RELATION AND THE

BETWEEN SOUND INTENSITY LATENCY AND AMPLITUDE OF THE BRAINSTEM AUDITORY EVOKED RESPONSE

JANIS A. WOLFE

Century Medical Plaza, Tucson, Arizona PAUL SKINNER

University of Arizona, Tucson JOHN BURNS

Hughes Aircraft, Los Angeles, California

This study investigated the relation of peak amplitude and latency to signal intensity for the brainstem auditory evoked response (BSAER). One thousand clicks were presented to obtain each averaged response. Responses were obtained to clicks presented at sensation levels of 15, 20, 30, 40, 50, 60, and 70 dB. Five adult males who demonstrated normal hearing served as subjects. Latency and amplitude for various wavelets were plotted against signal intensity. A consistent trend of decreased peak latency occurred with increased signal intensity. Contrary to previous reports, the amplitude of Wavelet V showed a linear growth with increased signal intertsity. A number of investigators have considered the relation between signal intensity and peak amplitude of the late response (Keidel and Spreng, 1965; Davis and Zerlin, 1966; Davis, Bowers, and Hirsh, 1968; Antinoro, Skinner, and Jones, 1969) and the middle response (Antinoro, Shimota, and Skinner, 1970; Madell, 1969; Thornton, Mendel, and Anderson, 1977.) The brainstem auditory evoked response (BSAER) was descxibed by Jewett and Williston (1971) as a series of seven wavelets with a latency range of about 4 to 9 msec, depending on stimulus level and subject age (Hecox and Galambos, 1974). These investigators found no trend between signal intensity and response amplitude for the various wavelets. A report by Starr and Achor (1975), however, indicated that Wavelet V increased in a fairly linear manner with increased stimulus sensation level. Our study investigated systematically the relation of peak amplitude and latency to signal intensity for the BSAER. 401

Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx

METHOD

Instrumentation Ongoing electroencephalic activity was picked up by silver-disc electrodes attached to the vertex and earlobe. A ground electrode was attached to the opposite earlobe. The neuroelectrical potentials were amplified by preamplifiers (Tektronix, Model 122) at a gain of 106 with a band pass set at 80 to 3425 Hz (at 3-dR down points) before oscilloscopic display and a computer analysis of averaged transients (Nmemotron Model 400 C). The summed responses were written out on an X-Y recorder (Hewlett-Packard Model 7053 A). Signal presentation and control were maintained as follows: A waveform generator (Krohn-Hite, 5100 A) was used to synchronize the sweep of the computer and to trigger a pulse generator (Tektronix, Model 162). Acoustic transients (clicks) were produced by the pulse generator with a rise time of about 0.5/zsec and a duration of 100 ~sec at a rate of approximately 10/sec. The signals were lead to the necessary attenuation and matching network and presented monaurally through an earphone (Sharpe HA-10 A).

Procedure A total of 1000 clicks was presented to obtain each averaged response. Responses were obtained without stimuli (control conditions) and with clicks presented at sensation levels of 15, 20, 30, 40, 50, 60, and 70 dR. The order of presentation of different sensation levels was randomized. Data were collected usually for three replications at each sensation level during three different test sessions.

Subjects Five young male undergraduate students served as subjects. All subjects demonstrated normal hearing levels from 250 to 8000 Hz. Zero-dR SL was established for each individual and this level was used as a reference in all sessions. The subjects reclined on a cot in a double-walled, sound-treated booth. They were asked to remain still throughout the procedure and to minimize neck and facial movement. A pillow was provided for a head rest. All subjects but one reported that they slept during most of the session. RESULTS The emergence and development of the various wavelets were observed as the stimulus sensation level was increased from 15- to 70-dR SL. The lowest sensation level at which Wavelets IV, V, and VI clearly emerged was 15 dR, however all subiects did not yield clear wavelets consistently at that level. As reported by others, Wavelet V was the most prominent and consistent. Wavelets IV and VI frequently were not detectable at 15-dR SL. Wavelets I through 402 1ournal of Speech and Hearing Research

21 401-407 June 1978

Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx

VII could be observed at 20-dB SL and wavelet V was prominent. At sensation levels of 40 dB and greater, all seven wavelets were distinctive. The emergence of the BSAER to increasing stimulus sensation levels for one subject is shown in Figure 1. A complete set of responses was obtained for all but one subject at all sensation levels for three sessions. The data were pooled for analysis based on equal sample numbers. Wavelet latency was plotted against signal intensity at each dB sensation level. These data appear in Figure 2. Except for latency values recorded for the lower sensation levels, a distinctive trend of decreased latency for increased intensity appeared for all seven wavelets. Group means and standard deviations for wavelet latency with intensity for all wavelets except VII are presented in Table 1. dB SL

-9"rr i

~

~

LATENCY GROUPMEANS

7O -vr

6O

"v-

50~

6

E

"rv"

Z

4O

-- 5 )Z w F~ 4

Ii

2.0

_J

r.,..

