation have spurred research aimed at determining auditory abilities fromelectrophysiological responses includingtransiently evoked (click and tone-burst)and steady-state evoked brainstem and cor-tical auditory evoked potentials (CAEPs).A large part of this research has usedacoustic stimuli, although there is nowan evidence base for using electricalstimulation (through the implant elec-trode) to evoke these responses for use

in estimating hearing abilities in personswith cochlear implants.

Vander Werff and associates (Chapter 7)discuss the strengths and weaknesses ofusing ASSRs to determine hearing sensi-tivity, the starting point for any hearingaid or cochlear implant prescriptive pro-cedure. Dimitrijevic and Cone-Wesson(Chapter 12) describe ways in whichASSR may be used to infer suprathresh-old hearing abilities, some of which maybe relevant to determining the benefitsof amplification or cochlear implantation.In the first segment of this chapter (“TheRole of Auditory Steady-State Responses inFitting Hearing Aids”), Zenker and Bara-jas propose a procedure for using (acous-tically evoked) ASSR testing for fittinghearing aids. This method uses ASSRthreshold, amplitude, and amplitudegrowth data to estimate the hearing aid

13_Rance_241-264 1/13/08 10:09 AM Page 259


gain, maximum power output (MPO), andcompression as a function of frequency.

Would there be any benefit to usingelectrically evoked ASSRs for the pur-pose of fitting (“mapping”) a cochlearimplant? Have there been any attemptsto do so? The answer to the first questionis best considered in light of the expe-rience with other electrically evokedauditory potentials, such as eighth nerveaction potentials (Abbas, Brown, Hughes,& Gantz, 1999; Brown, 2003; Hughes,Brown,Abbas,Wolaver, & Gervais, 2000a),ABR (Hughes, Brown, Abbas, Wolaver, &Gervais, 2000b; Kileny & Zwolan, 2004),middle latency and cortical evoked poten-tials (Kileny, 1991; Kileny, Zwolan, Boerst,& Telian, 1997; Sharma, Dorman, & Spahr,2002). The answer to the second ques-tion is a qualified yes, and the one pub-lished attempt is described subsequently.

The need for electrophysiologicalmethods for the purposes of verificationof implant function and to determineelectrical stimulation levels (i.e., thresh-old, comfortable and uncomfortableloudness settings) has been addressed bycochlear implant manufacturers. Themanufacturers have developed the hard-ware and software for recording eighthnerve action potentials through theimplant itself in conjunction with a com-puter interface used for the implantdevice programming. That is, the surgi-cally implanted cochlear implant elec-trode to provide an electrical stimulus tothe auditory nerve (spiral ganglion cells)and also as the recording electrode forthe compound nerve action potentialresulting from the electrical stimulus.The computer interface provides amethod for display of the potentials andsome rudimentary analyses. It is possibleto determine the (electrical stimulus)

threshold of the auditory nerve actionpotential for a given electrode location.Such electrically-evoked auditory nerve-action potential (E-AP) thresholds can beused to guide initial stimulation settings.It is the case that the E-AP threshold doesnot approximate perceptual (electricalstimulation) threshold, but the presenceof the E-AP indicates that the eighthnerve is stimulable at that electrode site.Furthermore, the E-AP threshold and theperceptual threshold have a fairly con-stant relationship across the electrodearray, so determination of the amount of“offset” of the electrical from the percep-tual threshold at one electrode site canbe used to estimate the perceptualthreshold across the array (i.e., set the“T” level).This is the major clinical appli-cation of the E-AP: to estimate perceptualthreshold for the purpose of mapping. Inresearch, the E-AP also can be used toinvestigate refractoriness/adaptation forelectrical stimulation, and to estimate thespread of current in the cochlea whichhas implications for designing more(spectrally) precise stimulation schemes.