6.0

100

HD

(Msec)

1SO

1 10

I I 15 20

l 30

1 40

t 50

1 60

I 70

dB re SENSATION LEVEL

FICURE 1. ( L e f t ) A selected set of responses that shows the growth of the BSAER with increasing stimulus sensation levels. FICU]E 2. (Right) L a t e n c y for each of the seven BSAER wavelets with increasing stimulus sensation levels. WOLFE ET AL. : Measures of the Brainstem AER

Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx

403

TAnLE 1. Group means and standard deviations for latency of each of the seven BSAER wavelets with increasing sensation levels. (Latency in milliseconds). dB Sensation Level Wave~t

I

III IV

VI VII

20

Mean SD

1.38 0.74 2.35 0.77 4.08 0.97 5.60 0.63 6.67 0.42 8.34 1.70 10.00 0.57

30

40

50

60

70

2.09 0.47 3.18 0.52 3.96 0.61 5.11 0.55 5.99 0.28 7.81 0.27 9.90* -

1.57 0.54 2.49 0.61 3.48 0.65 4.46 0.64 5.56 0.47 7.46 0.40 9.20 0.83

1.23 0.37 2.05 0.48 3.09 0.48 4.02 0.53 5.27 0.64 7.03 0.45 9.26 0.24

0.93 0.31 1.97 0.34 3.11 0.38 4.05 0.46 4.98 0.46 6.92 0.53 8.69 0.65

0.84 0.30 1.96 0.25 2.96 0.25 4.17 0.23 4.79 0.29 6.47 0.28 8.28 0.29

*Wavelet VII latency was based on four subiects. Amplitudes were calculated from the lowest sensation level at which each of the wavelets was consistently detectable, however considerable response variability existed among subiects. No consistent trends were noted for any of the components except Wavelet V. Wavelet V was noted to be the most consistent component of the BSAER recorded at the various sensation levels. It commonly is so identified and thus is the wavelet of primary interest. Peak amplitude of Wavelet V was plotted against sigual intensity and essentially a linear relation can be observed in Figure 3. Moreover, the correlation for this relation was calculated at r - 0.95. Individual-subiect and pooled-group means and standard deviations for three complete sets of data for each subiect for Wavelet V amplitude versus dB sensation level are presented in Table 2. Individual data generally are consistent with the linear trend between group means for response amplitude and dB SL increments.

DISCUSSION T h e results of this study concur with latency trends of previous studies of the BSAER (Jewett and Williston, 1971; Hecox and Galambos, 1974; Starr and Achor, 1975), however the latency values are slightly shorter. The seeming linear relation revealed between Wavelet V amplitude growth at increased sensation levels contradicted the Jewett and Williston, Hecox and Galambos studies, b u t concurred with the Starr and Achor study. The data of this study were collected from a few subiects and may prove to be an unreliable indicator of the relation. Although a linear trend apparently fits the data well, a slight 404

Journal of Speech and Hearing Research

21 401-407 June 1978

Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx

.4-

AMPLITUDE

WAVE-]Z GROUPMEANS

/ s j

U) ,4.m 0

Z Iii (:3

/J< /

Fm 1 Q_

L"

:E

/j .#/

I

I0

I

\

BEST FIT LINE

1

I

..... I

I

15" 20 30 40 50 dB re SENSATION LEVEL

1

60

!

70

FmuaE 3. Wavelet V amplitude with increasing stimulus sensation levels. TABLE 2. Individual- and pooled-group means and standard deviations for amplitude of Wavelet V with increasing stimulus sensation levels (amplitude in microvolts).

Subject

1

Group

Mean SD

dB Sensation Levels 40 50

20

30

0.229 0.069 0.167 0.038 0.161 0.030 0.284 0.053 0.213 0.040

0.230 0.050 0.240 0.040 0.230 0.040 0.290 0.030 0.250 0.040

0.293 0.036 0.232 0.059 0.257 0.030 0.331 0.042 0.263 0.025

0.212 0.065

0.250 0.040

0.268 0.057

60

70

0.330 0.080 0.250 0.050 0.220 0.060 0.290 0.060 0.240 0.080

0.340 0.040 0.270 0.040 0.280 0.050 0.280 0.080 0.350 0.040

0.480 0.020 0.380 0.070 0.270 0.020 0.270 0.030 0.360 0.030

0.260 0.080

0.300 0.060

0.350 0.090

d i p h a s i c i t y m a y be o b s e r v e d at a b o u t 40- to 50-dB SL. Davis a n d H i r s c h (1977) h a v e s p e c u l a t e d t h a t diphasicity occurs a r o u n d these sensation levels a n d possibly reflects the response of t w o populations, the i n n e r a n d o u t e r h a i r cells. WOLFE ET AL. : Measures of the Brainstem AER

Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx

405

Thus the input-output function of Wavelet V may be worthy of careful observation. Subject state, signal parameters, and the band-pass characteristics of the amplifiers vary among experiments. These factors may also affect the latency and amplitude of the responses (Suzuki, 1976). It should be noted, however, that a study by Cobb, Skinner, and Burns (1978) revealed that signal rise time affects BSAER response amplitudes significantly. They observed a linear trend between Wavelet V response amplitude and sensation level for click stimuli but not for short tone bursts. Earlier studies show a strong relation between Wavelet V latency and signal intensity and a weak relation between Wavelet V amplitude and signal intensity. Presumably, these findings have led investigators using the BSAER to focus on the latency relation and to discount that of amplitude. Peak amplitude is an important criterion in the detection of the BSAER and failure to evaluate the response amplitude, signal intensity relation may prove to be a significant oversight. The data obtained in this study were used to devise a BSAER template (Figure 4). The template is based on the amplitude and latency pooled group .50

1

1

1

I

1

1. . . .

1

'

'

lu

.45

- 9

.40

- 8

F-" V

--I l'rl

"

9~

7

I,I

o_

0

.50

6

I .z5

./"-

5 ~,

W

~-

-