Given the fact that E-AP technologiesare well developed, is there a need forelectrically evoked ASSR (E-ASSR)? TheASSR for modulation rates of 70 Hz andabove is primarily a brainstem response.But is there any advantage for E-ABR overE-AP? The experience with E-ABR (Brown,2002; Hughes et al., 2000b) is similar to that with E-AP: The threshold of theE-ABR falls within the dynamic range ofthe listener but may well be closer to thecomfortable listening levels than it is toperceptual threshold. Yet the presenceof an E-ABR indicates that the stimulusprovided by the implant is stimulatingbrainstem auditory nuclei and pathways.So the presence of an E-ASSR (for high


13_Rance_241-264 1/13/08 10:09 AM Page 260


modulation rates), in conjunction withan E-AP, would give additional informa-tion with regard to brainstem auditoryfunction but probably would not provideany additional information that could beused for device programming. Further-more, unlike acoustically evoked ABRs oraction potentials, in comparison with theASSR, the issue of frequency specificity isnot relevant, because the cochlear placeof stimulation is determined by the choiceof implant electrode stimulated.

Firzst and associates have completedseminal work on electrically evoked ABR(E-ABR), middle latency response (E-MLR),and E-CAEP in deaf adults who useimplants to hear (Firszt, Chambers, &Kraus, 2002; Firszt, Chambers, Kraus, &Reeder, 2002). The work of these inves-tigators shows that the latency andamplitude of the E-MLR had the highestcorrelations (among the E-AEPs) withspeech perception abilities. This findingsuggests that the E-ASSR for carriersmodulated at 40 Hz, for which the gen-erators are thought to be the same asthose for MLR, may provide some infor-mation relevant to speech perceptionabilities. This is speculative at best.

The overriding technical issue inrecording an evoked potential usingelectrical stimulation is that of electricalartifact created by the implant. Single or several cycles of biphasic electricalpulses are used to evoke transient poten-tials such as the E-AP or E-ABR. Eventhen, the stimulus artifact often swampsthe electrophysiological response. Toovercome artifact issues, the E-AP isderived from a forward masking para-digm in which the electrical artifacts of both the masker and the probe aresubtracted from the average containing a response to the probe stimulus. In

the case of E-ABR, the stimulus artifactobscures early components of theresponse, and wave V may be the onlycomponent visualized. E-ABRs and laterresponses such as E-MLR and E-CAEPalso require the use of radiofrequency filters and electrophysiological preampli-fiers that are not overloaded by the electrical artifact created by the implant.For E-ASSR, the biphasic pulses must bepresented in a steady-state (continuous)manner. This means that the stimulusartifact is present during the entirerecording epoch and is orders of magni-tude greater than the response.

Menard and colleagues (2004) tackledthe technical problem of recording ASSRsfor electrical stimulation provided by animplant. The Lyon group recorded ASSRsto amplitude-modulated steady-state bi-phasic pulses electrical pulses providedby an MXM Digisonic cochlear implant.The modulation rates were between 70and 85 Hz. These investigators variedboth pulse width (duration) and inten-sity to derive thresholds and responseamplitude input-output functions. Theyreasoned that stimulus artifact wouldshow a linear growth with intensity andduration, whereas physiological responseswould show nonlinear growth functions.In this way, they estimated the portion ofthe response that was physiological ver-sus the artifactual portion. This methodtook advantage of the charge densitybasis of neural excitability. Of note, thismethod of estimating the portion ofresponse due to stimulus artifact andthat portion due to neural excitation wasused only for higher-intensity and longer-duration stimuli. The investigators madean assumption that the responses ob-tained for short pulse width durationswere free of artifact. They compared the


13_Rance_241-264 1/13/08 10:09 AM Page 261


E-ASSR thresholds (in this case, the min-imum pulse width duration that resultedin an ASSR) with perceptual thresholdsobtained as a function of pulse width dura-tion. A linear relationship was observedbetween perceptual threshold and ASSRthreshold, and the differences betweenthe two measures were on the sameorder of magnitude as those seen forE-ABR. This is not surprising: With themodulation rates used, the E-ASSR wasgenerated primarily at the brainstem.The investigators further suggest that theE-ASSR input-output functions show sat-uration, and this saturation point mayprove useful in the estimation of comfort-able listening levels but is yet untested.(This concept is similar to that underly-ing the Zenker and Barajas method [seeChapter 13] of using ASSR amplitude toestimate MPO for hearing aids.)

The question remains: Is there anadvantage to using E-ASSR over otherevoked potential methods? One advan-tage may be that several electrode sitescould be tested simultaneously by usingdifferent modulation rates for each. Giventhat programming cochlear implantsrequires the setting of threshold andcomfortable listening levels for up to 22electrodes, this could be advantageous incomparison with the single electrodetest techniques (such as E-AP or E-ABR).This advantage may be outweighed bythe technical demands of separating thesteady-state stimulus artifact from thesteady-state response.

References

Abbas, P. J., Brown, C. J., Hughes, M. L., &Gantz B. J. (2000). Electrically evoked com-pound action potentials (EAP) recorded

from subjects who use the NucleusC124M device. Annals of Otology, Rhinol-ogy, and Laryngology, 185, 6–9.

Brown, C. J. (2003). Clinical uses of electri-cally evoked auditory nerve and brain-stem responses. Current Opinion in Oto-laryngology and Head & Neck Surgery,11(5), 383–387.

Firszt, J. B., Chambers, R. D., & Kraus, N.(2002). Neurophysiology of cochlear im-plant users. II: Comparison among speechperception, dynamic range and physiolog-ical measures. Ear and Hearing, 23,516–531.

Firszt, J. B., Chambers, R. D., Kraus, N., &Reeder, R. M. (2002). Neurophysiology ofcochlear implant users. I: Effect of stimu-lus current level and electrode site on theelectrical ABR, MLR and N1-P2 response.Ear and Hearing, 23, 502–515.

Hughes, M. L., Brown, C. J., Abbas, P. J.,Wolaver, A. A., & Gervais, J. P. (2000a).Comparison of EAP thresholds with MAPlevels in the Nucleus 24 cochlear impant:Data from children. Ear and Hearing, 21,164–174.

Hughes, M. L., Brown, C. J., Abbas, P. J.,Wolaver, A. A., & Gervais, J. P. (2000b). Therelationship between EAP and EABRthresholds and levels used to program theNucleus 24 speech processor: Data fromadults. Ear and Hearing, 21, 151–163.

Kileny, P. R. (1991). Use of electrophysiologicmeasures in the management of childrenwith cochlear implants: Brainstem, middlelatency and cognitive (P300) responses.American Journal of Otology, 12, 37–41.

Kileny, P. R., & Zwolan, T. A. (2004) Pre-operative transtympanic electrically evokedauditory brainstem response in children.International Journal of Audiology,43(Suppl. 1), S16–S21.

Kileny, P. R., Zwolan,T.A., Boerst,A., & Telian,S. A. (1997) Electrically evoked auditorypotentials: Current clinical applications inchildren with cochlear implants. Ameri-can Journal of Otology, 18, S90–S92.

Menard, M., Gallego, S.,Truy, E., Berger-Vashon,C., Durrant, J. D., & Collet, L. (2004). Audi-


13_Rance_241-264 1/13/08 10:09 AM Page 262


tory steady-state thresholds in cochlearimplant patients. International Journalof Audiology, 43(Suppl. 1), S39–S43.

Sharma, A., Dorman, M. F., & Spahr, A. J.(2002). A sensitive period for the develop-

ment of the central auditory system inchildren with cochlear implants: Implica-tions for age of implantation. Ear andHearing, 23, 532–539.


13_Rance_241-264 1/13/08 10:09 AM Page 263


13_Rance_241-264 1/13/08 10:09 AM Page 264


auditory steady-state responses and hearing device fitting · 2012-08-02 · tion of hearing...

Documents